What can be done to reduce this burden? Increasingly sophisticated methods for detecting disease in embryos during pregnancy will help, and these have recently taken another step forward with the development of accurate, non-invasive methods based on analysing foetal DNA in the blood of pregnant mothers (an article in the BMJ this week demonstrates the feasibility of this approach for a non-Mendelian disease, Down syndrome; and the same group showed late last year that this approach can also be applied to effectively any known disease-causing mutation). Yet these approaches detect disease after pregnancy has already begun.

Disease mutations can also be detected in embryos prior to implantation, for prospective parents undergoing IVF. But IVF remains an expensive, arduous and invasive procedure, and thus a weapon of last resort for most parents-in-waiting; as Armand Leroi notes drily in an exceptional 2006 article in EMBO Reports: “nature has contrived a cheap, easy and enjoyable way to conceive a child; IVF is none of these things.” (While Leroi goes on to argue that the challenges of IVF are less severe for young couples with no fertility problems, it still seems fairly implausible that this will become the default mode of reproduction in the near future.)

However, for some classes of Mendelian disease it’s possible to move the screening one step back. Recessive diseases are insidious things. The mutations that cause them lurk undetected – each of us carry perhaps 5 to 10 of them – as their carriers are protected by the presence of a healthy second copy of the affected gene. These mutations can thus wait silently for generation after generation, until a carrier is unlucky enough to fall for someone who carries the same mutation, or another mutation in the same gene. The children of such a couple will each have a 25% chance of inheriting one damaged copy of the gene from each parent and thus developing the disease.

The ability of a recessive mutation to pass silently from generation to generation means that many children born with recessive diseases have no family history. And while certain marriage practices (notably serial first-cousin marriage) can dramatically increase the risk of having a child with a recessive disease, these diseases can also explode into appearance in families with no obvious risk factors.

However, the fact that both parents must carry mutations in the same gene to pass a recessive disease to their children raises the possibility of detecting risk before a couple has even conceived children. For instance, one could screen both members of a couple for a panel of known mutations, an approach currently offered by US company Counsyl (disclaimer: my wife and I both accepted free tests from Counsyl in 2009). However, while a panel containing all known Mendelian mutations could detect a substantial fraction of all genetic disease (Leroi again), it can never eliminate the risk, because many Mendelian mutations remain undiscovered. However, one could go one step further: rather than simply look for known mutations, one could examine the entire sequence of all genes known to be associated with Mendelian diseases, and thus identify new mutations lurking in the same gene.

In an article published today in Science Translational Medicine a group of US researchers describe a high-throughput approach for doing precisely that.

Rather than review the technical aspects of the paper in great detail, I’ll just hit the main points I took from it:

• The 448 genes selected for the screening panel are the product of walking a political tightrope. While the authors indicate that the marginal costs of adding extra sequence mean that the optimal cost-benefit ratio comes from including a very broad range of diseases, they have excluded diseases that might trigger controversy (such as deafness and adult-onset disorders).
• The authors do a commendable job of comparing multiple technologies for capturing disease genes and for sequencing the captured DNA; the final product is a combination of Agilent SureSelect for sequence capture and Illumina HiSeq for sequencing.
• This approach allowed them to detect ~95% of the genetic variants in their target genes with very high accuracy. In other words, they might miss around 5% of disease-causing mutations in these genes, but the ones they find are probably real.
• Perhaps the single most important message from the paper, which I hope to expand on in a future post, is that disease mutations reported in the literature are depressingly enriched for false positives. The authors suggest that 27% of mutations in their samples that overlap with entries the largest database, the Human Gene Mutation Database, turned out to be the result of sequencing errors or mistakes in the literature (e.g. common polymorphisms that have been falsely reported to be disease-causing).
• The researchers found that the 104 sequenced samples contained on average 2.8 known disease-causing mutations in the surveyed genes. This fits with expectations: each of us is likely walking around with 5-10 of these severe disease-causing variants in total, which we will only know about if (1) we get our genomes sequenced, or (2) we’re unlucky enough to have children with a partner who carries mutations in the same gene.
• The authors estimate the cost of their test at $378, which they note is “approximating that expended on treatment of severe recessive childhood disorders per U.S. live birth”. In other words, offering this type of screen across the US population as a whole would be roughly cost-neutral from a healthcare stand-point, while simultaneously reducing the number of children dying from genetic diseases.

Given the plummeting costs of sequencing and the economies of scale, the cost-benefit ratio for this type of screening panel will continue to drop. In this context it seems inevitable that some form of sequencing approach will ultimately be implemented as routine for young parents-to-be.

Matt’s underlying argument is that while sequencing costs will continue to drop, obtaining a complete genome sequence that is sufficiently accurate for medical interpretation will require additional expenses (increased sequence coverage to ensure accuracy, all of the computation required to stitch the raw data into a useable form, and paying doctors to perform the interpretation) that will keep the cost of medical sequencing well above the arbimagical US$1,000 threshold. Instead, Herper argues, we will likely see medical-grade genomes stay above $10,000, or at least above the $2,000 currently forked out for MRI scans.

There’s certainly some depressing truth here. I believe Herper is right that if you intend to access your genome sequence via a traditional medical route, it will certainly cost more than $1000 for the foreseeable future – if indeed you can get access to it at all, which is by no means guaranteed. The costs of clinical sequencing will include even more overheads than Herper notes in his short post: for instance, even as the accuracy of high-throughput sequencing technology improves, there will still be a need for variants with major medical impact to be independently validated in clinical labs, and custom assays don’t come cheap.

However, many of these extra costs of clinical sequencing will be further inflated by regulatory demands (at least some of which will be arbitrary and pointless), and many will only apply if you obtain your genome through the medical system. Individuals with the motivation to seek alternative routes will be able to obtain a perfectly serviceable genome sequence at a substantially lower price: they’ll have to be cautious in how they interpret the results, of course (although, importantly, this is also true of a medical test), but it will be possible to obtain substantial and potentially extremely useful information from your own genome without having to pass through the clinical toll-booths.

The key obstacle will be the development of cheap (or free), intuitive tools for annotating large-scale genetic information. Purchasing the sequencing itself will be trivial: even if the FDA succeeds in crushing innovation in direct-to-consumer genetic testing in the US, there will be plenty of companies abroad (especially in East Asia) willing to convert a mailed saliva sample into an assembled genome sequence. What the individual needs to do with that sequence is to (1) validate that the sequence they receive is in fact their own genome; (2) extract medically useful variants; (3) confirm that these variants are real; and (4) figure out how that information should be used to make health and lifestyle decisions.

Only one of these steps (the final one) will require direct consultation with the medical profession. The first will require comparing the sequence with independent genetic data (such as a genome scan from a company like 23andMe) to check that the two match the same individual (i.e. you), and also to provide an indicator of the global quality of the sequence. The second step is currently extremely challenging, but we can expect tremendous innovation in genome interpretation software and databases of functional variants over the next few years that will gradually simplify and improve this process. The third step will require sending another DNA sample to a company that does affordable, custom assays of a small number of genetic regions of interest using an independent technology. And the final step will involve discussing the results with everyone who might be able to tell you something useful about them, including your family and your doctor.

None of this is simple, but it will become easier with time. As the retail costs of sequencing drops, a substantial niche will develop for innovators providing affordable, intuitive, accurate interpretation tools (embryonic versions already exist: see, for instance, Promethease or Enlis Genomics). Open-source academic software built for large-scale sequencing projects will be adapted for use by non-specialists. The increasing availability of large-scale computing power (for instance, via Amazon EC2), coupled with this intuitive software, will make even compute-intensive analyses available to the educated, motivated lay-person. (Incidentally, tracking and fostering the development of these tools is one of the motivations behind the Genomes Unzipped project – and you’ll hear more about our plans in this area in 2011).

Done carefully, there’s no reason why a DIY genome couldn’t be every bit as useful (or indeed as useless, in many cases) as one obtained through the doctor-as-gatekeeper route. As Dan Vorhaus argued after last year’s sample mix-up at 23andMe, it is likely that clever DIY genomicists will be better at picking up certain kinds of errors (such as sample swaps) than clinical labs would. In terms of accuracy, while retail genomes may not reach the same quality standards as those generated by clinical labs, increased competition and innovation in the direct-to-consumer space will mean they’re unlikely to lag too far behind – and judicious use of independent validation of important variants would in most cases raise the reliability to a level is equivalent to, or even higher than, a medical-grade test. Of course, for the very small number of genetic variants per genome that might require urgent, serious action – such as BRCA1 breast cancer-associated mutations – individuals can always fork out for a clinical test.

What proportion of people will take this DIY route? This isn’t for everyone. Those wealthy enough to blithely fork out $10,000 for a medical genome interpretation, or sufficiently unwell to be able to convince their insurance company or public health system to pay for it, will by and large simply take the expensive medical route. Of the remainder, relatively few people will be sufficiently motivated to develop the background knowledge required to make sense of their genome, even if the analysis software is relatively intuitive. But for the non-trivial fraction of the population who want to know about their genomes, but don’t want to pay the inflated costs associated with medical-grade sequencing – and I know this is a category that many readers of Genetic Future fall into – this will be an attractive and feasible option.

In addition, there will be advantages to the DIY approach beyond the lower cost. People who actively engage in the process of constructing useful information from raw sequence data – regardless of how intuitive the software is for doing it – will automatically learn important lessons about the nature of genetics (just as anyone who has given more than a casual glance at their own 23andMe profile has automatically learnt something important about the probabilistic nature of genetic risk factors for common diseases). They will also have opportunities to ask and answer fascinating questions that would be irrelevant in a purely medical consultation about your genome: for instance, what does your genetic information tell you about your ancestry?

They’ll be able to ask these questions because they will own the data. How easy do you think it will be to obtain your raw genome sequence from your doctor to use to satisfy your own curiosity? How many forms and disclaimers will you need to sign? How many times will you need to listen to someone tell you that the data files are just too large, that the formats are inaccessible to lay-people, that your request is extremely unusual and will need to be considered for months by a hospital committee in the name of “health data privacy”? Anyone who has ever tried to get access to their own medical records will know how tedious and shrouded in unnecessary mystery this process can be; imagine how much larger the obstacles will loom when the system has the additional excuse of large, complex file formats to throw in your path.

Importantly, your genome is just one contributor to your present and future health. DIY genomics will, I hope, be part of a larger ongoing trend towards individuals taking greater personal responsibility for tracking and maintaining their own wellness – a task, incidentally, which modern healthcare systems are spectacularly ill-equipped to perform, something that seems unlikely to change substantively in the near future. As Western populations age, this broader shift of responsibility will be essential for healthcare systems to survive.

But I digress. My point here is simply this: Herper is perfectly correct that the overheads imposed (for a mixture of valid and arbitrary reasons) by medical-grade testing will ensure that clinical genomes remain expensive. But for those of us willing to learn the skills required to go outside the system, the $1000 genome is rapidly approaching. We just need to be ready to make the most of what it contains – and to reap the benefits of accessing that information as an active, engaged participant rather than a passive recipient.

These claims would be fascinating, if true. However, while the article makes some (scattered) valid points, its central claim (that the results of GWAS suggest that genetics plays little or no role in the causation of common diseases) is entirely false, and the authors rely on a combination of distortions and statistical misunderstandings to make their case.

In a comment over at the Huffington Post found by Keith Grimaldi, one of the authors explains the key messages and motivations of his analysis:

We have just reported that genetics now demonstrat­es that genes cannot be the cause of common diseases:

http://www­.bioscienc­eresource.­org/commen­taries/art­icle.php?i­d=46

That means environmen­t must be the entire cause of ill health, i.e. junk food, pollution, lack of exercise, etc. The reason we wrote an article about human genetics (when we are a food and agricultur­e website) is that we believe that if people live right, agricultur­e and therefore the planet will more or less fix itself. [my emphasis]

This quote is illuminating in a number of ways. Firstly, it shows that there is no nuance in this argument: the authors aren’t attempting to argue that genes play a smaller role in common disease than geneticists expected, but rather that genetics plays no role whatsoever. 

For each disease, even if a person was born with every known ‘bad’ (or ‘good’) genetic variant, which is statistically highly unlikely, their probability of contracting the disease would still only be minimally altered from the average.

Studies of human twins estimate heritability (h²) by calculating disease incidence in monozygotic (genetically identical) twins versus dizygotic (fraternal) twins (who share 50% of their DNA). If monozygotic twin pairs share disorders more frequently than do dizygotic twins, it is presumed that a genetic factor must be involved. A problem arises, however, when the number resulting from this calculation is considered to be an estimate of the relative contribution of genes and environment over the whole population (and environment) from which the twins were selected. This is because the measurements are done in a series of pairwise comparisons, meaning that only the variation within each twin pair is actually being measured. Consequently, the method implicitly defines as environment only the difference within each twin pair. Since each twin pair normally shares location, parenting styles, food, schooling, etc., much of the environmental variability that exists between individuals in the wider population is de facto excluded from the analysis. In other words, heritability (h²), when calculated this way, fails to adequately incorporate environmental variation and inflates the relative importance of genes. [my emphasis]

The last fifteen years, coinciding with the rise of medical genetics, have seen unprecedented sums of money directed at medical research. At the same time, research on pollution, nutrition and epidemiology has not benefited in any comparable way.

[…]

This same mindset is accurately reflected in the media where even strong environmental links to disease often receive little attention, while speculative genetic associations can be front page news.

Even as a direct beneficiary of money thrown at medical genetics over the last five years, and someone who blogs entirely about news in the genetic domain, I freely acknowledge that these criticisms have merit. Genetic dissection of common disease is valuable, and will be (and indeed already has been) fruitful in generating new therapies, but it is nonetheless true that research into environmental risk factors and interventions to minimise morbidity is woefully under-funded and under-reported relative to its potential benefit.

Personal genomics company 23andMe has made some fairly major announcements this week: a brand new chip, a new product strategy (including a monthly subscription fee), and yet another discount push. What do these changes mean for existing and new customers?

The new chip

• Increased coverage of drug metabolizing enzymes and transporters (DMET) as well as other genes associated with response to various drugs. 
• Increased coverage of gene markers associated with Cystic Fibrosis and other Mendelian diseases such as Tay-Sachs. 
• Denser coverage of the Human Leukocyte Antigen region, which contains genes related to many autoimmune conditions.

I summarised some of the key motivations for members of the group in my Unzipped announcement post:

• we want to share the results of scientific analysis of our own genomes, and as proponents of open data access most of us believe that doing good science means releasing complete data for others to investigate;
   
• we hope that releasing our data publicly will help to guide useful discussions about genetic privacy and the benefits, risks and limitations of genetic information in general;
   
• many of us believe that the ideal resource for genetic research is large open-access, non-anonymous research databases such as the Personal Genome Project, and that sharing linked genetic and trait information openly with the wider community is a public good – and we hope that our own experiences will encourage others to participate in open research projects;
   
• we all believe that many of the fears expressed about the dangers of genetic information are exaggerated, and see this project as an opportunity to have a constructive public discussion about the truth behind these fears;
   
• given the ease with which a dedicated snoop could obtain genetic information surreptitiously (via shed skin, hair or saliva, for instance), some of us argue that the whole notion of genetic privacy is illusory anyway – while releasing our data online makes it easier for people to get hold of it, this is a difference of degree rather than kind.

In my second big piece of news for the day, I’m pleased to announce that Genetic Future will shortly be moving to a  the brand new Wired Science Blogs network.

The American Society of Human Genetics’ annual meeting has now kicked off in San Francisco. The usual terrifyingly large conference center, dizzying collection of stalls and posters as far as the eye can see are all very much in evidence, as is the migrating herds of human geneticists wondering the landscape looking for food in the perpetually busy local restaurants.

The first session I attended was given the perplexing title of “Yes Virginia, Family Studies Really Are Useful for Complex Traits in the Next-Generation Sequencing Era”. The session was in honour of the 80th birthday of statistical geneticist Robert Elston, whose development of the Elston-Stewart algorithm in the 1970s kickstarted the field of parametric linkage. It covered the use of next-generation sequencing family studies in the discovery of the hypothesised rare risk variants that everyone hopes to find. This isn’t really a new topic, and I last wrote about it at ASHG 2010, but it had a very different feel to two years ago.

Read the rest of this entry | Read comments

The big paper

Out this week in Nature is our new big paper on the genetics of inflammatory bowel disease. We genotyped over 75 thousand individuals and discovered 71 new loci and god am I sick of writing this sentence. In what may seem like slightly desperate behaviour, I have posted four different bits of writing on four different websites (five if you include this post) following up on this paper. You can read my general thoughts on the Sanger Institute website, what I think this publication means for the biology of IBD on the International IBD Genetics Consortium’s website, what I think these loci do (or do not) mean for genetic risk prediction at Genomes Unzipped, and some thoughts on how we went about visualising the results on my team’s website.

In my defense this is the Big Paper from my PhD, and is probably the most significant marker of what I have managed to get done over the last four years, so I do have plenty of things to say on the subject!

My PhD and beyond

Read the rest of this entry | Read comments

Ewan Birney reported on results from the ENCODE project, a truly massive consortium project to look at the role of non-coding functional DNA. My take-home message was that a chunk of the genome significantly larger than the entire exome can be confidently said to be bound by a protein in at least one cell line. This includes bound transcription factor motifs and DNase1 footprints: these aren’t wooly definitions, and certainly miss out lots of important non-coding annotations. Ewan also presented evidence implying that we have only captured half of what we eventually could if we sequenced all human cell types. Overall, we can guess that for every base pair that codes for part of an exon, there will be another four base pairs responsible for binding proteins. A few other ENCODE talks dug deeper into the data, including one from Mark Gerstein about using transcription factor binding to construct gene networks. Interestingly, he showed that networks connected via distal regulation (i.e. regulation via proteins bound outside the promoter) are very different to those formed by proximal regulation.

As well as cataloging this variation, researchers are also getting better and better at figuring out the mechanisms of genome regulation, and there were quite a few talks that really dug down into the specific dynamics of the genome and its assorted bound products.

Read the rest of this entry | Read comments

As usual, there were a few talks in the first session describing large consortium analyses of human complex disease. For instance, I presented work (on behalf of the International IBD Genetics Consortium) on the Inflammatory Bowel Disease immunochip project. We have genotyped tens of thousands of cases from 15 different countries, and discovered a host of new common loci for IBD. I’ll be writing about this project on Genomes Unzipped soon. On the other end of the allele frequency spectrum, Mark McCarthy reported on some of the next-generation sequencing projects that are going on in Type 2 Diabetes; these have less samples (something like 7K cases in total), but generated high quality calls for a large number of rare variants. Mark reported on a few interesting hints of rare T2D associations, but his overall conclusion was that we will need tens of thousands of samples to be well powered to find rare variants that underlie common disease. We will need to go beyond just sequencing a few thousand samples, and start designing well-powered replication studies to follow up what we find.

But I wanted to talk about some more non-standard, and slightly cleverer studies of non-human phenotypes that I found interesting. Three speakers described pretty nifty studies that used the particular properties of non-human sequence data to do some well-powered sequencing experiments that wouldn’t be possible with humans.

Read the rest of this entry | Read comments

To cut straight to the chase, the first session (High Throughput Genomics and Genetics) included a total of four different talks on sequencing the genomes of single cells. A clear theme if ever there was one.

Two of these talks discussed creating personal genetic maps of recombination by directly sequencing single sperm cells. Adam Auton went first – his dataset was an example of a particularly tricky setup, including massive amplification biases and a whopping 75% of his samples showing evidence of contamination. Despite this, he showed that with the right QC and statistical model you can still get a good map that is very similar to existing maps, even given these problems. Stephen Quake presented a somewhat less difficult sperm sequencing study, which used a special microfluidic chip (previously used for chromosome separation) to separate out the sperm cells. Low coverage sequencing could produce a map that closely recapitulated existing maps, and higher coverage could be used to estimate rates of aneuploidy and de novo mutation.

Read the rest of this entry | Read comments

I am lucky enough to be staying on site again this year, rather than having to endure the daily death-defying suspensionless bus ride from one of the satellite hotels. However, the cabin that I have been put in this year breaks my current record for the longest climb from lecture theater to front door:

At its heart, this paper is a challenge to a common assumption used in statistic genetics: the assumption of additive genetic risk. This states that genetic risk factors act independently of each other, with each variant increasing genetic risk by the same amount regardless of what other risk factors are present*. Of course, this is clearly a spherical cow situation, we know that the cell is full of complex interactions of various sorts, and a mutation cannot help but be effected in some way by the rest of the genome. But mathematically the assumption simplifies much of the complexity of statistical genetics, and allows you to do a number of things that would be very hard otherwise. We generally think additivity is a good approximation; it doesn’t matter if it is slightly wrong, and we’d have picked up if it were very wrong.

Zuk et al’s claim is that it is possible that additivity is wrong, that did not pick it up, and that it really does matter. This blog post will discuss the specific arguments that Zuk et al make against additivity. Some of the broader implications of the research is discussed over at Genomes Unzipped, and in particular looking at what this does and doesn’t say about total (not additive) heritability.

Read the rest of this entry | Read comments

A new group blog, res gerendae, came online at the end of last year, and is interesting for a few reasons. Firstly, it is written by Classics students (those who study Roman, Greek and other societies of the Classical antiquity), who aren’t really known for their blogging. Secondly, the blog is open to all the Classics grads of Cambridge University, potentially acting as a voice for all students of the Faculty of Classics.

So far you can read about graffiti then and now, the real location of Troy and a surprisingly technical reading of a postcard.

Lets hope other student bodies take inspiration from this, and do their bit to get students involve in writing about their field.

For me, the hightlight of yesterday was (somewhat obviously) the Individual Resequencing for Complex Trait Genetics session. This was organised by Mark Daly and Ben Neale of Mass. Gens. Analytic and Translational Genetics Unit, and gathered together a number of the Big Men (all men, unfortunately) of disease association together to talk about the many and varied Next-Generation sequencing studies they have been working on. I’ve summarised some of what was said below.

As always, you can find more coverage of ICHG on twitter (@lukejostins for me, #ICHG2011 for aggregated coverage).

Read the rest of this entry | Read comments

The debate was pretty lively, and in parts both enlivening and depressing. Below are a few points that I found interesting.

As always, you can find more coverage of ICHG on twitter (@lukejostins for me, #ICHG2011 for aggregated coverage).

Different perspectives

Joris Veltman described his exome sequencing of 500 individuals with intractable disease, and noted that there has been much success, and very little evidence of harm. Ségolène Aymé mentioned NIH targts that hope to see almost all genetic diseases diagnosed by 2020, and new treatments for rare diseases to be developed simultaneously. There seemed to be a solid consensus across the panel that sequencing should be rolled out as a standard tool in the diagnosis of genetic diseases, provided that the approach is a targeted one, restricted to finding the pathogenic mutation(s) causing the disease.

More controversial was the role of sequencing of healthy individuals, and the general return of data to patients or doctors for any reason other than directly diagnosing a genetic disease. Rade Drmanac, chief scientific officer of Complete Genomics, was obviously strongly in favour of everyone having their genome sequenced, and made it clear that Complete Genomics intends to start offering sequencing to doctors in the future. In his vision, genomes are sequenced at birth, and an initial analysis of immediately actionable results (e.g. potential genetic diseases) is passed to the doctor and patient, with further analyses being carried out if and when they are required.

Michael Hayden immediately dismissed this as hype. He pointed out how unable the US is to handle medical sequencing, with no good systems of reimbursement, a massive shortage of genetic councilors, and a general lack of training in the medical profession.While more positive in general, Louanne Hudgins also expressed worries about the lack of knowledge of genetics among doctors, with some truly scary examples of MDs failing to understanding even the most basic genetics.

Read the rest of this entry | Read comments

On episode #356 of the science show This Week in Virology, Stephanie joins the super professors to discuss the gut virome of children with serious malnutrition, caterpillar genes acquired from parasitic wasps, and the effect of adding chemokines to a simian immunodeficiency virus DNA vaccine.

You can find TWiV #356 at www.twiv.tv.

On episode #355 of the science show This Week in Virology, the TWiV team considers the effect of a Leishmaniavirus on the efficacy of drug treatment, and the human fecal virome and microbiome in twins during early infancy.

You can find TWiV #355 at www.twiv.tv.

The protozoan parasite Leishmania, transmitted to humans by the bite of a sandfly, may cause disfiguring skin lesions. A virus within the parasite appears to increase the risk of treatment failure with anti-leishmania drugs.

A double-stranded RNA virus was found over 20 years ago to infect different species of Leishmania, with up to 50% of clinical isolates infected. Leishmaniavirus (LRV) causes a chronic infection with little effect on the parasite. In mouse models, infection of Leishmania with LRV is associated with increased parasite replication and disease severity. The double-stranded RNA genome of LRV appears to be sensed by the mammalian innate immune system, leading to overproduction of cytokines and a hyper-inflammatory response. Similarly, the dsRNA of Trichomonas vaginalis virus is also sensed by the innate immune system, leading to inflammatory complications.

Two independent studies have been done to assess the consequence of LRV infection in human cases of leishmaniasis. In one study, presence of LRV was determined in Leishmania braziliensis isolated from 97 patients in Peru and Bolivia. The patients were treated with pentavalent antimonials or amphotericin B, and the outcome was determined as ‘cured’ or ‘failure’. Thirty-two (33%) Leishmania isolates were found to contain LRV.   Treatment failed in 33% of the patients (18 of 54). There were fewer drug failures in the LRV negative isolates (9 of 37, 24%) than in the LRV positive isolates 9 of 17, 53%). These observations demonstrate that presence of LRV is associated with a significant increase in the risk of treatment failure.

In the second study, carried out in French Guiyana, 58% of 75 patients with Leishmania guyanensis infection had LRV in the parasite. All the patients with LRV-negative Leishmania were cured after one or two treatments with pentamidine, while 12 of 44 LRV-positive patients (27%) had persistent infections requiring treatment with other drugs. In addition, presence of LRV was associated with high levels of inflammatory cytokines within lesions.

The results of both studies show that infection of Leishmania with LRV is associated with drug treatment failure and persistent infection. Determining whether LRV is present in infected patients could therefore guide better treatment strategies. How the presence of the virus leads to such consequences is unknown. The effect might be a consequence of higher parasite numbers associated with LRV infection, which simply overcome already marginal drugs. The host inflammatory response caused by the dsRNA of LRV might also play a role. Understanding the precise mechanism might allow the development of drugs that overcome the effects of LRV. It might also be useful to develop drugs that target LRV, thereby improving the efficacy of anti-Leishmania drugs.

On episode #354 of the science show This Week in Virology, the esteemed doctors of TWiV review a new giant virus recovered from the Siberian permafrost, why influenza virus gain of function experiments are valuable, and feline immunodeficiency virus.

You can find TWiV #354 at www.twiv.tv.

The Sabin infectious, attenuated poliovirus vaccines are known to cause vaccine-associated paralysis in a small number of recipients. In contrast, the Salk inactivated vaccine does not cause poliomyelitis. Why are the Sabin vaccines still used globally? The answer to this question requires a brief visit to the history of poliovirus vaccines.

The inactivated poliovirus vaccine (IPV) developed by Jonas Salk was licensed for use in 1955. This vaccine consists of the three serotypes of poliovirus whose infectivity, but not immunogenicity, is destroyed by treatment with formalin. When prepared properly, IPV does not cause poliomyelitis (early batches of IPV were not sufficiently inactivated, leading to vaccine-associated outbreaks of polio, the so-called Cutter incident). From 1955 to 1960 cases of paralytic poliomyelitis in the United States dropped from 20,000 per year to 2,500.

While Salk’s vaccine was under development, several investigators pursued the production of infectious, attenuated vaccines as an alternative. This approach was shown to be effective by Max Theiler, who in 1937 had made an attenuated vaccine against yellow fever virus by passage of the virulent virus in laboratory mice. After many passages, the virus no longer caused disease in humans, but replicated sufficiently to induce protective immunity. Albert Sabin capitalized on these observations and developed attenuated versions of the three serotypes of poliovirus by passage of virulent viruses in different animals and cells. In contrast to Theiler’s yellow fever vaccine, which was injected, Sabin’s poliovirus vaccines were designed to be taken orally – hence the name oral poliovirus vaccine (OPV). As in a natural poliovirus infection, Sabin’s vaccines would replicate in the intestinal tract and induce protective immunity there and in the bloodstream.

Sabin began testing his attenuated vaccines in humans in 1954. By 1957 there was evidence that the virus that was fed to volunteers was not the same as the virus excreted in the feces. As Sabin writes:

It was evident, however, that as in the young adult volunteers, the virus in some of the stool specimens had a greater neurovirulence than the virus originally swallowed in tests in monkeys.

What Sabin did not know was whether the change in neurovirulence of his vaccine strains constituted a threat to the vaccine recipients and their contacts, a question that could only be answered by carrying out larger clinical trials. Many felt that such studies were not warranted, especially considering the success of IPV in reducing the number of paralytic cases. Sabin notes that his friend Tom Rivers, often called the father of American virology, told him to ‘discard the large lots of OPV that I had prepared into a suitable sewer’.

Despite the opposition to further testing of OPV in the US, others had different views. An international committee of the World Health Organization recommended in 1957 that larger trials of OPV should be carried out in different countries. Sabin’s type 2 vaccine was given to 200,000 children during an outbreak of polio in Singapore in 1958, and follow-up studies revealed no safety problems. In Czechoslovakia 140,000 children were given OPV and subsequent studies revealed that the virus spread to unimminized contacts but did not cause disease.

Perhaps the most important numbers came from trials of OPV in the Soviet Union. Sabin had been born in Russia and had close contacts with Soviet virologists, including Mikhail Chumakov, director of the Poliomyelitis Research Institute in Moscow. Chumakov was not satisfied with the results of IPV trials in his country and asked Sabin to send him OPV for testing. By the end of 1959 nearly 15,000,000 people had been given OPV in different parts of the Soviet Union with no apparent side effects. Dorothy Horstmann, a well known virologist at Yale University, was sent to the Soviet Union to evaluate the outcome of the trials. Horstmann writes:

It was clear that the trials had been carefully carried out, and the results were monitored meticulously in the laboratory and in the field. By mid-1960 approximately 100 million persons in the Soviet Union, Czechoslovakia, and East Germany had received the Sabin strains. Of great importance was the demonstration that the vaccine was safe, not only for the recipients, but for the large numbers of unvaccinated susceptible who must have been exposed as contacts of vaccines.

The results obtained from these trials in the Soviet Union convinced officials in the US and other countries to carry out clinical trials of OPV. In Japan, Israel, Chile, and other countries, OPV was shown to be highly effective in terminating epidemics of poliomyelitis. In light of these findings, all three of Sabin’s OPV strains were approved for use in the US, and in 1961-62 they replaced IPV for routine immunization against poliomyelitis.

As soon as OPV was used in mass immunizations in the US, cases of vaccine-associated paralysis were described. Initially Sabin decried these findings, arguing that temporal association of paralysis with vaccine administration was not sufficient to implicate OPV. He suggested that the observed paralysis was caused by wild-type viruses, not his vaccine strains.

A breakthrough in our understanding of vaccine-associated paralysis came in the early 1980s when the recently developed DNA sequencing methods were used to determine the nucleotide sequences of the genomes of the Sabin type 3 vaccine, the neurovirulent virus from which it was derived, and a virus isolated from a child who had developed paralysis after administration of OPV. The results enumerated for the first time the mutations that distinguish the Sabin vaccine from its neurovirulent parent. More importantly, the genome sequence of the vaccine-associated isolate proved that it was derived from the Sabin vaccine and was not a wild-type poliovirus.

We now understand that every recipient of OPV excretes, within a few days, viruses that are more neurovirulent that the vaccine strains. This evolution occurs because during replication of the OPV strains in the human intestine, the viral genome undergoes mutation and recombination that eliminate the attenuating mutations that Sabin so carefully selected by passage in different hosts.

From 1961 to 1989 there were an average of 9 cases (range, 1-25 cases) of vaccine-associated paralytic poliomyelitis (VAPP) in the United States, in vaccine recipients or their contacts, or 1 VAPP case per 2.9 million doses of OPV distributed (illustrated). Given this serious side effect, the use of OPV was evaluated several times by the Institute of Medicine, the Centers for Disease Control and Prevention, and the Advisory Committee on Immunization Practices. Each time it was decided that the risks associated with the use of OPV justified the cases of VAPP. It was believed that a switch to IPV would lead to outbreaks of poliomyelitis, because: OPV was better than IPV at protecting non-immunized recipients; the need to inject IPV would lead to reduced compliance; and IPV was known to induce less protective mucosal immunity than OPV.

After the WHO began its poliovirus eradication initiative in 1988, the risk of poliovirus importation into the US slowly decreased until it became very difficult to justify routine use of OPV. In 1996 the Advisory Committee on Immunization Practices decided that the US would transition to IPV and by 2000 IPV had replaced OPV for the routine prevention of poliomyelitis. As a consequence VAPP has been eliminated from the US.

OPV continues to be used in mass immunization campaigns for the WHO poliovirus eradication program, because it is effective at eliminating wild polioviruses, and is easy to administer. A consequence is that neurovirulent vaccine-derived polioviruses (VDPV) are excreted by immunized children. These VDPVs have caused outbreaks of poliomyelitis in areas where immunization coverage has dropped. Because VDPVs constitute a threat to the eradication campaign, WHO has recommended a global transition to IPV. Once OPV use is eliminated, careful environmental surveillance must be continued to ensure that VDPVs are no longer present before immunization ceases, a goal after eradication of poliomyelitis.

As a virologist working on poliovirus neurovirulence, I have followed the vaccine story since I joined the field in 1979. I have never understood why no cases of VAPP were observed in the huge OPV trials carried out in the Soviet Union. Had VAPP been identified in these trials, OPV might not have been licensed in the US. Global use of OPV has led to near global elimination of paralytic poliomyelitis. Would the exclusive use of IPV have brought us to the same point, without the unfortunate cases of vaccine-associated paralysis? I’m not sure we will ever know the answer.

Update: As recently as 1997 DA Henderson, architect of smallpox eradication, argued that developed countries should not use IPV, because it ‘implies accepting the potential of substantial penalties while reducing but not eliminating, an already extremely small risk of vaccine-associated paralytic illness’.

ASM Live will be broadcast from ICAAC/ICC 2015 in San Diego, CA, where host Michael Schmidt, PhD, Professor and Vice Chairman of Microbiology and Immunology at the Medical University of South Carolina, and co-host of This Week in Microbiology, will interview researchers about their work.

Streaming will take place at the San Diego Convention Center, Room 29B, and meeting registrants are encouraged to attend. You can watch ASM Live at microbeworld.org. Content will also be archived immediately on YouTube and MicrobeWorld for future viewing.

On episode #353 of the science show This Week in Virology, the TWiVniacs discuss twenty-eight years of poliovirus shedding by an immunodeficient patient, and packaging of the innate cytoplasmic signaling molecule cyclic GMP-AMP in virus particles.

You can find TWiV #353 at www.twiv.tv.

An immunodeficient individual has been excreting poliovirus in his stool for 28 years. Such chronic excreters pose a threat to the poliovirus eradication program.

Since its inception in 1988 by the World Health Organization, the poliovirus eradication program has relied on the use of the infectious, attenuated vaccine strains produced by Albert Sabin. These viruses are taken orally, replicate in the intestine, and induce protective immunity. During replication in the gut, the Sabin strains lose the mutations that prevent them from causing paralysis. Nearly every individual who receives the Sabin vaccine strains excretes so-called vaccine-derived polioviruses (VDPVs) which are known to have caused outbreaks of poliomyelitis in under-immunized populations.

Immunocompromised individuals who produce very low levels of antibodies (a condition called agammaglobulinemia) are known to excrete VDPVs for very long periods of time – years, compared with months in healthy individuals. Seventy-three such cases have been described since 1962. These individuals receive the Sabin vaccine in the first year of life, before they are known to have an immunodeficiency, at which time they must receive antibodies to prevent them from acquiring fatal infections.

The most recently described immunocompromised patient was found to excrete poliovirus type 2 vaccine for 28 years (the time period is determined by combining the known rate of change in the poliovirus genome with sequence data on viruses obtained from the patient).  The VDPV is neurovirulent (causes paralysis in a mouse model), antigenically drifted, and excreted in the stool at high levels.

Because the polio eradication plan calls for cessation of vaccination at some future time, these immunocompromised poliovirus shedders pose a threat to future unimmunized individuals. The global number of such patients is unknown, and there is no available therapy to treat them – administration of antibodies does not clear the infection. The development of antivirals that could eliminate the chronic poliovirus infection is clearly needed (and ongoing). It will also be necessary to conduct environmental surveillance for the presence of VDPVs – they can be identified by properties that distinguish them from VDPVs produced by immunocompetent vaccine recipients.

While the WHO eradication plan now includes a shift to using inactivated (Salk) poliovaccine, this strategy would not impact the existing immunocompromised poliovirus shedders. Should a VDPV from these individuals cause an outbreak of polio in the post-vaccine era, it will be necessary to control the outbreak with Salk vaccine, or an infectious poliovirus vaccine that cannot revert to virulence during replication in the intestine. Polioviruses with a recoded genome are candidates for the latter type of vaccine.

Image credit: Jason Roberts

When I became Peter Palese’s first Ph.D. student in 1976, his laboratory at Mt. Sinai School of Medicine in New York City was in dire need of shelves. The laboratory benches (pictured) had no room for storing the many bottles of reagents that I was beginning to generate.

When I told Peter that his laboratory needed shelves, he told me to get them. I looked in the yellow pages (there was no internet at the time) and found a dealer in downtown Manhattan. When I told Peter, he responded, to my amazement, that he would drive me there to get them.

Within minutes we were in Peter’s car, driving far downtown to a small garage filled with shelf parts. The proprietor loaded some into Peter’s car and we returned to Mt. Sinai, where we carried the metal pieces up the elevator into the laboratory. There I bolted them together and made one set of shelving for the low lab bench, and a second set for a higher bench. The photo of Peter shows what the shelves looked like just a few years ago.

A few days ago Matt Evans, a professor in the same department as Peter, told me that the Palese lab was being renovated and that the shelves that I had built were being discarded. He knew that I had made the shelves because I talk about them when I give a seminar. Matt was kind enough to send photos of the shelves, and of the low lab bench on which they once rested. It’s amazing that the shelves lasted for 39 years!

There is a reason why I’m writing this story, and why I tell it whenever I give a seminar. I had selected Peter to be my Ph.D. mentor, which meant that I listened to him. When he told me to build the shelves, I built them. When I went to David Baltimore’s laboratory as a postdoctoral fellow, I listened to David. When either Peter or David asked me to do something, I did it. I selected them as mentors because they had more experience in virology than I had and I wanted to train with them. Therefore when they asked me to jump, I said, ‘how high’?

I am not saying that mentors should have the ability to make their trainees do whatever they ask. Of course it is important to question; if you don’t understand why your mentor is asking you to do a specific experiment, by all means ask them! Being an effective trainee means listening and questioning. But remember that in the end they have more experience doing science than you have. If you don’t want to listen, why did you pick them in the first place?

I have been very lucky to have trained many fine virologists during my 33 years at Columbia University. Most have listened to my suggestions. The best have always asked questions, leading to many wonderful two-way interactions. But some chose to consistently ignore my suggestions for experiments or projects. I have seen similar conduct in other laboratories here and elsewhere. This behavior baffles me. To this day, if Peter or David asked me to do something, I would immediately comply. They will always be my mentors.

Update: Gayathri, a Ph. D. student in the Palese lab, sent me a before and after photo of the shelves.

On episode #352 of the science show This Week in Virology, Vincent meets up with Michele Banks in Washington, DC to discuss her career as a creator of science-themed art.

You can find TWiV #352 at www.twiv.tv.

On episode #351 of the science show This Week in Virology, the Masters of the ScienTWIVic Universe discuss a novel poxvirus isolate from an immunosuppressed patient, H1N1 and the gain-of-function debate, and attenuation of dengue virus by recoding the genome.

You can find TWiV #351 at www.twiv.tv.

Image credit

The individuals who believe that certain types of gain-of-function experiments should not be done because they are too dangerous (including Lipsitch, Osterholm, Wain-Hobson,) cite the 1977 influenza virus H1N1 strain as an example of a laboratory accident that has led to a global epidemic. A new analysis shows that the reappearance of the 1997 H1N1 virus has little relevance to the gain-of-function debate.

Human influenza viruses of the H3N2 subtype were circulating in May of 1977 when H1N1 viruses were identified in China and then Russia. These viruses spread globally and continue to circulate to this day. The results of serological tests and genetic analysis indicated that these viruses were very similar to viruses of the same subtype which circulated in 1950 (I was in the Palese laboratory in 1977 when these finding emerged). Three hypotheses were suggested to explain the re-emergence of the H1N1 virus: a laboratory accident, deliberate release, or a vaccine trial.

Rozo and Gronvall have re-examined the available evidence for the origin of the 1977 H1N1 virus. While there is ample documentation of the extensive work done during the 1970s in the Soviet Union on biological weapons, there is no evidence that Biopreparat had attempted to weaponize influenza virus. The release of the 1977 H1N1 virus from a biological weapons program is therefore considered unlikely.

It is more likely that the 1977 H1N1 virus was released during testing of influenza virus vaccines. Many such trials were ongoing in the USSR and China during the 1960s-70s. C.M. Chu, a Chinese virologist, told Peter Palese that the H1N1 strain was in fact used in challenge studies of thousands of military recruits, an event which could have initiated the outbreak.

The hypothesis that the 1977 H1N1 virus accidentally escaped from a research laboratory is formally possible, but there are even less data to support this contention. Shortly after this virus emerged, WHO discounted the possibility of a laboratory accident, based on investigations of Soviet and Chinese laboratories. Furthermore, the H1N1 virus was isolated at nearly the same time in three distant areas of China, making release from a single laboratory unlikely.

It is of interest that with the onset of the gain-of-function debate, which began in 2011 with the adaptation of influenza H5N1 virus to aerosol transmission among ferrets, the ‘laboratory accident’ scenario for the emergence of the 1977 strain has been increasingly used as an example of why certain types of experiments are ‘too dangerous’ to be done (See graph, upper left). For example, Wain-Hobson says that ‘1977 H1N1 represented an accidental reintroduction of an old vaccine strain pre-1957, probably from a Russian research lab’. Furmanski writes that ‘The virus may have escaped from a lab attempting to prepare an attenuated H1N1 vaccine’. In the debate on gain-of-function experiments, the laboratory escape hypothesis is prominently featured in public presentations.

The use of an unproven hypothesis to support the view that some research is too dangerous to do is another example of how those opposed to gain-of-function research bend the truth to advance their position. I have previously explained how Lipsitch incorrectly represented the results of the H5N1 ferret transmission studies. We should not be surprised at this tactic. After all, Lipsitch originally called for a debate on the gain-of-function issue, then shortly thereafter declared that the moratorium should be permanent.

Rozo and Gronvall conclude that the use of the 1977 influenza epidemic as a cautionary tale is wrong, because it is more likely that it was the result of a vaccine trial and not a single laboratory accident:

While the events that led to the 1977 influenza epidemic cannot preclude a future consequential accident stemming from the laboratory, it remains likely that to this date, there has been no real-world example of a laboratory accident that has led to a global epidemic.

On episode #350 of the science show This Week in Virology, Vincent speaks with Katherine High about her career and her work on using viral gene therapy to treat inherited disorders.

This episode is drawn from one of twenty-six video interviews with leading scientists who have made significant contributions to the field of virology, part of the new edition of the textbook Principles of Virology.

You can find TWiV #350 at www.twiv.tv.

I have a soft spot in my heart for Lassa virus: a non-fictional account of its discovery in Africa in 1969 inspired me to become a virologist. Hence papers on this virus always catch my attention, such as one describing its origin and evolution.

Lassa virus, a member of the Arenavirus family, is very different from Ebolavirus (a filovirus), but both are zoonotic pathogens that may cause hemorrhagic fever. It is responsible for tens of thousands of hospitalizations, and thousands of deaths each year, mainly in Sierra Leone, Guinea, Liberia, and Nigeria. Most human Lassa virus outbreaks are caused by multiple exposures to urine or feces from the multimammate mouse, Mastomys natalensis, which is the reservoir of the virus in nature. In contrast, outbreaks of Ebolavirus infection typically originate with a crossover from an animal reservoir, followed by human to human transmission. Despite being studied for nearly 50 years, until recently the nucleotide sequences of only 12 Lassa virus genomes had been determined.

To remedy this lack of Lassa virus genome information, the authors collected clinical samples from patients in Sierra Leone and Nigeria between 2008 and 2013. From these and other sources they determined the sequences of 183 Lassa virus genomes from humans, 11 viral genomes from M. natalensis, and two viral genomes from laboratory stocks. All the data are publicly available at NCBI. Analysis of the data lead to the following conclusions:

• Lassa virus forms four clades, three in Nigeria and one in Sierra Leona/Liberia (members of a clade evolved from a common ancestor).
• Most Lassa virus infections are a consequence of multiple, independent transmissions from the rodent reservoir.
• Modern-day Lassa virus  strains probably originated at least 1,000 years ago in Nigeria, then spread to Sierra Leone as recently as 150 years ago. The lineage is most likely much older, but how much cannot be calculated from the data.
• The genetic diversity of Lassa virus in individual hosts is an order of magnitude greater than the diversity of Ebolavirus. Furthermore, Lassa virus diversity in the rodent host is greater than in humans, likely a consequence of the longer, persistent infections that take place in the mouse.
• The gene encoding the Lassa virus glycoprotein is subject to high selection in hosts, leading to variants that interfere with antibody binding.
• Genetic variants that arise in one rodent are not transmitted to another.

Perhaps the most important result from this work is the establishment of laboratories in Sierra Leone and Nigeria that can safely collect and process samples from patients infected with Lassa virus, a BSL-4 pathogen.

In midsummer 1986, five years after starting my poliovirus laboratory at Columbia University, I received a letter from Frederick L. Schaffer, a virologist at the University of California, Berkeley, asking if I would like to have his collection of poliovirus stocks. He was retiring and the samples needed a home, otherwise they would be destroyed. Of course I jumped at the opportunity to have a bit of virology history.

Three boxes full of dry ice arrived in the laboratory in August 1986. In them were sixty-six containers of different polioviruses that Dr. Schaffer had collected over the years. All three poliovirus serotypes were represented, both wild type and vaccine strains. The tubes were marked with dates ranging from 1948 to 1965. Most had come into Dr. Schaffer’s hands from well known poliovirologists, including Jonas Salk, Albert Sabin, Igor Tamm, Renato Dulbecco, and Charles Armstrong.

All of the samples were in glass containers, either tubes with a screw cap or a rubber stopper held in place with tape (pictured). A few were in glass bottles of the type that were used to grow cells. The collection held not a single plastic tube: these were the days before plastics entered the virology world. All were identified by hand-written labels on white cloth tape. Some of the labels in the photo read: Polio 1 Brunhilde (1963), Polio 2 MEF1, Polio 2 P712, HeLa P3, 10-21-62. Nearly all the samples were virus-containing cell culture supernatants.

Perhaps the most amazing sample was a specimen labeled ‘Lansing poliomyelitis virus, 9/27/50, passage 379, C. Armstrong, NIH’. Within the glass vial was a intact mouse brain still attached to the spinal cord (there were no cell phones in 1986, hence no photo). The Lansing type 2 strain of poliovirus had been adapted to grow in mice by Charles Armstrong, and this was apparently the 379th intracerebral mouse-to-mouse passage! It was especially exciting for me to receive this sample, because my laboratory had been studying the ability of the Lansing strain of poliovirus to infect mice. I believe that the note accompanying the tube is in Dr. Armstrong’s writing.

I have since transferred most of the samples to plastic tubes which are stored in a -70C freezer. There is a trove of information to be obtained by studying these samples, but there are few poliovirologists left who are interested. Once poliomyelitis is eradicated – perhaps within the next 10 years – these samples, and similar ones throughout the world, will have to be destroyed.

On episode #349 of the science show This Week in Virology, Vincent, Alan and Rich explain how to make a functional ribosome with tethered subunits, and review the results of a phase III VSV-vectored Ebolavirus vaccine trial in Guinea.

You can find TWiV #349 at www.twiv.tv.

An Ebolavirus vaccine has shown promising results in a clinical trial in Guinea. This vaccine has been in development since 2004 and was made possible by advances in basic virology of the past 40 years.

The ability to produce the Ebolavirus vaccine, called rVSV-EBOV, originates in the 1970s with the discovery of the enzyme reverse transcriptase, the development of recombinant DNA technology, and the ability to rapidly and accurately determine the sequence of nucleic acids. These advances came together in 1981 when it was shown that cloned DNA copies of RNA viral genomes (a bacteriophage, a retrovirus, and poliovirus), carried in a bacterial plasmid, were infectious when introduced into mammalian cells. Production of an infectious DNA copy of the genome of vesicular stomatitis virus (VSV) was reported in 1995. In their paper the authors noted:

Because VSV can be grown to very high titers and in large quantities with relative ease, it may be possible to genetically engineer recombinant VSVs displaying foreign antigens. Such modified viruses could be useful as vaccines conferring protection against other viruses.

This technology was subsequently used in 2004 to produce replication competent VSV carrying the genes encoding the glycoproteins of filoviruses, which others had shown are the targets of neutralizing antibodies. When injected into mice, these recombinant viruses induced neutralizing antibodies that were protective against lethal disease after challenge with Ebolavirus.

In a series of experiments done over the next 10 years, rVSV-EBOV was shown to protect nonhuman primates from lethal disease. In these experiments, animals were injected intramuscularly with the vaccine and challenged with Ebolavirus. The vaccine induced protection against lethal disease and prevented viremia. Extensive studies of the VSV vector in ~80 nonhuman primates showed no serious side effects, and only transient vector viremia.

The rVSV-EBOV was originally developed by Public Health Agency of Canada, and subsequently licensed to NewLink Genetics. Financial support has been provided from Canadian and US governments and others. From 2005 to the present, the NIH Rocky Mountain Laboratory in Hamilton, Montana has also been involved in this work, particularly with nohuman primate challenge studies. In November 2014 Merck entered an agreement with NewLink to manufacture and distribute the vaccine.

In August 2014, well into West Africa Ebolavirus outbreak, Canada donated 800 vials of vaccine to WHO, which then established the VSV Ebola Consortium (VEBCON) to conduct human trials.

The results of Phase I trials of rVSV-EBOV in Africa (Gabon, Kenya) and Europe (Hamburg, Geneva) were published on 1 April 2015. These trials comprised three open-label, dose-escalation trials, and one randomized, double blind controlled trial in 158 adults. Each volunteer was given one injection of 300,000 to 50 million plaque-forming units of rVSV-EBOV or placebo. No serious vaccine related events were reported, but immunization was accompanied by fever, joint pain, and some vesicular dermatitis. A transient systemic infection was observed, followed by development of Ebolavirus-specific antibody responses in all participants, and neutralizing antibodies in most.

The interim results of a phase III trial of rVSV-EBOV, begun on 23 March 2015 in Guinea, have just been published. It is a cluster-randomized trial with a novel design that is modeled on the ring vaccination approach used for smallpox eradication in the 1970s. In ring vaccination, individuals in the area of an outbreak are immunized, in contrast to treating a larger segment of the population. During this trial, when a case of Ebolavirus infection was identified, all contacts and contacts-of-contacts were identified. Some of these individuals were immediately immunized intramuscularly with 2 x 10⁷ PFU, and others (randomly chosen) were immunized three weeks later. The primary outcome was Ebolavirus disease confirmed by PCR. As new cases arose in other areas (clusters), these were treated in the same way, hence the name of cluster-randomized trial.

The press has widely reported that the vaccine was ‘100% protective’. This outcome sounds much better than is represented by the data, so let’s look at the numbers.

Zero cases of Ebolavirus disease were observed in 2,014 immediately vaccinated people, while 16 cases were identified in those given delayed vaccine (n=2,380). These numbers were used to calculate the vaccine efficacy of 100%. While statistically significant, the numbers are small.

More telling are the results obtained when we consider all individuals eligible for immunization, not just those who were immunized (some were excluded for a variety of reasons). Of 4,123 eligible individuals, 2,014 were immunized as noted above, but 2,109 did not receive vaccine. Eight cases of Ebola virus disease were noted in the non-immunized population. This number is small, a consequence of the fact that the outbreak is waning.

On the basis of these interim results, the data and safety monitoring board decided that the trial should continue. However because the board felt that the vaccine is a success, they decided to curtail randomization of subjects into immediately vaccinated and delayed vaccinated groups. Now all contacts and contacts-of-contacts will immediately receive vaccine. As a consequence of this change, it will not be possible to improve the accuracy of vaccine efficacy. For example, when many more individuals are immunized in the future, many fewer that 100% might be protected from disease.

There are two lessons I would like you to remember from this brief history of an Ebolavirus vaccine. Developing a vaccine takes a long time (minimum 11 years for rVSV-EBOV) and depends on advances made with both basic and clinical research.  Don’t believe anyone who says that this vaccine was made in a year. And always look at the numbers when you hear that a vaccine has 100% efficacy.

On episode #348 of the science show This Week in Virology, Vincent and Rich discuss fruit fly viruses, one year without polio in Nigeria, and a permissive Marek’s disease viral vaccine that allows transmission of virulent viruses.

You can find TWiV #348 at www.twiv.tv.

A permissive vaccine prevents disease in the immunized host, but does not block virus infection. Would a permissive vaccine lead to the emergence of more virulent viruses?

This hypothesis is based on the notion that viruses which kill their hosts too quickly are not efficiently transmitted, and are therefore removed by selection. However a vaccine that prevents disease, but not viral replication in the host, would allow virulent viruses to be maintained in the host population. It has been suggested that in this scenario, viruses with increased virulence would be selected if such a property aids transmission between hosts.

On the surface this hypothesis seems reasonable, but in my opinion it is flawed. One problem is that increased transmission might not always be associated with increased virulence. The more serious flaw lies in making anthropomorphic assessments of what we think viruses require, such as concluding that increased viral transmission is a desired trait. Our assumptions fail to recognize the main goal of evolution: survival. Evolution does not move a virus along a trajectory aimed at perfection. Change comes about by eliminating those viruses that are not well adapted for the current conditions, not by building a virus that will fare better tomorrow. All the viruses on Earth today transmit well enough, or they would not be here; yet some kill their hosts clearly much faster than others. The fact is that humans have little understanding of what drives virus evolution in large populations. Our assumptions of what constitute the selective forces are usually tainted by anthropomorphism.

This long preamble is an introduction to a series of findings which are purported to support the idea that permissive vaccines (the authors call them ‘leaky’ and ‘imperfect’ vaccines but I dislike both names because they imply defects) can lead to the selection of more virulent viruses. The subject of the paper is Marek’s disease virus (MDV), a herpesvirus that infects chickens. MDV is shed from feather follicles of infected chickens and is spread to other birds when then inhale contaminated dust. Vaccines have been used to prevent MDV infection since the early 1970s. These vaccines prevent disease, but do not block viral replication, and vaccinated, infected birds can shed wild type virus. The virulence of MDV has been increasing since the 1950s, initially from a paralytic disease, to paralysis and death. The authors wonder if the use of permissive Marek’s vaccines has lead to the selection of more virulent viruses.

To address their hypothesis, the authors inoculate vaccinated or unvaccinated chickens with a series of MDV isolates that range from low to high virulence. Unvaccinated chickens inoculated with the most virulent MDV died within a week and shed little virus. In contrast, most vaccinated birds survived infection with virulent viruses, and shed virus for the length of the experiment, 56 days.

A transmission experiment was done to determine if shed virus could infect other birds. The authors infected vaccinated or unvaccinated birds and asked if sentinel, unvaccinated chickens became infected. Unvaccinated birds died within 10 days after infection with virulent MDV, and did not transmit infection. In contrast, vaccinated birds survived at least 30 days, and co-housed sentinel animals became infected and died.

The experiments are well done and the conclusions are clear: more virulent Marek’s disease viruses replicate longer in vaccinated than unvaccinated chickens, and can be readily transmitted to other chickens. But these results do not prove that more virulent MDV arose because of permissive vaccines. Nor do the results prove in general that leaky vaccines lead to selection of more virulent viruses. The results simply show that a vaccine that does not prevent replication will allow transmission of virulent viruses.

To prove that vaccinated chickens can allow the selection of more virulent viruses, vaccinated chickens could be infected with an avirulent virus, and the shed virus collected and used to infect additional, vaccinated birds. This process could be repeated to determine if more virulent viruses arise. While the results of this gain-of-function experiment would be informative, they would be done in a controlled laboratory setting which would not duplicate all the selective forces present on a poultry farm.

The authors note that most human vaccines do prevent replication of infecting virus. They do not mention the one important exception: the Salk poliovirus vaccines. People who are immunized with the Salk vaccine can be infected with poliovirus, which will then replicate in the intestines, be shed in the feces, and transmitted to others. This behavior has been well documented in human populations, yet the virulence of poliovirus has not increased for the 60 years during which the Salk vaccine has been used.

I do not feel that these experimental results have general implications for the use of any animal vaccine. It is unfortunate that the work has been covered in many news sources with the incorrect implication that vaccines may be responsible for the emergence of more virulent viruses.

I am pleased to announce the publication by ASM Press of the fourth edition of our virology textbook, Principles of Virology. Two years in the making, this new edition is fully updated to represent the rapidly changing field of virology.

Principles of Virology has been written according to the authors’ philosophy that the best approach to teaching introductory virology is by emphasizing shared principles. Studying the phases of the viral reproductive cycle, illustrated with a set of representative viruses, provides an overview of the steps required to maintain these infectious agents in nature. Such knowledge cannot be acquired by learning a collection of facts about individual viruses. Consequently, the major goal of this book is to define and illustrate the basic principles of animal virus biology.

This edition is marked by a change in the author team. Our new member, Glenn Rall, has brought expertise in viral immunology and pathogenesis, pedagogical clarity, and down-to-earth humor to our work. Although no longer a coauthor, our colleague Lynn Enquist has continued to provide insight, advice, and comments on the chapters.

A major new feature is the inclusion of 26 video interviews with leading scientists who have made significant contributions to the field of virology. These in-depth interviews provide the background and thinking that went into the discoveries or observations connected to the concepts being taught in this text. Students will discover the personal stories and twists of fate that led the scientists to work with viruses and make their seminal discoveries.

Principles of Virology is ideal for teaching the strategies by which all viruses reproduce, spread within a host, and are maintained within populations. It is appropriate for undergraduate courses in virology and microbiology as well as graduate courses in virology and infectious diseases. I have used previous editions of this textbook to build my Columbia University virology course. Volume I: Molecular Biology covers the molecular biology of viral reproduction. Volume II: Pathogenesis & Control addresses the interplay between viruses and their host organisms. The two volumes can be used for separate courses or together in a single course. Each includes a unique appendix, glossary, and links to Internet resources such as websites, podcasts, and blogs.

PoV4 goes on sale the week of 24 August 2015. If you are thinking about using the book for your course, reserve your review copy today at http://www.asm.org/pov.

Watch the video below to hear authors Jane Flint, Vincent Racaniello, Glenn Rall, and Ann Skalka talk about the making of PoV 4.

On episode #347 of the science show This Week in Virology, Vincent, Alan, and Rich discuss the virus behind rose rosette disease, and fatal human encephalitis caused by a variegated squirrel bornavirus.

You can find TWiV #347 at www.twiv.tv.

Image credit

Foot-and-mouth disease virus (FMDV) infects cloven-hoofed animals such as cattle, pigs, sheep, goats, and many wild species. The disease caused by this virus is a substantial problem for farmers because infected animals cannot be sold. Transgenic pigs have now been produced which express a short interfering RNA (siRNA) and consequently have reduced susceptibility to infection with FMDV.

FMDV is classified in the picornavirus family which also contains poliovirus and rhinoviruses. The virus is highly contagious and readily spreads long distances via wind currents, and among animals by aerosols and contact with farm equipment. Infection causes a high fever and blisters in the mouth and on the feet – hence the name of the disease. When outbreaks occur, they are economically devastating. The 2001 FMDV outbreak in the United Kingdom was stopped by mass slaughter of all animals surrounding the affected areas – an estimated 6,131,440 – in less than a year.

Vaccines against the virus can be protective but they are not an optimal solution. One problem is that antigenic variation of the virus may thwart protection. In addition, countries free of FMDV generally do not vaccinate because this practice would make the animals seropositive and prevent their export (it is not possible to differentiate between antibodies produced by natural infection versus immunization). Furthermore, if there were an outbreak of foot-and-mouth disease in such countries, the rapid replication and spread of the virus would make vaccination ineffective – hence culling of animals as described above is required. Clearly other means of protecting animals against FMDV are needed.

Synthetic short interfering RNAs (siRNA) have been shown to block viral replication in cell culture and in animals. To achieve such inhibition, short synthetic RNAs complementary to viral sequences are produced in cells. Upon infection, these siRNAs combine with the cellular RNA-induced silencing complex (RISC) which then targets the viral RNA for degradation.

To determine if siRNA could be used to protect pigs from foot-and-mouth disease, a complementary viral sequence was first identified that blocks FMDV replication in cell culture by ~97%. A vector containing this siRNA sequence was then used to produce transgenic pigs. Such animals not only express the antiviral siRNA, but as the encoding vector is present in germ cells, it is passed on to progeny pigs.

Expression of the siRNA was confirmed in a variety of transgenic pig tissues, including heart, lung, spleen, liver, kidney, and muscle. In fibroblasts produced from transgenic pigs, virus replication was reduced 30 fold. When transgenic pigs were inoculated intramuscularly with FMDV, none of the animals developed signs of disease such as fever or blisters of the feet and nose. In contrast, control non-transgenic pigs developed high fever and lesions. Viral RNA levels in the blood of transgenic pigs were 100-fold lower than in control animals. At 10 days post-infection no viral RNA was detected in heart, lung, spleen, liver, kidney, and muscle, while high levels were observed in these organs from non-transgenic controls.

These results show that siRNAs can protect transgenic pigs from FMDV induced disease. An important question that must be answered is whether transgenic pigs still contain enough virus to transmit infection to other animals. In addition, siRNAs are short – 21 nucleotides – and a mutation in the viral genome can block their inhibitory activity. Therefore it would be important to determine if mutations arise in the FMDV genome that lead to resistance to siRNAs.

Even if transgenic siRNA pigs do not transmit infection, and viral resistance does not arise, I am not sure that consumers are ready to accept such genetically modified animals.

Episode #346 of the science show This Week in Virology was recorded at the 34th Annual Meeting of the American Society for Virology, where Vincent, Rich, and Kathy spoke with Joan Steitz, a tireless promoter of women in science and one of the greatest scientists of our generation.

You can find TWiV #346 at www.twiv.tv.

On episode #345 of the science show This Week in Virology, the TWiVonauts review how the weather affects West Nile virus disease in the US, benefit of B cell depletion for ME/CFS patients, and an autoimmune reaction induced by influenza virus vaccine that leads to narcolepsy.

You can find TWiV #345 at www.twiv.tv.

Patients with myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) showed clinical improvement after extended treatment with the anti-B-cell monoclonal antibody rituximab. This result suggests that in a subset of patients, ME/CFS might be an autoimmune disease.

Rituximab is a monoclonal antibody against a protein on the surface of B cells known as CD20. When the antibody is given to patients, it leads to destruction of B cells, which are the producers of antibodies, proteins that are made by the immune system to counter infections. The drug has been approved by the US Food and Drug administration to treat diseases of B cells such as lymphomas, leukemias, and autoimmune conditions.

ME/CFS is a disease of unknown etiology and mechanism that includes symptoms of severe fatigue, post-exertional malaise, pain, cognitive and sleep problems that affects 0.1-0.2% of the population. A previous randomized, phase II trial of rituximab treatment showed clinical benefit in 20 of 30 patients. The improvements were evident 2-8 months after treatment, leading the study authors to suggest that remission requires elimination of long-lived antibodies after depletion of B cells.

The current study was done to determine the effects of sustained treatment with rituximab. Included patients (29) were 18-66 years of age and diagnosed with ME/CFS according to Fukuda 1994 criteria. All were given rutiximab infusions two weeks apart, then at 3, 6, 10, and 15 months, and followed up for 36 months. Self-reported symptoms were recorded every second week and used to calculate scores for fatigue (comprising post-exertional malaise, need for rest, daily functioning), pain (muscle, joint, and cutaneous pain and headache) and cognitive scores (concentration ability, memory disturbance, mental tiredness).

Clinically significant responses were found in 18/29 patients (64%), with a lag of 8-66 weeks. After 36 weeks 11 of 18 responding patients were still in clinical remission. Nine patients from the placebo group in the previous study were included in this trial; of these, six had clinical improvement.

These results show that some ME/CFS patients benefit from ablating B cells. The delayed response, coupled with the relapse after cessation of treatment and B cell regeneration, suggests that antibodies are involved in the pathogenesis of the disease. Because onset of ME/CFS in many patients correlates with a viral infection, it is possible that antibodies to viral proteins may cross-react with self proteins, leading to autoimmune reactions that cause disease. Treatment with rituximab would lead to reduced levels of such antibodies, thereby reducing symptoms.

These results warrant trials of larger numbers of ME/CFS patients in other countries (this study was carried out in Norway) to determine if ablation of B cells would have a similar effects elsewhere. It would also be useful to determine the total repertoire of antiviral antibodies produced by ME/CFS patients. Such antibodies can be identified using the newly developed VirScan assay, which requires a small amount of blood and is relatively inexpensive. The results will indicate whether certain viral infections in a large population of ME/CFS patients predispose to the illness. Furthermore, the results may also be used to guide efforts to determine whether such antibodies react with human cellular proteins. A similar approach was used to determine that antibodies to an influenza virus protein cross react with a neuropeptide receptor, leading to narcolepsy.

While these findings are promising, they also show that not all ME/CFS may involve autoimmune pathogenesis. Other creative approaches will be needed to determine the cause of disease in individuals who do not respond to rituximab.

Episode #344 of the science show This Week in Virology was recorded at the Glasgow Science Festival microTALKS, where Vincent spoke with Ruth, Glen, and Esther about their research on viruses and Hodgkin lymphoma, adenovirus structure and entry into cells, and interactions between arthropod borne viruses and their hosts.

You can find TWiV #344 at www.twiv.tv.

Back in 2013 I built a Wall of Polio in my laboratory – a large stack of six-well cell culture plates that have been used to measure the concentration of polioviruses in various samples by plaque assay. It became a focal point of the lab at which many guests came to have their photographs taken. Sadly, the Wall fell twice. Now a new Wall – version 3.0 – has been completed.

The new Wall of Polio is in my office at Columbia University Medical Center, where it will not annoy the Fire Inspector (the former Wall partially blocked an aisle). Furthemore, the new Wall is glued together, so it will not come apart. Its construction is documented in the photographs below. The Wall of Polio 3.0 is built with 1,464 six-well plates of HeLa cells that were used to determine the titer of poliovirus. We have also already had a number of visitors to Wall 3.0.

Because the Wall is impressive, it attracts attention, which can then be used to explain the plaque assay and determining virus titer. Therefore it is simply another tool that I used to teach the world about virology.

When you visit, expect that I will ask to photograph you before the Wall. Only a few have refused.

On episode #343 of the science show This Week in Virology, the TWiVerinoes discuss the potential for prion spread by plants, global circulation patterns of influenza virus, and the roles of Argonautes and a viral protein in RNA silencing in plants.

You can find TWiV #343 at www.twiv.tv.

Chronic wasting disease is a prion disease of cervids (deer, elk, moose) that is potentially a threat to human health. A role for environmental prion contamination in transmission is supported by the finding that plants can take up prions from the soil and transmit them to animals.

A concern is that prions of chronic wasting disease could be transmitted to cows grazing in pastures contaminated by cervids. Consumption of infected cows would then pass the disease on to humans. When deer are fed prions they excrete them in the feces before developing clinical signs of infection, and prions can also be detected in deer saliva. In the laboratory, brain homogenates from infected deer can transmit the disease to cows.

To determine whether prions can enter plants, wheat grass roots and leaves were exposed to brain homogenates from hamsters that had died of prion disease. The plant materials were then washed and amounts of prions were determined by protein misfolding cyclic amplification. Prions readily bound these plant tissues, at low concentrations and after as little as 2 minutes of incubation. Mouse, cervid, and human prions also bound to plant roots and leaves. When living wheat grass leaves were sprayed with a 1% hamster brain homogenate, prions could attach to the leaves and be detected for 49 days.

To determine if prions in plants could infect animals, plants were exposed to brain homogenates, washed thoroughly, and then fed to hamsters. The positive control for this experiment was to feed hamsters the brain homogenates. All animals fed infected plants or brain homogenates succumbed to prion disease.

Plants can also take up prions from animal waste. This conclusion was reached by incubating leaves and roots for 1 hour with urine or feces obtained from prion-infected hamsters or cervids. Prions were readily detected in these samples, even after extensive washing.

Experiments were also done to examine whether plants could take up prions from the soil. Barley grass plants were grown on soil that had been mixed with hamster brain homogenate, and then 1-3 weeks later, stem and leaves were assayed for the presence of prions. Small amounts of prions were detected in stems from all plants, while 1 in 4 plants contained prions in leaves, at levels that should be able to infect an animal.

These results show that prions can bind to plants and be taken into the roots, where they may travel to the stem and leaves. Therefore it is possible that prions excreted by deer could pass on to other animals, such as grazing cows, or even humans consuming contaminated plants (illustrated – image credit). Cooking plants will not eliminate infectivity, just as cooking contaminated beef did not halt the spread of bovine spongiform encephalopathy. Keeping cervids out of grazing or growing fields should be considered as a way to manage the risk of prions entering the human food chain.

On episode #342 of the science show This Week in Virology, the TWiVniks discuss the structure of a virus that reproduces in an extreme environment, long-term consequences of Ebolavirus infection, and VirScan, a method to identify the different virus infections you have had in your lifetime.

You can find TWiV #342 at www.twiv.tv.

Virologist Richard Elliott passed away on 5 June 2015. I have known Richard since 1979 and I would like to provide some personal recollections of this outstanding virologist. A summary of his work can be found at the MRC-University of Glasgow Centre for Virus Research science blog.

I first met Richard in 1979 when he joined Peter Palese’s laboratory at Mt. Sinai School of Medicine in New York City. We overlapped for only about a year but it was enough to get to know him: he was a hard-working, enthusiastic virologists and a good friend. We shared many beers in New York City. At the end of 1979 I went off to David Baltimore’s laboratory where in 1981 I produced an infectious DNA clone of poliovirus. It proved very difficult to make infectious DNAs of negative strand RNA viruses, and it was Richard who was the first to accomplish this feat in 1996 for a virus with a segmented genome. This work was very important as it showed that infectious DNA clones were not limited to RNA viruses with monopartite genomes.

I remained in contact with Richard over the years but I did not see him in person until the 2010 meeting in Edinburgh of the Society for General Microbiology. The following year he joined me, Connor Bamford, Wendy Barclay, and Ron Fouchier for TWiV #177 recorded in Dublin. Schmallenberg virus had just emerged as a new pathogen of livestock, and he discussed his work on this virus.

I next saw Richard in a 2011 meeting of the Brazilian Society for Virology. When I arrived in Brazil it was quite hot, and I found Richard sitting by the pool, reviewing manuscripts in his bathing suit. I snapped a few photos of him and put them on Facebook. Later that evening he said his laboratory had asked why pornographic photos of him were on the internet – he was shirtless in my pictures (with Grant McFadden in the photo).

Richard visited New York in the summer of 2014 but we were unable to connect. Early this year Richard had agreed to join TWiV again for an episode from Glasgow. Sadly he became too ill to participate and died on the Friday before I traveled to Scotland. While there I briefly visited the Elliott lab at the University of Glasgow MRC-Centre for Virus Research, nearly a week after his death. I’m happy that I made the lab members smile:

I can still remember Richard telling me how to spell his name: two ls, two ts. Richard was an excellent virologist, mentor, and friend. I will miss him.

Update: Corrected to reflect the fact that Richard produced the first infectious DNA of a segmented (-) strand RNA virus.

On episode #341 of the science show This Week in Virology, Vincent returns to the University of Glasgow MRC-Center for Virus Research and speaks with Emma, Gillian, and Adam about their ebolavirus experiences: caring for an infected patient, working in an Ebola treatment center in Sierra Leone, and making epidemiological predictions about the outbreak in west Africa.

You can find TWiV #341 at www.twiv.tv.

Transmissible spongiform encephalopathies (TSEs) are rare human neurodegenerative disorders that are caused by infectious proteins called prions. A naturally occurring variant of the human prion has been found that completely protects against the disease.

A protective variant of the prion protein was discovered in the Fore people of Papua New Guinea. Beginning in the early 1900s, the prion disease kuru spread among Fore women and children as a result of ritual cannibalism of the brains of deceased relatives. When cannibalism stopped in the late 1950s, kuru disappeared.

Survivors of the kuru epidemic are heterozygous for a prion protein gene (prnp) with a unique amino acid change not seen in other populations, a change at position 127 from glycine to valine (G127V). The G127V change was always seen together with methionine at 129. Heterozygosity for M and V at amino acid 129, which is protective against prion disease, is found in humans all over the world.

Transgenic mice were used to determine if the G127V change in prion protein protects against disease. These mice lack the murine prnp gene (which encodes the normal prion protein) and contain a copy of either the wild type human prnp gene, or one with changes at amino acids 127 and 129. The mice were then inoculated intracerebrally with brain extracts from individuals who died of kuru. Mice with wild type human prnp were susceptible to infection. In contrast, transgenic mice heterozygous for the variant prnp (G127M129/V127M129) were completely resistant to infection. The mice were also resistant to infection with prions from cases of another human TSE, Creutzfeldt-Jacob disease.

Prnp transgenic mice were also challenged with variant Creutzfeldt-Jacob disease prions. This novel TSE arose after consumption of beef from animals with the prion disease bovine spongiform encephalopathy (BSE). These mice were susceptible to infection with vCJD prions, not a surprising result given that the Fore people were never exposed to BSE prions. However, mice homogygous for the altered prnp (V127M129/V127M129) were completely resistant to infection with vCJD prions – as resistant as mice with no prnp genes.

The protective effect of the M129V polymorphism is thought to be a consequence of inhibition of protein-protein interactions during prion propagation (i.e. the conversion of normal prion to pathogenic prion). How the G127V change confers protection is unknown.

These results show that the G127V change confers resistance to kuru and was likely selected as a consequence of the epidemic. If kuru had not been stopped by the abolition of cannibalism, it likely would have been self-limiting, as individuals with resistance to the disease, caused by the G127V change, repopulated the Fore people.

On episode #340 of the science show This Week in Virology, the TWiV teams reviews a MERS-coronavirus serosurvey and an outbreak in South Korea, and constraints on measles virus antigenic variation.

You can find TWiV #340 at www.twiv.tv.

Did you ever wonder what different virus infections you have had in your lifetime? Now you can find out with just a drop of your blood and about $25.

Immune defense systems of many hosts produce antibodies in response to virus infections. These large proteins, which are generally virus specific, can block or inhibit virus infection, and persist at low levels for many years after the initial infection. Hence it is possible to determine whether an individual has had a virus infection by looking for anti-viral antibodies in the blood. Up to now the process of identifying such antibodies has been slow and limited to one or a few viruses. A new assay called VirScan allows unbiased searches for all the virus antibodies in your blood, providing a picture of all your past infections.

To identify the human antivirome, DNAs were synthesized encoding proteins from all viruses known to infect humans – 206 species and over 1000 strains. These DNAs were inserted into the genome of a bacteriophage, so that upon infecting bacteria, the viral peptides are displayed on the phage capsid. These ‘display’ phages were then mixed with human serum, and those that were bound by antibodies were isolated. The DNA sequence of the phage genomes were then determined to identify the human virus bound by the antibodies.

This method was used to assay samples from 569 humans. The results show that each person had been exposed to an average of 10 viruses, with a range from a few to over 20 (two individuals had antibodies to 84 different virus species!). The most frequently identified viruses included herpesviruses, rhinoviruses, adenoviruses, influenza viruses, respiratory syncytial virus, and enteroviruses. The overall winner, found in 88% of samples, is Epstein-Barr virus.

These results are not unexpected: all of us are infected with at least a dozen viruses at any time, and the viruses identified in this study known to infect much of the human population. What was surprising is the absence of some common viruses, such as rotaviruses, and the ubiquitous polyomaviruses. According to serological surveys, the most common human viruses are the small, single-stranded DNA containing anelloviruses. Yet the related torque teno virus was only found in 1.7% of samples. These differences are likely due to a combination of technical and biological issues (e.g., failure of antibodies to certain viruses to persist in serum).

This new assay may one day become a routine diagnostic tool that is used along with complete blood counts and chemistries to know if a patient’s signs and symptoms might be attributable to a past virus infection. VirScan technology is not limited to virus infections – it can be used to provide a history of bouts with bacteria, fungi, and parasites.

VirScan might also allow us to determine which virus infections are beneficial, and which contribute to chronic diseases such as autoimmune or neurodevelopmental disorders or cancer. The assay can be used to conduct unbiased population-based studies of the prevalence of virus infections and their possible association with these diseases. Such connections were not previously possible with antibody assays that search for one virus at a time. This approach was not only inefficient, but required guessing the responsible virus.

Some other findings of this study are noteworthy. As expected, children had fewer virus infections than adults. HIV-positive individuals had antibodies to more viruses than HIV-negative individuals, also expected given the damage done by this virus to the immune system. Frequencies of anti-viral antibodies were higher outside of the United States, possible due to differences in genetics, sanitation, or population density. In most samples, there was a single dominant peptide per virus, although there were occasional differences among populations. This information might be useful for improving vaccines, or tailoring them to specific countries or regions.

Update: It would be very informative to use VirScan to search for antibodies against viruses that are not known to infect humans. Other animal viruses, plant viruses, insect viruses: to which do a significant fraction of humans respond? The information might identify other viruses that replicate in humans and which might constitute future threats (or present benefits).

On episode #339 of the science show This Week in Virology, tre TWiV amici present three snippets and a side of sashimi: how herpesvirus inhibits host cell gene expression by disrupting transcription termination.

You can find TWiV #339 at www.twiv.tv.

On episode #337 of the science show This Week in Virology, Vincent meets up with Michael and Steve to discuss their finding of a transmissible tumor in soft-shell clams associated with a retrovirus-like element in the clam genome.

You can find TWiV #337 at www.twiv.tv.

On episode #336 of the science show This Week in Virology, the TWiVsters explore mutations in the interferon pathway associated with severe influenza in a child, outbreaks of avian influenza in North American poultry farms, Ebolavirus infection of the eye weeks after recovery, and Ebolavirus stability on surfaces and in fluids.

You can find TWiV #336 at www.twiv.tv.

A major goal of viral oncotherapy – the use of viruses to destroy tumors –  is to design viruses that kill tumor cells but not normal cells. Two adenoviruses provide perfect examples of how this specificity can be achieved.

Adenovirus CG0070, designed to treat bladder cancer, and adenovirus Oncorine, for head and neck tumors, replicate only in tumor cells. The selectivity is caused by mutations introduced into the viral genomes.

When adenovirus infects a cell, the first event is synthesis of mRNA that encodes the E1 proteins. These proteins are needed to start cellular DNA synthesis. Most cells in our bodies are not dividing, an environment not conducive to viral replication. The adenovirus E1 proteins solve this problem. The E1A protein binds the cellular Rb (retinoblastoma) protein, which is normally bound to members of the E2f family of transcription factors (illustrated, upper left). Binding of E1A to Rb frees E2f which goes on to induce the transcription of cell genes needed for DNA synthesis and cell division.

The genome of CG0070 (illustrated below) has been modified so that the promoter for mRNA synthesis of the E1 proteins is replaced by the viral E2f promoter. This promoter requires E2f transcription factors for activity; hence the promoter does not function in non-dividing cells in which Rb is bound to E2f. However, many tumors lack Rb, and E2f is always available. CG0070 will replicate in such tumor cells.

The genome of adenovirus Oncorine lacks the early region protein E1b-55K. The function of this viral protein is to bind the cellular protein p53, which would otherwise halt division and induce death of the infected cell. Binding to p53 leads to its degradation, allowing the virus to execute its 24 hour reproductive cycle. Adenovirus lacking the E1b-55K protein will not replicate in normal cells. However, the virus will replicate in p53 deficient tumors.

Oncorine has been licensed in China for the treatment of head and neck tumors, while CG0070 is in phase III clinical studies for the treatment of bladder cancer. Both oncolytic adenoviruses were developed by using knowledge of fundamental aspects of viral replication, yet another illustration of how basic research can lead to clinical applications.

It is the first week in May, which means that the spring semester has just ended at Columbia University, and my annual virology course is over.

Each year I teach an introductory undergraduate virology course that is organized around basic principles, including how virus particles are built, how they replicate, how they cause disease, and how to prevent infections. Some feel that it’s best to teach virology by virus: a lecture on influenza, herpesvirus, HIV, and on and on. But this approach is all wrong: you can’t learn virology by listening to lectures on a dozen different viruses. In the end all you will have is a list of facts but you won’t understand virology.

I record every one of my 26 introductory lectures as a videocast, and these are available on the course website, or on YouTube. If you have listened to my lectures before, you might be wondering what is new. I change about 10% of each lecture every year, updating the information and adding new figures. This year I’ve also added two new lectures, on on Ebolavirus and one on viral gene therapy.

Once you have taken my introductory course, then you will be ready for an advanced course on Viruses. A course in which we go into great detail on the replication, pathogenesis, and control of individual viruses. I am working on such a course and when it’s ready I’ll share it with everyone.

I want to be Earth’s virology professor, and this is my introductory virology course for the planet.

On episode #335 of the science show This Week in Virology, the TWiVumvirate discusses a whole Ebolavirus vaccine that protects primates, the finding that Ebolavirus is not undergoing rapid evolution, and a proposal to increase the pool of life science researchers by cutting money and time from grants.

You can find TWiV #335 at www.twiv.tv.

Rhinovirus is the most frequent cause of the common cold, and the virus itself is quite common: there are over 160 types, classified into 3 species. The cell receptor has just been identified for the rhinovirus C species, which can cause more severe illness than members of the A or B species: it is cadherin-related family member 3.

On episode #334 of the science show This Week in Virology, the TWiVles talk about endogenous viruses in plants, sex and Ebolavirus transmission, an outbreak of canine influenza in the US, Dr. Oz, and doubling the NIH budget.

You can find TWiV #334 at www.twiv.tv.

About eight percent of human DNA is viral: it consists of retroviral genomes produced by infections that occurred many years ago. These endogenous retroviruses are passed from parent to child in our DNA. Some of these viral genomes are activated for a brief time during human embryogenesis, suggesting that they may play a role in development.

There are over 500,000 endogenous retroviruses in the human genome, about 20 times more than human genes. They were acquired millions of years ago after retroviral infection. In this process, viral RNA is converted to DNA, which then integrates into cell DNA. If the retroviral infection takes place in the germ line, the integrated DNA may be passed on to offspring.

The most recent human retroviral infections leading to germ line integration took place with a subgroup of human endogenous retroviruses called HERVK(HML-2). The human genome contains ~90 copies of these viral genomes, which might have infected human ancestors as recently as 200,000 years ago. HERVs do not produce infectious virus: not only is the viral genome silenced – no mRNAs are produced – but they are littered with lethal mutations that have accumulated over time.

A recent study revealed that HERVK mRNAs are produced during normal human embryogenesis. Viral RNAs were detected beginning at the 8-cell stage, through epiblast cells in preimplantation embryos, until formation of embryonic stem cells (illustrated). At this point the production of HERVK mRNA ceases. Viral capsid protein was detected in blastocysts, and electron microscopy revealed the presence of virus-like particles similar to those found in reconstructed HERVK particles. These results indicate that retroviral proteins and particles are present during human development, up until implantation.

Retroviral particles in blastocysts are accompanied by induction of synthesis of an antiviral protein, IFITM1, that is known to block infection with a variety of viruses, including influenza virus. A HERVK protein known as Rec, produced in blastocysts, binds a variety of cell mRNAs and either increases or decreases their association with ribosomes.

Is there a function for HERVK expression during human embryogenesis? The authors speculate that modulation of the ribosome-binding activities of specific cell mRNAs by the viral Rec protein could influence aspects of early development. As Rec sequences are polymorphic in humans, the effects could even extend to individuals. In addition, HERVK induction of IFITM1 might conceivably protect embryos against infection with other viruses.

The maintenance of open reading frames in HERV genomes, over many years of evolution, suggests a functional role for these elements. Evidence for such function comes from the syncytin proteins, which  are essential for placental development: the genes encoding these proteins originated from HERV glycoproteins. However, not all endogenous retroviruses are beneficial: a number of malignant diseases have been associated with HERV-K expression.

On episode #333 of the science show This Week in Virology, Vincent returns to Vanderbilt University and meets up with Ben, Megan, Bobak, and Meredith to learn about life in the Medical Scientist Training Program, where students earn both an MD and a Ph.D.

You can find TWiV #333 at www.twiv.tv.

We also learned that my grandmother married Walter A. Snell, who also appears to be a member of the same Snell family.

Did Helen knowingly marry someone who was her cousin or possibly her half-brother? Did she always know who she was biologically?

Is My Surname Actually SNELL?

On February 17, 1932, Helen’s first husband Frank Bettinger passed away unexpectedly at the age of 59:

On April 9, 1932, just 52 days after the death of Frank, the widow married Walter A. Snell:

While my assumption was that the genetic matching with the Snell family was due to Helen’s father possibly being a Snell, another very real possibility was that the relationship was through Walter A. Snell instead of Helen. This seemed especially likely in view of how quickly Helen married Walter after her first husband’s death.

Were Helen and Walter already intimately involved prior to Frank’s death? Could they have been involved in July of 1915 when Helen’s son (my grandfather) was conceived?

Y-DNA to the Rescue

I’ve done a lot of Y-DNA testing. STR testing, SNP testing, Full Y, Geno 2.0, 23andMe, Y-Seq, you name it. At no point did I think anything was out of the ordinary, nor did I have a Snell match. But that doesn’t prove anything.

However, about a year ago, I sought out a distant Bettinger relative to do a 67-marker Y-DNA test. We’re both descendants of the most distant known Bettinger ancestor, Philip Bettinger. I’m a descendant of Philip’s son Georg/George, and my sixth cousin is a descendant of Philip’s son Leonard:

You can see Frank Bettinger – husband of Helen Johnson and my great-grandfather – in dark red in the graphic.

When the results came back, we were a very close Y-DNA match, having a genetic distance of just 2 (one fast marker and one slow marker).

Additionally, there is autosomal DNA sharing between these two lines, with a fifth cousin once removed (5C1R) relationship between the two lines sharing a 17 cM segment on chromosome 15. Although not definitive, the autosomal DNA matching and Y-DNA matching align well.

So no, I don’t have Snell Y-DNA.

Helen Johnson’s Snell DNA

So if we continue with the hypothesis that Helen’s father was a Snell, can we come up with some possibilities about who he might have been?

Here’s the tree I’ve reconstructed using the two Bettinger test-takers and the four Snell descendants:

The variety of descendants makes for a compelling hypothesis. For example, Match #1 is closest to the two Bettingers genetically (and genealogically, according to this reconstructed tree), while Match #2 and Match #3 are a bit more distant, and Match #4 is the most distant genetically (and genealogically, according to this reconstructed tree). This leads me to think that Edmond Cooley Snell is Helen’s grandfather, rather than another brother or sister of Edmond and Sylvester.

Helen’s father, therefore, could very likely be a child of Edmond Cooley Snell. Based on my preliminary research, Edmond Snell married twice and had nine children, six with his first wife Sophronia Parmelee and three with his second wife Mary Roberts.

Scenario #1

If Helen’s father is a child of Sophronia Parmelee, he will be a full sibling to Mary Elizabeth Snell and Julia Snell. The two Bettinger test-takers would then be third cousins (3C) with Match #1, and third cousins once removed with Match #2 and Match #3.

Scenario #2

If Helen’s father is a child of Mary Roberts, he will be a half sibling to Mary Elizabeth Snell and Julia Snell. The two Bettinger test-takers would then be half-3C with Match #1, and half-3C1R with Match #2 and Match #3.

The following chart shows the two possible relationships for the matches (the genealogical relationship with Match #4 is unchanged in either scenario), the expected amount of sharing for each relationship, and the actual sharing (in cM) between the two Bettinger test-takers and these matches:

Reading the Chart: For example, for Match #1, the relationship could be either 3C or Half-3C. 3C should share about 53 cM in common, and Half-3C should share 27 cM in common. Bettinger test-taker #1 shares 90 cM with Match #1, and Bettinger test-taker #2 shares 44 cM with Match #1.

For each of Match #1 through #3, the results for both Bettinger test-takers suggest that the relationship is not a half relationship but a full relationship, and thus that Marley’s grandmother might have been Sophronia Parmelee.

Unfortunately, all of this research is far too preliminary – and the DNA evidence not conclusive enough – to preclude Edmond Snell’s second wife Mary Roberts as the grandmother. Hopefully with additional evidence, either paper trail or DNA, we will have more conclusive support for one possibility over the other.

Edmond Cooley Snell’s Sons

In the next post, we’ll look at Edmond Cooley’s four sons and who could have been Helen’s father. What were they doing in the summer of 1888 when Helen was conceived?

And if there’s time, we’ll examine the possibility that Helen might have known her biological heritage. Although cousins have frequently married throughout history, it was usually because they knew they were cousins; they interacted with each other and/or lived in the same town.

But if Helen was indeed biologically a Snell, and if she didn’t know it, what are the odds that she randomly reclaimed a birthright she was denied and randomly lived with her biological surname for the last 50 years of her life?

What are the odds that the surname carved into Helen’s gravestone for eternity randomly could be the surname she should have been born with?

The post The Search For Helen’s Roots – Part II appeared first on The Genetic Genealogist.

She was born on March 2, 1889 in Mexico, Oswego County, New York, the unnamed daughter of a “Minerva D. Johnson” (age 20 and born in nearby New Haven, Oswego County, NY) and an unknown father. (New York State Department of Health, birth certificate 8040 (1889), no name; Office of Vital Statistics, Albany).

She died at the age of 93 in 1983 in Watertown, New York. Visiting the elderly Helen (by then known as Marley) is one of my earliest childhood memories.

In an attempt to find Helen’s ancestors, I’m using DNA that I graciously obtained from four of Helen’s grandchildren (my father, two of his sisters, and their first cousin). Last week, I uncovered some possible clues that have raised more questions than I could have ever thought possible. And when DNA is involved, that’s really saying something!

Discovering a Clue Using ICW

In late August 2015, AncestryDNA had just recently launched the new In Common With tool, and I decided to inspect my aunt and my father’s matches. I selected my aunt’s closest unknown match, a predicted 4th cousin that we’ll call R.T., for review:

I quickly reviewed his tree and saw a few surnames that I recognized, although it’s pretty rare to review a tree and not see surnames I recognize. I did notice the following in R.T.’s tree, however:

The “Snell” surname is connected to Helen in a way I’ll talk about at the end of this post (and in later posts), but none of these names seemed familiar. So I decided to see where this Snell family might have been located by clicking on Edmond Cooley Snell:

And then, instantly and without any such expectation, Edmond was an enormous person of interest! Edmond died in 1866 in New Haven, the very same small town where Helen’s mother was reported to have been born in 1869. And just a few miles from where my great-grandmother was born in 1889.

Next I clicked on the SHARED MATCHES button to see all the predicted 4th cousins or closer that my aunt shares with R.T. (Note that AncestryDNA’s In Common With Tool only shows 4th cousins or closer; you can share many more distant matches with someone , but they won’t be identified by the tool).

The SHARED MATCHES showed the following list of individuals:

• My father (my aunt’s brother) – no big surprise there, my father and my aunt share R.T. in common. In fact, R.T. is a predicted 3rd cousin to my father;
• Myself – again, no surprise;
• One of my children – no surprise; and
• Two predicted 4th cousins that we’ll call Susanne and John.

1. Predicted 4th Cousin Susanne

I clicked on the first match, Susanne, hoping to see a connection to the SNELL surname in some way, but unfortunately I didn’t see anything. I did, however, see this:

One of Susanne’s ancestors, Edna M. Whitney, was born in 1903 in New Haven. Yet another connection to the small town of New Haven, Oswego County, New York!

I wanted to dig into this family quickly to see if I could tie it in any way to the Snell family. A quick internet search uncovered the wonderful Whitney Research Group, including a page describing the family of Burton C. Whitney of New Haven, who was the father of Willis Whitney (who married Anna Belle Daniels).

To my shock, the wife of Burton C. Whitney was reported to be none other than Julia SNELL, daughter of “Edmond Cooley and Cynthia (Parmele) Snell.” Of course this would all have to be verified, but it was all quickly becoming a very tantalizing clue to my great-grandmother’s biological ancestry.

Predicted 4th Cousin John

Returning to John, the other predicted 4th cousin that my aunt shared in common with R.T., I reviewed his family tree as well and found the following:

Once again, Edmond Snell was in the family tree of this cousin! (As you’ll see, although this Mary Elizabeth Snell is the same Mary Elizabeth Snell found in R.T.’s tree, they’re descended through different daughters of Mary Elizabeth Snell, and the accounts appear to be administered by different people).

Confirming my Father’s Matches

I had been using my aunt’s results, so I went to my father’s account and checked to see if he had all the same results (and potentially some additional ones). He did indeed have the same results, the only difference being that he was actually a predicted 3rd cousin to R.T. where my aunt was a 4th cousin.

One More Clue…

Then I searched for SNELL in the match list of both my aunt and my father. Since AncestryDNA’s In Common With only shows predicted 4th cousins or closer, it was possible that there were other SNELL relatives from the same close family that didn’t share enough DNA to be 4th cousins or closer, but who still might be able to add to this research.

Perhaps not surprisingly, both my father and my aunt had matches with SNELL in their trees. One match, the very first “Distant Cousin” match in my aunt’s list of matches containing SNELL in their trees, whom we’ll call Jacob, was promising. Reviewing his tree, I saw the following:

Again a little beginning research, and I discovered that Sylvester Snell (reported to be born in Paris, Oneida County, New York in 1805) was reported to be the brother of Edmond Snell (possibly born in Paris, Oneida County, New York in 1802).

Could Helen Johnson be a Descendant of Edmond Cooley SNELL?

So now I had four different SNELL cousins, all of whom appeared to be distinct (none of the kits were administered by any of the others). And at least R.T., Susanne, and John were in common with each other, my father, and my aunt:

• R.T. – predicted 3rd cousin to Dad, 4th cousin to aunt
• Susanne – predicted 4th cousin to Dad & aunt
• John – predicted 4th cousin to Dad & aunt
• Jacob – predicted distant cousin to aunt

Of these, the descendants of Edmond Cooley Snell (R.T., Susanne, and John) were shared by both my father and my aunt, and were 3rd to 4th cousins (we’ll look at the amount of shared DNA in the next post). Meanwhile, the descendant of Sylvester Snell (Jacob) was only shared with my aunt and it was a much more distant relationship.

This pattern – together with the amounts of DNA shared by these relatives – make me think that it is very likely that my father and my aunt are descendants of Edmond Cooley Snell, and quite possibly through their grandmother Helen Johnson.

A Shocking Discovery

Unfortunately, concluding that Helen Johnson was a SNELL descendant would be an enormous leap here, for several reasons.

First, it is still possible that R.T., Susanne, John, and Jacob are related to my father and his sister via an entirely different family and surname that the paper trail hasn’t yet uncovered.

Second, there isn’t any real evidence that these individuals are related to my father and his sister via their paternal grandmother. There are 3 other grandparents through whom they could share this relationship.

But.

There is something, an undeniable fact that I haven’t yet revealed.

This undeniable fact makes the SNELL family extremely tempting as Helen’s biological family, yet at the same time calls the entire delicate hypothesis into question.

On April 9, 1932, Helen – a very recent widow at the age of 43 – married Walter A. Snell, grandson of Edmond Cooley Snell.

To be continued…

.

The post The Search For Helen’s Roots appeared first on The Genetic Genealogist.

We have multiple tools for collaboration of genetic matches. At GEDmatch and DNAGedcom, for example, we have many third-party tools that assist our efforts. The companies also offer tools that allow us to sift through our matches to find the clues we need. Family Tree DNA, for example, has an In Common With (“ICW”) tool and a Matrix tool that allow users to see what matches they share in common with another person.

Announcing Shared Matches

On August 26^th, AncestryDNA announced the “Shared Matches” tool, which is an ICW tool that shows you all the close matches you share in common with someone (usually, but not always, a genetic match) in the AncestryDNA database.

You can check for a Shared Match with any of your genetic matches, all the way to the last person in your list of matches. You can even check for Shared Matches with people who have no tree or only a private tree! However, because of the limitation discussed below, you will expect to find fewer and fewer Shared Matches as you work your way down your list.

Before we launch into this, be sure to see “New AncestryDNA Common Matches Tool: Love It!” by Diahan Southard at Lisa Louise Cooke’s Genealogy Gems for some absolutely beautiful illustrations!

Fourth Cousins or Closer

There is a very important limitation of Shared Matches that has created most of the confusion about the new tool: the tool will only show fourth cousins or closer. In other words, for a Shared Match to show up, that Shared Match must be a fourth cousin or closer to BOTH you and the person you are checking for shared matches with.

Let’s use an example. My name is Blaine, and I have a great-aunt named Amy who has tested at AncestryDNA. I see Amy in my match list, so I click on the VIEW MATCH button to see Amy’s profile. There, I click on SHARED MATCHES to see the shared matches we have in common. I get a list that includes Chris. In order for Chris to appear in that list, he must satisfy two criteria:

1. Chris MUST be a fourth cousin or closer to Blaine; AND
2. Chris MUST be a fourth cousin or closer to Amy.

If Chris is a distant cousin to Amy, he won’t appear in our Shared Matches list EVEN IF he is a match shared in common with Amy and Blaine.

In the follow graphic, which works off the same example, Blaine and Amy both have an array of matches. Some of those matches are shared in common, as shown in the region highlighted in red. Of those, only Chris and David are fourth cousins or closer to BOTH Blaine and Amy, so only Chris and David will show up in the Shared Matches list. While Blaine and Amy share Gil, Hu, and Jill in common, they are all more distant matches and won’t show up in the Shared Matches list. Further, close matches like Franny and Egbert won’t show up because they aren’t shared by both Blaine and Amy.

As you can see from the above, do not use the absence of a match from an ICW group as evidence!

Shared Matches Groups

Amy, Blaine, and Chris create a “Shared Matches Group.” Note that this is NOT a triangulation group, and is not even a pseudo-triangulation group. But it is a group nonetheless, and should be explored for clues.

As one commenter pointed on in an ISOGG Facebook group thread, once you can show an ICW group to a match, they might be more likely to share information, upload to GEDmatch, or otherwise collaborate.

Using a real-life example, I have 120 genetic relatives at the fourth cousin or closer range. My great-uncle has 187 genetic relatives at the fourth cousin or closer range. My great-uncle and I have 10 Shared Matches (3 immediate family members I’ve personally tested, 2 second cousins, a third cousin, and 4 fourth cousins).

Why Fourth Cousins or Closer?

AncestryDNA explains that “Shared matches work best for closer relationships so you’ll only see “high confidence” matches (4th cousins and closer to each of you) in the list.”

There are certainly plusses and minuses to limiting Shared Matches to fourth cousins (“4C”) or closer.

On the one hand, for those without significant endogamy, limiting Shared Matches to the 4C or closer range results in greater confidence that the Shared Matches Group is based on a common shared ancestor. Of course it’s not a guarantee, but it is more likely than a scenario where you’re adding in much more distant relatives. The more distant the relationship, the more opportunity for a relationship through multiple lines, and/or through different lines.

On the other hand, several people have been underwhelmed by the new tool because they have found very few ICW matches, usually due to having very few genetic matches at the 4C or closer range. Hopefully everyone’s results will continue to improve as the database continues to grow.

Use Caution

Now remember, ICW status among three (or more) people IS NEVER confirmation or proof that ALL three people share common ancestors, much less that they share the same segment of DNA. (Although I’m a proponent of Genetic Networks so sharing the same segment of DNA (“triangulation”) is not always a requirement; however, ICW alone is never sufficient without more).

Using the example above, Chris may be related to Blaine and Amy through the same line or through unrelated lines. Unfortunately, this problem can be exacerbated by endogamy.

Using Shared Matches with DNA Circles and New Ancestor Discoveries (“NAD”)

During the discussion of Shared Matches on the ISOGG Facebook group, one commenter noted that you could check your shared matches not only with your genetic matches, but also with anyone in a DNA Circle or a New Ancestry Discoveries group.

In the NAD below, for example, the test-taker is only genetically related to people in the orange circle groups. However, the test-taker can check for Shared Matches ANYONE in any of the circles:

Clicking on one of the NAD members with whom you don’t share DNA will look like this:

Not surprisingly, most of the time you won’t share a match ICW these people. However, once in a while you will, and that might help you solve the mystery of the NAD.

Other Uses

Several people have indicated that they use the AncestryDNA Shared Matches tool together with the Snavely ICW feature to expand the ICW analysis.

Do you have any other uses or successes with Shared Matches? If so, feel free to leave a comment below.

Nope, It’s Not a Chromosome Browser…

This is not a chromosome browser and you won’t get segment data. But that doesn’t mean that the Shared Matches tool is useless or that it can’t provide valuable clues. Indeed, many people have already posted about how the tool has helped them.

So beware the naysayers, and wring this tool out for everything it’s worth!

Other Resources

• “See Your DNA Matches in a Whole New Way” on the Ancestry blog
• “Introducing AncestryDNA New ‘Shared Matches’ Feature” an Ancestry video at YouTube
• “AncestryDNA Shared Matches” by Angie Bush at DearMYRTLE
• “Finally!” at The Legal Genealogist
• “AncestryDNA FINALLY introduces new “In Common With” feature, SHARED MATCHES” at Roots & Recombinant DNA
• “New AncestryDNA Common Matches Tool: Love It!” by Diahan Southard at Lisa Louise Cooke’s Genealogy Gems
• “Ancestry Shared Matches Combined With New Ancestor Discoveries” at DNAexplained
• “AncestryDNA Adds Shared Matches Tool” at genealogyinsider

The post AncestryDNA Announces New IN COMMON WITH Tool appeared first on The Genetic Genealogist.

Out of curiosity, I generated a similar table with my own data:

There are many interesting data points in the table. For instance, between the 7th and 8th generations, I drop from knowing 71% of all of my ancestors to knowing just 51% of my ancestors. At 10 generations, with 2046 total ancestors in all generations, I only know a quarter of them. And while I feel very confident for the first 6 or 7 generations; after that I’m much less confident with my family tree.

So this is an interesting exercise, but is a table like this at all applicable to genetic genealogy?

The Importance of Knowing Your Family Tree

Having a chart like this for your own family tree, or at least some knowledge of the information in such a chart, is a vital consideration in one’s determination of confidence in a conclusion made using DNA.

I recently stated the following in a thread in the ISOGG Facebook group:

“No atDNA paper or proof argument should EVER make a conclusion based on shared segments without at least a sentence or two about the lack of known overlap in other family lines. Or, alternatively, addressing known overlap. That is a fatal error.”

My point was that whenever we make a conclusion about a particular ancestor or ancestral couple based on segments of DNA shared with a relative, we absolutely must address whether we do, or could, share other ancestors with that relative.

For example, when I’m reviewing someone’s conclusion, I need to know whether they’ve at least considered the possibility that they could share DNA from another line instead of or in addition to the line on which they’ve focused. I need that information to evaluate their conclusions.

Now, maybe I’ll go one step further. Perhaps I’d also like to see how likely it is that they might share DNA via more than one recent line of their family tree; one vital factor in that probability determination is the extent to which they and their match have researched their family tree. And this is the topic I’d like to get other people’s thoughts on.

Of course there are many caveats, but ultimately they don’t really change my question. For example, we all know that family trees are notoriously error prone and can be subject to misattributed parentage. As a result, even the most complete family tree can be misleading. Or maybe your family lines are so diverse that you can be fairly certain there’s no unknown overlap affecting your conclusion. In either case, providing that information will help others evaluate your conclusions.

I know what some of you are thinking right now: “I’m not going to publish my conclusions anyway.” Or perhaps: “It doesn’t matter what others think of my conclusions.” Maybe you aren’t writing for a publication, and maybe you’re doing this for yourself, but ultimately your genealogical conclusions will be judged and evaluated. Whether it’s a relative that we decide to share it with, or a relative going through our files after we’re gone, all of our conclusions are ultimately evaluated.

An Example

Let’s use a quick example.

Joe, Julie, and John are all predicted to be about 5th cousins with each other (although I use 3 cousins, it could be 4 or 5, or even more). They all share a 22 cM segment of DNA on chromosome 3 which they’ve triangulated to a shared 4th great-grandparent, George and Susan (Gold) Silber. The triangulation of the segment of DNA is a definite plus, and it looks like a good conclusion.

But let’s factor in their family trees:

• Joe: an experienced genealogist, John knows 85% of his family tree out to the 6th generation
• Julie: an intermediate genealogist, Julie knows about 45% of her family tree out to the 6th generation
• John: a beginning genealogist, John knows about 10% of his family tree out to the 6th generation

How confident do you feel about their conclusion now? Julie is missing 55% of her family tree and John is missing 90% of his family tree. Doesn’t that increase the likelihood that their shared ancestor could be located within an unknown area of their family trees?

Is this information we should have when we evaluate a genealogical hypothesis or conclusion using DNA?

Questions

Does this example suggest that beginning genealogists shouldn’t make conclusions using DNA? Of course not! Does this example suggest that we need to ignore any conclusion unless you have 100% of your tree completed? Of course not!

Does this suggest that everyone should create and share a chart like the one I’ve provided when they want others to evaluate their genetic genealogy conclusion? Perhaps, yes. I can say that it only took about 30 minutes for me to create these graphs, so the burden is extremely low (especially when compared to the hundreds of hours spent on DNA!) and the return on investment is extremely high.

So what are your thoughts on the matter?

.

The post How Much of Your Family Tree Do You Know? And Why Does That Matter? appeared first on The Genetic Genealogist.

The ISOGG wiki page about TribeCode offers some information about the test, gleaned mostly from Facebook postings by the company. For example, the test apparently uses an Illumina low-coverage sequencing technology and tests at least 12 million markers throughout the genome. More exact details of the sequencing aren’t yet found on the TribeCode website.

Around Thanksgiving of 2014 I ordered the test on sale from approximately $79, and received my results a couple of months later.

My Ancestry

After receiving notification that your results are ready, you’ll log in and see the introductory page:

The introductory page provides links to all the various tools, including:

1. Ethnicity Composition – Overview
2. Maternal lineage (mtDNA)
3. Paternal lineage (Y-DNA)
4. Ethnicity Composition – European View
5. Chromosome Painting (Ethnicity Composition – Chromosome View)
6. mtDNA Tree
7. Y-DNA Tree
8. Jewish Ancestry
9. Principal Component Analysis – PCA

I’ll look at most of these tools below, analyzing them in view of tools I’m familiar with at other testing companies.

Maternal Lineage

According to TribeCode’s website, the test “sequences your full mitochondrial DNA, giving you the most comprehensive and accurate maternal lineage information possible.” My mtDNA results show as A2w1, which is the furthest branch of the mtDNA tree that I can currently be mapped to (other than still-new research showing that I am A2w1b, see “An mtDNA Journey – Discovering My mtDNA in a Research Paper“):

For comparison, Family Tree DNA assigns me to the Haplogroup A2w, and 23andMe assigns me to A2.  Family Tree DNA’s full mtDNA sequence could classify me as A2w1 and A2w1b, but their tree will have to be updated with the most recent information. 23andMe does not sequence the full mtDNA genome, and a quick look at the SNPs they test indicates that they are probably unable to classify me as A2w1.

TribeCode provides an mtDNA tree interface, which is based on PhyloTree (the most recent version of which is from February 2014):

My branch is very accurate, minus the very recent discovery of new branches under A2w, including my branch of A2w1b.

In the tree, it appears that green means positive for the SNP and red means negative for the SNP. For example, I know that my mtDNA line likely back-mutated C16111T more than 1,500 years ago, and I am negative for that SNP.

Unfortunately, TribeCode does not provide the mtDNA sequence, other than what can be gleaned from the tree.

Paternal Lineage

According to TribeCode’s website, an algorithm called YFitter is used to determine the test-taker’s paternal lineage. YFitter is a statistical tool that can be used, for example, to determine Y-DNA haplogroup using low-coverage Y-DNA sequencing data.

My Y-DNA results, which TribeCode has based on the ISOGG Y-DNA tree, show as R1b1a2a1ac1a2a:

Looking at the ISOGG Y-DNA tree, I see that R1b1a2a1a1c1a2b is R-L1, which I know is my terminal SNP:

TribeCode provides a Y-DNA tree interface with the same green and red indication:

For comparison, Family Tree DNA assigns me to Haplogroup R-L1 (due to SNP testing imported from The Genographic Project), and 23andMe assigns me to R1b1b2a1a1* (an old way to say somewhere below S21/U106).

Unfortunately, TribeCode does not provide the Y-DNA sequence, other than what can be gleaned from the tree.

Ethnicity Composition

The Ethnicity Composition tool compares your DNA to 62 reference populations (I couldn’t find the list), and is intended to reflect where your ancestors lived approximately 500 years ago.

There are two confidence levels, “Confident” and “Standard.” Confident mode shows all ethnicities equaling 3% or more of your overall composition, while Standard mode shows ethnicities equaling 1% or more of your total composition.

At the “Confident” confidence level, my ethnicity is 100% European:

At the “Standard” confidence level, my report adds African, Native American, and East Asian ancestry:

This is very similar to my 23andMe ethnicity report, which is 95.5% European, 3.1% Native American and Asian, and about 1% African (at the “Speculative” confidence level). As we’ll see below, these percentages are not surprising in light of the Chromosome View.

From time to time, TribeCode will introduce “Experimental” tools. For example, a few months ago, the TribeCode ethnicity algorithm was not picking up Native American ancestry, and they had an Experimental tool to detect Native American ancestry. Since that time, it appears that the tool has been incorporated into their standard ethnicity algorithm.

The Experimental “European View” tool shows only European ancestry estimates, and groups all non-European estimates into an “Others” category. In the Standard mode, I have 4.63% Others. It’s interesting to note how different these European estimates are from the European estimates provided above, which is likely a combination of both different labels, and different reference populations.

TribeCode also offers a Chromosome View which, like 23andMe and GEDmatch, shows chromosomes 1-22 with your personal calls for the four broadest ethnicity categories painted onto them:

This is a comparison of my first six chromosomes at TribeCode and at 23andMe:

It is possible to look at one chromosome at a time. Here I’m zooming in on Chromosome 2:

And then you can zoom into a portion of a chromosome. For example, here I’m zooming in on the right end of Chromosome 2:

When looking at a single chromosome, you can see the locations of genes along the length of the chromosome (shown by the multi-colored tic marks along the top of the view). Hovering over a tic mark will bring up the name of the gene:

And clicking on the tic mark will bring up a box with information about the gene:

I’m a very big fan of the Chromosome View at 23andMe and GEDmatch, and I very much appreciate having that view at TribeCode. With a few rare exceptions (mostly for adoptees), the overall ethnicity percentages are not very helpful for genealogical research. In contrast, knowing actual segments of DNA that have been assigned to a particular ethnicity could be very helpful to your research.

For example, because I know what grandparent my African and Native American segments come from, I can quickly assign those segments to her. And because I’ve tested my mother, who has additional segments that I don’t have, by process of elimination I can tentatively map those segments I didn’t inherit from Mom to her father, my grandfather. So the Chromosome View at 23andMe helped me map a significant percentage of my genome.

But how do I get the start and stop positions for these segments?

If you’ve tested at 23andMe, you can download a spreadsheet of those segments using DNAGedcom:

You might also be able to get this information from TribeCode’s browser, at least a rough approximation. For example, by clicking on the tic marks denoted by the red arrows below, I can get a gene pop-up with a location.

The tic marks reveal the following approximate start and stop locations for these two segments:

• African – 24,397,972 to 42,669,157
• Native American – 211,341,429 to 241,570,676

From 23andMe via DNAGedcom:

• African – 27,184,446 to 31,000,643
• Native American – 217,482,379 to 240,900,227

Not very exact, but close. The first challenge, and it is a very big one, is what segments to examine and where to place the start and stop positions. For example, without the benefit of the 23andMe comparison (or perhaps a calculator from GEDmatch using AncestryDNA or FTDNA data), it would be difficult to conclude what segments I thought were significant enough to assign.

Jewish Ancestry

TribeCode also has an experimental tool called “Jewish Ancestry.” According to the website, it can “accurately estimate if you have one, two, three, or four Ashkenazi Jewish grandparents.” From the website:

If you are of European descent, an analysis of your DNA sequencing results may indicate whether you have Ashkenazi Jewish ancestry. Your results are compared to SNP alleles in hundreds of self-described Jewish and non-Jewish individuals from this study, as well as another data set from a large cohort in New York.

I have no known Jewish ancestry:

Conclusions

I enjoyed testing with TribeCode and reviewing my results. I think the Y-DNA and mtDNA designations, at least for my results, were extremely accurate. For my Y-DNA, for example, it even determined my terminal SNP.

The ethnicity estimate from TribeCode also appears to be accurate, at least when looking at the four broadest categories. I wouldn’t rely in any way on the sub-regional categories, including the “European View,” but that’s a symptom of the science, not TribeCode. You shouldn’t be relying on sub-regional categories at any testing company.

My concern with TribeCode is that they don’t yet return raw data to customers. According to their Facebook page they’re still considering it, but are concerned about the size of the files which are much larger than typical genetic genealogy files. I’m not positive, but my suspicion is that TribeCode might be concerned about returning low-coverage sequencing to customers (if indeed it is low-coverage sequencing), as it will knowingly contain many sequencing errors due to the low-coverage nature. That doesn’t have a significant impact on the various tools, but it would be difficult to adequately explain to consumers who would find and then only focus on the errors. However, according to Genetic Genealogy Standard #3, genealogists believe that they have an inalienable right to their raw data. Hopefully TribeCode will review their current policy and provide raw data to consumers.

.

The post A Review of TribeCode by Centrillion Biosciences appeared first on The Genetic Genealogist.