Omics! Omics!

Notes On Instrument APIs Critiques I Delegated to Claude

2026-09-03T10:41:22.883-04:00

In my previous post about the Model Hardware Standard from Anthropic, I mentioned some exploration I've been running on instrument APIs. Given our Catalyst Agent skills, a model such as Claude can evaluate what gaps exist between a given API and what we would require to integrate into our RAC autonomous laboratory environment. I've run only a few APIs through, and I'm not going to reference which ones, but I have seen some patterns. The whole process has been very thought-provoking.Read more »

Model Hardware Standard: What Is It? Seriously, That's What I'm Asking.

2026-08-31T23:26:27.548-04:00

Anthropic, the company behind Claude, announced a new initiative called the Model Hardware Standard (MHS). It's a way to enable intelligent agents to talk to laboratory instrumentation - well, that's what it's billed as. But as first surfaced in a tweet by Sina Booeshaghi, if they want to be a standard they're taking a counterproductive path to it, as the announcement was only that. A blogpost extolling the goals of MHS is given accompanied by testimonials from two early access sites , but no open and substantive description of MHS itself that would allow evaluating what it can and cannot do. Read more »

Recombination-Based Autonomous Protein Structure-Function Search

2026-08-27T09:55:34.112-04:00

An interesting preprint showed up last week in the autonomous agent-driven experimentation space - an area I'm paid to be very interested in. This didn't run on our hardware, but we certainly could reproduce the experiment on our autonomous laboratory installation Nebula if the preprint had all the critical details (I'll try not to kvetch constantly about that, but it is an annoying brake on my eagerness to reproduce some or all the results). Anyone could run it on our Cloud Lab - I haven't priced it out but this should come in at a very reasonable number. What I found particularly clever is the solution to the problem of keeping protein engineering by AI in a reasonable cost envelope - and this is accomplished by relying on chimeras which can be a generated from one round of DNA synthesis and then clever shuffling.

Learning protein function through autonomous experimental interaction

, ,

https://www.biorxiv.org/content/10.64898/2026.08.14.744985v1
Read more »

SIMB Notebook: The Frontiers of Protein Engineering

2026-08-20T21:51:41.067-04:00

The Society for Industrial Microbiology (SIMB) annual meeting in Austin the first week of the month was great fun. I caught up with two Warpers, saw some other contacts from that era and met a bunch of new friends. Plus I showed the flag for autonomous laboratories with a talk in a session on applying AI to the field - and got a chance to talk to multiple users of our systems at customer sites. Murphy's law struck in that our PR on four academic placements of autonomous labs came out about 20 hours after my talk. In addition to the session I appeared in, I attended a keynote on protein engineering, a session on enzyme engineering, and a whole day of talks on natural product research - plus all the posters. I have a few notes that are maddeningly incomplete & I apologize in advance for the quality of attributions. But here are a few observations on the space on protein engineering, which I am trying to level up to "barely dilettante" grade.Read more »

HiFi's Not Very Pacific Waters

2026-08-18T23:40:33.114-04:00

PacBio had a major shakeup early this month. First came a large round of layoffs, which caught several individuals who were very much the face of innovation on social media. Next came the sudden retirement of CEO Christian Henry, who has appeared in this space multiple times, and the elevation of CFO Mark Van Oene to the top seat.

When PCR Gives You Lemons...

2026-08-03T00:07:05.434-04:00

I wish to highlight an interesting new preprint from a group at Danish Technical University which explores the schmutz in PCR reactions, demonstrates some surprising conclusions, shows how their results might be very relevant to a common use of amplicon sequencing, and demonstrate a specific utility running PCR on in cases where typical PCR absolutely can't succeed. Put another way, how often do people patent what was previously considered uninteresting experimental artifacts?

Dad Had Manuals, I Have An Agent

2026-07-28T14:29:13.933-04:00

My father was a very methodical person, and that particularly applied when he decided to purchase the second IBM PC clone to appear on the market. A whole host of manuals appeared on his desk and when I started skimming them it was clear he had been there first. Sentences highlighted, marginal notes and of course editing marks to suit his sense of grammar and style. The hardware, operating system, and APL programming language by far represented the most complex technology in the house - easily more complicated than everything else put together.Read more »

Element's Curious Patentsieg

2026-07-21T22:02:50.860-04:00

Last week news broke that Element Biosciences had won a patent lawsuit against Illumina, and that it affected MiSeq. MiSeq???? How did a patent battle in 2026 affect a platform launched nearly a decade and a half ago? And which had been launched about five years before Element incorporated. Which is being put out to pasture, though the MiSeqDx version What witchcraft is this? I've done limited digging - I really dislike trying to dig through patents - and it is a strange case.Read more »

Fiber-Seq: Plasmids Model Chromosomes Well

2026-07-20T10:54:22.618-04:00

A recent publication in Molecular Cell involving Fiber-Seq caught my attention, though I've since discovered this preprinted over a year ago. One of the challenges in exploring the effects of genetic variations is the cost & effort to generate collections of variation which can be profiled by available methods. This paper explores using Fiber-Seq on transfected plasmids and finds that, with important caveats, transfected plasmids evoke the same patterns of chromatin protection as the corresponding nuclear DNA. This opens up the possibility of exploring large sets of variants for their effect on chromatin structure and either feeding that data into models or using it to understand clinical non-coding variation. Or why not both?Read more »

The Simulator That Isn't There

2026-07-10T09:14:08.954-04:00

Okay, this is one of those cases where I kvetch about other people not having built the tools I want to use - perhaps a bit of a insufferably egotistical position. I mean, what gives me the chutzpah to do that? My title, according to me, is "Automation Visioneer", so that gives me the right - though again, self-given title, self-given right. But I hope you'll bear with me, because I think (there again, me! me! me!) it is an interesting observation on a curious gap in the software world - one that could make a very interesting project for someone so inclined. Oh, and this is one of the "a bit of promoting the product" entries - which means I get to blog on company time.Read more »

Countable Labs Raises Funds, Heads to Clinic

2026-07-07T11:44:18.444-04:00

I caught up by teleconference recently with Countable Labs CEO Giovanna Prout, fresh off Countable Labs' $26M equity fundraising - a round which was oversubscribed. Top on my agenda: what is Prout's vision to turn that money into products. And how will Countable balance building molecular tools vs. applying those tools to clinical problems?

Alex Hastie Returns to BioNano, Takes CSO Reins

2026-06-28T08:54:17.899-04:00

An interesting news item this week is that BioNano Genomics has appointed Alex Hastie as Chief Scientific Officer. Hastie is an obvious choice for CSO, as for many years he was very much the public scientific face of BioNano's development. . But in February 2025 he left BioNano and headed to Arima Genomics. But after a bit over a year at Arima and then a short stint at Inocras, Hastie has returned to the fold.Read more »

E.coli Doesn't Like Diets Either, At Least Genomic Ones

2026-06-24T23:19:01.344-04:00

Another interesting preprint that fits several recent themes - genome minimization, discontinuation of products - comes from a Japanese group that reports on significantly reducing the genome of E.coli. They tried for 1.7Mb but ended up settling - for now - with a 2.3Mb genome. That's only about half the size of the strain they started with, but not at all a small genome. Read more »

Shredding Genomes: Establishing Plausible Deletability

2026-06-23T10:43:39.133-04:00

There's an interesting new preprint from Jay Shendure's lab which introduces the technique of Shred-Seq, which is a Cas3-driven approach to generating large deletions across a large eukaryotic genome. By creating diverse collections of deletions, Shred-Seq enables identifying hidden non-coding sequences of functional importance.Read more »

The Commercial End for Fluidigm Technology?

2026-06-18T10:28:10.744-04:00

Last week brought news that Standard Biotools, which is what Fluidigm morphed into, is being acquired by Treeline Biosciences, a drug developer. The deal values Standard at about its cash position - and Treeline made it immediately clear this is a reverse merger to grab the cash and in effect go public by also grabbing Standard's stock listing. What Treeline has made very clear is that they have absolutely no interest in is Standard's existing tools portfolio.

Long Reads for Rare Diseases Hits New England Journal of Medicine

2026-06-13T17:03:14.345-04:00

Alexander Hoischen and colleagues have a brief piece in New England Journal of Medicine that published early this morning, in conjunction with the European Society of Human Genomics (ESHG) meeting. I'm now in the thrall of ESHG FOMO - I attended last year in part because it was immediately after London Calling and it seemed silly not to extend my trip (plus Alex had asked me personally to go), but this year I reluctantly decided my travel load was getting high and decided to make only one transatlantic flight in the May-June timeframe. Hopefully I picked the right one. The piece is the latest installment of a long running program to demonstrate that long read sequencing, specifically PacBio HiFi sequencing, can replace a battery of older tests when trying to diagnose rare genetic diseases.

Craig Venter Reflections: Building & Redesigning Genomes

2026-06-12T14:16:15.359-04:00

The last bit of Craig Venter's career a followed, often a bit green with envy, was his work to build genomes. This all revolved around the Mycoplasma genitalium they had knocked out a genome as a victory lap after completing the first bacterial genome. This bug is held up as the bacterium with the smallest genome which can be grown in a defined medium. To first reconstruct an entire genome, with watermarking sequences to prove it was done, required pushing every limit of the existing gene synthesis cloning technology. An effort so intense, that collaborator Dan Gibson became a verb! (See that gene cluster we want? We'll Gibson it into a BAC vector.")Read more »

Craig Venter Reflections: Human Genome

2026-06-11T13:09:44.273-04:00

Few bombshells have hit genomics like the day in May 1998 when Craig Venter and Applied Biosystems announced he would be launching Celera Genomics to sequence the human genome by a complete shotgun approach. A couple of colleagues at Millennium had just announced they were leaving to form a contract research organization which would perform high throughput sequencing on demand - they were certainly rocked back on their heels (though not blown over; Orion Genomics remains successful to this day with a focus on palm oil genomes). The public project was also rattled - Celera was promising to deliver a private genome much sooner. In some ways, it was the EST brouhaha all over again - except with important twists.Read more »

Craig Venter Reflections: Small Genomes

2026-06-10T09:46:41.542-04:00

In the early 1990s, Craig Venter left the NIH over disputes around the patenting of ESTs. Investors backed him, but then William Haseltine basically pushed him out of direct operation of Human Genome Sciences (HGS) and Venter went to start The Institute for Genome Research, far better known as TIGR. TIGR would attract an amazing array of talent both wet lab and dry lab - a pattern that Venter had established. I believe it was around then that Venter attracted both Clyde Hutchinson and Hamilton Smith to TIGR - two giants of microbiology (Smith had a Nobel Prize!), and they would collaborate with him for the rest of their lives. HGS would have human biology to themselves, but TIGR was free to explore other parts of life domain. And it would be in the sequencing of microbes that Venter and TIGR would radically shake up a world that, as I reviewed in the previous piece, was debating how many levels of physical maps to build.Read more »

Craig Venter's Genomics Revolution: Setting the Stage

2026-06-09T09:45:30.592-04:00

I am overdue in getting back to my very personal review of how the recently passed J. Craig Venter made a huge impact on genomics science. But especially after being at London Calling, I worry that many in genomics today just won't have the context to grasp how unusual his approaches were. I met some amazing graduate students there - and most likely they were born after the joint announcement by Venter and Francis Collins of the human genome initial completion. So with the risk of writing a "back in my day we walked uphill to school both ways" piece, I wish to sketch a debate that the genomics world was having in the early days of the Human Genome Project.

FoG to Roll into Boston Seaport Wednesday & Thursday

2026-06-01T23:22:25.545-04:00

The Festival of Genomics, Biodata and AI - thankfully known as FoG and not FogBA or FogBDA - will be at the Boston convention center this Wednesday and Thursday. It's an interesting conference that typically has two editions each year, one in London and one here in Boston. I'll be wandering around the floor part of both days - my beacon/hat to be announced on LinkedIn and Twitter that day (leaning hard towards my yellow University of Delaware hat). FoG works hard to not be your average scientific conference, and I'd like to run down a few things I'm anticipating. By the way, the world's largest autonomous lab is less than a mile from the conference site and tours are available during the conference - all you need to do is reach out to the tour coordinator to schedule one. And I'm that, so DM me on LinkedIn and I can set you up.

London Calling: Long-Term R&D

2026-05-31T23:07:55.266-04:00

Now to run through the long term technical projections from Oxford Nanopore's Tech Talk. Well, projection is probably not the correct word - hints and teasers might be closer to the mark. None of these pronouncements came with any hint of timing for when the community might be able to access them - indeed, CSO Lakmal Jayasinghe made it very clear that this section would have no timing information. Here is a link to the presentation: what is covered in this piece starts around the 44:40 mark.

London Calling: ONT's Near Term Launches

2026-05-30T22:39:20.412-04:00

I returned from London Calling 2026 much as I usually have - exhausted, more than a little hoarse, happy for discussions had and ruing connections missed. The meeting felt much like prior ones, except in all the ways it was a little different. The company continues its trajectory towards restraint and corporate maturity, with new CEO Francis Van Parys for the first time exhibiting his "evolution not revolution" messaging for the company's direction. ONT's technical presentation was divided into two parts: near term that came with release dates = none of them even as distant as the end of the summer - and long term which had no hint of timing. This piece will focus on the near term advancements and a separate one will cover the long term updates. The entire presentation is already available on YouTube.

London Calling 2026 Preview

2026-05-19T19:02:13.808-04:00

My series on Craig Venter is on a small pause since I didn't get it out before I arrived here in London for Oxford Nanopore's annual shindig at Old Billingsgate. On the plus side (well, I hope my readers find it a plus), I've sketched out an additional piece - I realized that our world of fast whole genome sequencing powered by long read technologies such as Oxford Nanopore may make it challenging for many to remember the genomics world of the 1990s and how there were great differences in opinion on how to go about sequencing genomes. But back to London Calling - this is an interesting moment in the history of the company and it shows in the program. So without more ado, here are some key topics and questions on my mind.

Personal Reflections on Craig Venter: Expressed Sequence Tags

2026-05-12T14:47:00.000-04:00

Venter's big Expressed Sequence Tag - soon known as ESTs - paper came out in June during my final summer as an intern; I photocopied it at my internship. I looked it up, and it had only a bit over 600 sequences - a small beginning to what would become an industry-wide exercise. That was a momentous time for me, as only the past December had I pivoted from thinking I would get a degree in experimental plant molecular biology to focusing on computational genomics. Venter was not the first person to feed an RNA library into a DNA sequencing workflow without prior screening; I believe Gregor Sutcliffe holds that milestone. But Venter did it on a much larger scale and with much more flash. He pitched it as a way to skim the cream of the human genome when the official Human Genome Project was just getting going. And as would characterize much of his career, it is the response of others in the community that was as much his contribution as his own work.