β.Log [ Computational Biology News ]

de-Bruijn assembler

2012-01-27T01:40:00.000+05:30

In the last post I talked about the overlap-layout-consensus (OLC) way of Genome assembly. The approach which is (really!) getting popular these days is the other one, de-Bruijn-graphs (DBG). It is based on the simple idea of converting the hard problem of finding Hamiltonian to relatively simpler Eulerian for assembling biological sequences. There is a beautiful tutorial like introduction co-authored by the 'father' of this idea, Prof. Pevzner.

To put it naively, in contrast with OLC approach where one represents the reads as vertex in graph which are connected if the end of the read representing the first vertex overlaps with the start of the second one, DBG approach has the overlapping parts of the reads as the vertices and edges representing the reads consisting of the those overlaps. Now all we have to do, well kind of, is to traverse this graph so that we visit edges once, which is what Eulerian path is, and it scales up much better than Hamiltonian making it a cool choice for the cqncers churning out massive amount of shorter reads .

I made a simple scripts to see this approach in action. It can be downloaded via:

wget http://misccb.googlecode.com/git-history/585ec7811a837a1f20ee37256abada7d6ee538a3/eulerovlpath.pl

To test is, one needs a simple genome (here 100 bases long), which can be done via PERL one-liner:

perl -e '@b=qw/A T G C/;while($l<100){print @b[int(rand(4))];$l++;}' > rangen.txt

generating 100 base pair uniformly random DNA string. The sequencing can be simulated via breaking up the above string into small (here 10 bases long) overlapping (here of 9 bases long) strings using PERL one-liner:

perl -ne 'for($c=0;$c < length;$c++){print substr($_,$c,10);print "\n"};' rangen.txt > rangen.txt.kmer

This reads file can be directly fed into the downloaded eulerovlpath.pl:

perl eulerovlpath.pl rangen.txt.kmer

generating the string:

GGGCGTGGGCATTCCAACGACTGAGTCTGCCCTAATTCGTTGAAGTACTATTCGCAACGACCGCCGTAGACAAATAAGAGTGTAGGGATTACCGGCATT

which is similar to what we started with :)

cat rangen.txt
AGGGCGTGGGCATTCCAACGACTGAGTCTGCCCTAATTCGTTGAAGTACTATTCGCAACGACCGCCGTAGACAAATAAGAGTGTAGGGATTACCGGCATT

Happy hacking !

Hamiltonian Assembler

2012-01-05T20:33:00.001+05:30

Just came across Andrew's Hamiltonian Cycle finder (HCF)

get:

 wget https://raw.github.com/ahh/ahh-toys/master/ham.sh

simple check:

 bash ham.sh 
 a b
 b c
 c a 
 [followed by Ctrl-D should produce] 

a b c

based purely on shell commands and thought of testing it out as a Genome assembler. The genome assembly problem is closely related to finding the shortest common superstring (S) of a given set of strings (s1, s2… sn). The superstring S corresponds to the genome and the set of strings being the short sequence reads produced by the sequencing machines.

The popular ways of approaching this problem are OLC and DBG (overlap-layout-consensus and de-bruijn-graph). OLC approach is similar to HCF or the popular textbook example, the traveling-salesman-problem :).

Typical bacterial genomes are strings of nucleotides (A,T,G,C) and circular. General approach to find it is to extract it from the organism, make multiple copies, shatter it into small manageable pieces, feed into a sequencer and generate list of overlapping reads (since we made multiple copies, it is highly likely that reads coming out will be overlapping and the sum-read-length/genome-size represents coverage). In OLC approach, one represents the reads as vertex in graph which are connected if the end of the read representing the first vertex overlaps with the start of the second one.

This graph is traversed to find a Hamiltonian Path and here is where one can use Adam's HCF :) Thus wrote a simple PERL script:

wget http://misccb.googlecode.com/git-history/9deb0b79350f92aa41735091dcc69688a9a45bf0/seqmerge.pl

and ran it:

perl seqmerge.pl 9 3 4 2

to create random small genome of a given length (9), break it into overlapping k-mers of given size (3) reflecting a given coverage (4), use the overlap (2) information to create list of connected k-mers. This is fed to the HCF and we compare the Simulated Genome to the HCF generated Genome to see if it works. For example:

perl seqmerge.pl 9 3 4 2
Simulated Genome CATTTTGAT
Generated k-mers CAT ATT TTT TTT TTG TGA GAT ATC TCA CAT ATT TTT
Genome file Genome.L9.K3.C4.O2.fa
Connectivity file Genome.L9.K3.C4.O2.cg
Running shell Hamiltonian path finder and writing output to Genome.L9.K3.C4.O2.cg.ham
Collapsed ham.sh output from Genome.L9.K3.C4.O2.cg.ham written to Genome.L9.K3.C4.O2.fa.ham
Assembled Genome TTTGATCAT

and as we can see the circular genome CATTTTGAT is same as TTTGATCAT, so it did work :)

Getting out gene list from Ensembl web query results with WWW::Mechanize

2011-07-20T04:51:00.000+05:30

Seems like the Ensembl API is not providing a way to get list of results (?) for web query, thus decided to write a simple perl script. So to get list of gene hits for Ensembl query 'histone h2a histone cluster 1' one can simply run:

perl getensidmech.pl 'histone h2a histone cluster 1'

and the quick and dirty code getensidmech.pl using WWW::Mechanize is there at http://research-code-base-animesh.googlecode.com/svn/trunk/code_lic/getensidmech.pl :)

Cancer classification using Google prediction API

2011-05-29T18:12:00.000+05:30

Google's prediction API is now available to everyone with nice tutorial on how to get going :) so I decided to check its performance using gene expression values for 5 selected genes using online-feature-selection from the classic AML/ALL dataset (training/test).
The dataset needed to be formatted (training/test) and copied to Google storage to get the default scripts working.
Coming to the results, well the training step:

./oauth-train.sh mlani/top5_tr.csv

gave a performance of about 3/4:
./oauth-check-training.sh mlani/top5_tr.csv
{
"kind": "prediction#training",
"id": "mlani/top5_tr.csv",
"selfLink": "https://www.googleapis.com/prediction/v1.2/training/mlani/top5_tr.csv",
"modelInfo": {
"modelType": "classification",
"classificationAccuracy": 0.75
},
"trainingStatus": "DONE"
}not too bad, considering they are using some sort of statistical learning and sample is not the recommended 10x of feature set...
Coming to testing part:
./oauth-predict.sh mlani/top5_tr.csv "1122,178,847,33,1018"
{
"kind": "prediction#output",
"id": "mlani/top5_tr.csv",
"selfLink": "https://www.googleapis.com/prediction/v1.2/training/mlani/top5_tr.csv/predict",
"outputLabel": "ALL",
"outputMulti": [
{
   "label": "ALL",
   "score": 0.899588
},
{
   "label": "AML",
   "score": 0.100412
}
]
}
so seems like the first example went well, putting ALL as ALL with ~90% score :)
To get it over all the test examples, needed to write a simple PERL script:
perl check-pred-googleapi.pl top5_te.csv
ALLALLALLALLALLALLALLALLALLALLALLALLALLALLAML1
ALLAML2
ALLALLALLAMLAMLAMLAMLAMLALL3
AMLALL4
AMLAMLALL5
AMLAMLAMLGot 29 out of 34(accuracy:0.852941176470588)
which I guess is pretty good considering black box like usage :)
Happy predicting !

PERL one liner to generate 1000 base genome

2011-05-22T00:47:00.001+05:30

perl -e '@b=qw/A T G C/;print ">Genome\n";while($l<1000){print @b[int(rand(4))];$l++;}'

Finding coding region with Fourier transform

2011-03-29T19:30:00.001+05:30

Been a long time since I posted anything here, well I hardly had anything worth posting ...
Recently a friend of mine asked for implementation of Genescan code presented in "Prediction of probable genes by Fourier analysis of genomic sequences" / "Ab initio gene identification: prokaryote genome annotation with GeneScan and GLIMMER" and thus I went back looking into my old posts here... and strangely there were none about it! So decided to post for future reference.
Running the code waexome.pl is straight forward, for example running it over ins.fas:

perl waexome.pl ins.fas 3 300

will split sequence(s) into window of 300 and look for periodicity at 3, generate one .fted.txt for each sequence in the ins.fas and also ins.fted.txt which will report regions > p2nr threshold (hardcoded to 4), naively these are the coding regions according to the Fourier transformation threshold as they show peak periodicity at 3 (size of codon). Thus the generated file ins.fas.fted.txt reports the coding region for this particular example :).

Harnessing the power of multicore

2008-07-29T00:35:00.003+05:30

Most of us are working on multi - core processors , but the code we generally write do not utilize the extra cpu given to us. I do lot of "Align EST and genomic DNA sequences" using est2genome from EMBOSS. This is a task which is embarrassingly amenable to parallelization. So how do we go about doing this using PERL? Well following a posting at PerlMonks, I got hold of Parallel::ForkManager:
wget http://search.cpan.org/CPAN/authors/id/D/DL/DLUX/Parallel-ForkManager-0.7.5.tar.gz
and extracted to:
/scratch/misc/parallel/Parallel
[if you have root access to your machine, I would recommend to install this module instead of going for 'use lib'].
Once this is done, all you need is follow the examples and get going :)
For example, I have EST in fasta format files ending with .fas. The genome I am aligning to is NC_010336.fna. For this I can write the following code:


  use strict;
  use warnings;
  use lib '/scratch/misc/parallel/Parallel';
  use LWP::Simple;
  use Parallel::ForkManager;
  system("ls -1 *.fas > list.tmp");
  my $command="est2genome";
  my $genome="NC_010336.fna";
  my @tasks;
  open(F,"list.tmp");
  while(< F >){chomp;push(@tasks,$_);}
  close F;  
  my $tasksize= @tasks;
  print "There are #  $tasksize \n";
  my $pm = new Parallel::ForkManager($tasksize); 
  foreach my $task (@tasks) {
    $pm->start and next; # do the fork
    system("$command $task $genome $task.$genome.out");
    $pm->finish; 
  }
  $pm->wait_all_children;

and store is as est2genome_parallel.pl and run it as:
perl est2genome_parallel.pl
above code can be downloaded from http://sharma.animesh.googlepages.com/est2genome_parallel.pl, enjoy :)

Blast /usr/share/dict/linux.words

2008-07-28T23:47:00.006+05:30

After reading 'Waq’s Words and World', where the author tries to find anagrams by scanning through dictionary, I got interested in checking which is the biggest word in /usr/share/dict/linux.words which can give a significant hit with blasp against swissprot. So I vimmed dict.pl and wrote:


#!/usr/bin/perl
while(<>){chomp;split(/\s+/);$c++;$dictast{length(@_[0])}.=">s.$c\n@_[0]\n";}
foreach $w (sort {$b<=>$a} keys %dictast){
   open(FI,">temp.blast.in");
   print "Blasting $w length word(s) file\n";
   print FI"$dictast{$w}";
   close FI;
   system("blastcl3 -p blastp -d swissprot -i temp.blast.in -o temp.blast.out");
   open(FO,"temp.blast.out");
   while(<>){
           print $_;
           if($_=~/^Sequences producing significant/){
                   close FO;  
                   die;
           }
   }
}

The program died aligining SPECTROCOLORIMETRY with SPDCAERCGIMRLMDTRY from ">sp|Q0U3Y6|DDI1_PHANO DNA damage-inducible protein 1":


Query: 1          SPECTROCOLORIMETRY 18
                  SP+C   C + R+M+TRY
Sbjct: 229        SPDCAERCGIMRLMDTRY 246

To replicate or explore-further, the code can be downleaded from http://sharma.animesh.googlepages.com/dict.pl ad run as:
perl dict.pl /usr/share/dict/linux.words
[ /usr/share/dict/linux.words is the dictionary file coming with fedora 9, you may have it in different location, to find, do 'find / -name dict' ].

clipped from idazuwaika.wordpress.com

Anyway, to avoid getting rusty with programming (as I have with football), I wrote a Java program to scan through list of English words and find anagram pairs. Below are the interesting ones.

antagonist stagnation

ascertain sectarian

bacterial calibrate

coordinate decoration

courteous outsource

eroticism isometric

excitation intoxicate

prettiness persistent

satirical racialist

shattering straighten

supersonic percussion

Kyoto Prize goes to Karp

2008-07-27T01:40:00.001+05:30

Prof. Richard M. Karp [ http://www.cs.berkeley.edu/~karp/biography.html ] won the prestigious Kyoto Prize this year in Information Science category [ http://en.wikipedia.org/wiki/List_of_Kyoto_Prize_winners ].Good to see a Bioinformatics hero getting something like Nobel.

clipped from www.bizjournals.com

UC-Berkeley professor Richard Karp has won the 2008 Kyoto Prize, Japan's equivalent of the Nobel Prize, in recognition of his lifetime achievements in the field of computer theory.

Karp is one of three laureates named Friday, entitling him to a cash gift of about $460,000, a diploma and a gold medal recognizing his work in defining the field of theoretical computer science.

Switiching via common structural framework

2008-07-27T01:17:00.001+05:30

Josh explores [ http://www.thinkgene.com/researchers-reveal-insights-into-hidden-world-of-protein-folding/ ] an article from Booth et al. titled 'Mechanism of lid closure in the eukaryotic chaperonin TRiC/CCT' [ http://www.nature.com/nsmb/journal/v15/n7/abs/nsmb.1436.html or http://www.scribd.com/doc/3321492/Mechanism-of-lid-closure-in-the-eukaryotic-chaperonin-TRiCCCT ].

clipped from www.thinkgene.com

Virtually all proteins have to be folded-some in complex configurations-in order to function properly, and many are known to require a molecule called a chaperone to fold them. Frydman estimates that perhaps 10 percent of the proteins needing chaperones must have one that, like TRiC, is part of the subset called chaperonins. Other work done in Frydman’s lab has shown that proteins that have very complex folds seem to require chaperonins.

Biopython 1.47

2008-07-27T01:01:00.003+05:30

Hope you guys are convinced to explore Python for you job after reading http://www.bioinformaticszen.com/2008/02/bioinformatics-zen-faq/ and more recently Mark Bieda's resentment at http://markbieda.wordpress.com/2008/06/10/i-wish-i-had-started-with-python-earlier/ and some reasons and his ways at http://markbieda.wordpress.com/2008/06/18/python-for-perl-programmers-and-bioinformatics-people/ on how to go about it. I have one more reason for you to follow the wave, the biopython's release of version 1.4.7 [ http://biopython.org/wiki/Download ] ... I am also noticing a recent rise in interest for the language LittleB based on lisp [story via Eureka].

clipped from biopython.org

The latest release is Biopython 1.47, released on 5 July 2008. Get it from our Download page.

BZ: Bioinformatics Career Survey 2008

2008-07-27T00:48:00.001+05:30

Mike is running a survey about Bioinformatics Career [ http://www.bioinformaticszen.com/2008/07/creating-a-picture-of-different-careers-in-bioinformatics/ ]. Although it is running for quite some time, but I just came back to blog-world. I would appeal to all the people in the domain to fill this. The more the samples, the better the survey and more over this is time of data driven science [ http://www.wired.com/science/discoveries/magazine/16-07/pb_theory ] though Deepak has some issues with this development [ http://mndoci.com/blog/2008/06/25/chris-anderson-you-are-wrong/ ].

clipped from www.bioinformaticszen.com

When I wrote a post about my opinion of doing a career in bioinformatics I got the impression from the comments that this was something many people wanted more information about. Pedro and I had some discussion, and thought it might be interesting to create an online survey, to get current researchers’ opinions of working in the field of bioinformatics. So, the below survey begins today (July 1) and is filled out by as many people as possible over the course the next month. The data is then released into the public domain at the start of next month (August 1). Anyone who is then interested can contribute back analysis of this data, so that on September 1 hopefully we can compile together lots of interesting statistics and graphs into a handy document discussing the highs and lows of career in bioinformatics.

A Book Review from Adaptive Complexity: Microcosm E. coli and The New Science of Life

2008-07-27T00:14:00.001+05:30

After reading this review [ http://www.scientificblogging.com/adaptive_complexity/e_coli_as_biologys_decoder_ring? ] from Michael White [ http://www.scientificblogging.com/mwhite74/feed ] on a book from Carl Zimmer [ http://blogs.discovermagazine.com/loom/ ] titled 'Microcosm: E. coli and The New Science of Life' [ http://www.amazon.com/gp/search?ie=UTF8&keywords=Zimmer%20e.%20Coli&tag=funnierthanyo-20&index=books&linkCode=ur2&camp=1789&creative=9325 ], I am quite eager to read this book.
Presently I am reading a book recommended by Myers [ http://scienceblogs.com/pharyngula/ ] titled 'Developmental Plasticity and Evolution' [ http://www.amazon.com/gp/product/0195122356 ] by Mary Jane West-Eberhard [ http://www.stri.org/english/scientific_staff/staff_scientist/scientist.php?id=35 ]. This along with the article referred in the above review 'Evolutionary Dynamics of Prokaryotic Transcriptional Regulatory Networks' [ http://dx.doi.org/10.1016/j.jmb.2006.02.019 ] by Madan Babu and Eisen talking about one of the heroes of bioinformatics at http://phylogenomics.blogspot.com/2008/06/connection-between-video-games-and.html , I am pretty much convinced that Evolutionary biology has lot of opportunity for Bioinformatics guys to be busy for eternity, not just "500 years of exciting problems" [ http://tex.loria.fr/historique/interviews/knuth-clb1993.html ]!

clipped from www.scientificblogging.com

For such a slender book, Microcosm covers a wide-ranging selection of science. Zimmer begins by recapping the key events in the history of molecular biology, events in which E. coli was frequently a central player. Once scientists realized that E. coli had genes just like animals and plants, this gut bacterium gradually became one of the favorite model research systems in the then hot, new science of molecular biology. Experiments in E. coli revealed how genes are structured, how DNA is replicated, and how genes are controlled. Marshall Nirenberg and his colleagues cracked the essentially universal genetic code using cellular components from E. coli. Joshua Lederberg used bacterial 'sex' to bring the formidable tools of genetics to bacterial studies. Today, E. coli is the most well-mapped organism on the planet.

Last Lecture

2008-07-26T21:58:00.001+05:30

Randy Pausch [ http://en.wikipedia.org/wiki/Randy_Pausch ] passed away yesterday morning. I admire him a lot and whenever I feel down, I watch his Last Lecture [ http://www.youtube.com/watch?v=ji5_MqicxSo ]. May his soul rest in peace ...

clipped from www.thelastlecture.com

454

2008-06-20T14:56:00.004+05:30

I have been trying to understand the pyrosequencing technology, popularly known as 454. The very first question which came to my mind is WHY it is called 454? Is it because .454 kasool Handgun is sleeker then Shotgun ( Cf. Whole Genome Shotgun Sequence ) and slowly becoming as powerful ... or is it because of the fact that "In one instrument run sequence a minimum of 20 million base pairs in 4.5 hours" ... just a conjecture, if you guys have a proof, please share !
Anyways coming back to the technology, what I understand (below is a crude attempt to visualize it) is that there are 4 containers containing A,T,G and C which sequentially pass over the bowl containing a mixture of Polymerase, Sulfurylase, Luciferase, Apyrase and the DNA fragment to be sequenced. The addition of the nucleotide will light up ( thanks to lucifer-ase ) the CCD (a thing close to what we have in a digital camera and the costliest part of the instrument). Since we know the culprit nucleotide container, peak telling how many of that nucleotide is added, we can sequentially proceed and keep adding letters to the reads... this is happening in a massively parallel fashion and that brings out the power of 454...

Biocomputing

2008-06-14T18:14:00.001+05:30

Haynes et al. in the article "Engineering bacteria to solve the Burnt Pancake Problem" [ http://www.jbioleng.org/content/2/1/8#IDAJQH0E ] demonstrate how one can take advantage of direction sense embedded in DNA by coding 5'-3' as 0 and 3'-5' as 1 binary coding to convert cells into computers.

clipped from www.msnbc.msn.com

Ron Weiss, an assistant professor of electrical engineering and molecular biology at Princeton University, said the new study provides a “nice demonstration that there’s some capability to instruct cells to carry out computational tasks. And I think this certainly brings a new aspect to what’s been demonstrated before.” Weiss, who wasn’t involved with the research, said the effort is another sign that the field is progressing, despite its relative infancy. “We really are at the beginning, at the vacuum tube stage or something like that, if you use the comparison with computer electronics,” he said.

International Conference on Functional Programming 2008

2008-06-07T21:44:00.002+05:30

The 11th ICFP ( http://icfpcontest.org/ ) Programming Contest is from Friday, July 11, 2008 to Monday, July 14, 2008. More details on what the contest is like, check out http://cpoucet.wordpress.com/2008/06/11/icfp-contest-2008/ .
Mail queries to Tim Sheard at sheard@cs.pdx.edu.

clipped from lambda-the-ultimate.org

The ICFP Programming Contest is one of the most advanced and
prestigious programming contests, as well as being a chance to show
off your programming skills, your favorite languages and tools, and
your ability to work as a team. The contest is affiliated with the
International Conference on Functional Programming. Teams consisting
of one or more participants, from any part of the world, using any
programming language, may enter.

Unofficial Google Shell

2008-06-03T18:45:00.003+05:30

I have been waiting for google shell. I thought they will name it gshell, but it turns out:
goosh.org - the unofficial google shell ( http://goosh.org/ ).
It is pretty neat and useful in its neonatal state itself, like to read a feed, you can type, for eg:

r http://www.google.com/reader/public/atom/user/11136726768885096694/label/cbn-roll

and gets default 4 latest feeds, or you can check out a ncbi accession by typing, for eg:

NP_001108376

and get to the Fugu fish refseq link.

The most useful thing is the fact that it it text based and quite easy to parse visually as well as via scripts.
Another cool feature for me right now is the shell translation, like to get the english word for the norwegian word 'hvor', all I have to do is type:

t no en hvor

which correctly says:
translating "hvor" from "no" to "en":

"where"

More discussion going on at http://tech.slashdot.org/article.pl?sid=08/06/02/222234&from=rss .
Enjoy!

clipped from goosh.org

Goosh goosh.org 0.4.3-beta #1 Mon, 02 Jun 08 22:28:01 UTC Google/Ajax

help

command aliases parameters function
web (search,s,w) [keywords] google web search
lucky (l) [keywords] go directly to first result
images (image,i) [keywords] google image search
wiki (wikipedia) [keywords] wikipedia search
clear (c) 
clear the screen
help (man,h,?) [command] displays help text
news (n) [keywords] google news search
blogs (blog,b) [keywords] google blog search
feeds (feed,f) [keywords] google feed search
open (o) <url> open url in new window
go (g) <url> open url
more (m) 
get more results
in (site) <url> <keywords> search in a specific website
load 
<extension_url> load an extension
video (videos,v) [keywords] google video search
read (rss,r) <url> read feed of url
place (places,map,p) [address] google maps search
lang 
<language> change language
addengine 

add goosh to firefox search box
translate (trans,t) [lang1] [lang2] <words> google translation

Conserved Domain Database (CDD) has been updated

2008-06-01T16:59:00.001+05:30

The CDD [ http://www.ncbi.nlm.nih.gov/Structure/cdd/cdd.shtml ] and the its search tool [ http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi ] has been recently updated. Also check out the latest approach for "automatically assigning subcellular locations" to protein from Newberg and Murphy [ http://pubs.acs.org/cgi-bin/abstract.cgi/jprobs/asap/abs/pr7007626.html ], more at http://www.sciencedaily.com/releases/2008/05/080512145232.htm .

clipped from beckerinfo.net

CQNCer , players in dna sequencing

2008-06-01T14:50:00.001+05:30

So finally we see Helicos in action (via http://microarray.wordpress.com/2008/04/11/entire-genome-sequencing-from-single-molecule-dna/ ) , http://www.sciencemag.org/cgi/content/full/320/5872/106 .
They have sequenced the M13 phage virus (about .3 mill bases) ... sometime back I was surveying the trends in sequencing and captured them at http://www.slideshare.net/sharma_animesh/cqncer .
These strategies are generally being used for reqsequencing or sequencing smaller genomes. Though I see that even smaller genomes like Stenotrophomonas maltophilia ( Steno, http://genomebiology.com/2008/9/4/R74 ) done the old way... more at http://www.biologynews.net/archives/2008/05/07/superbug_genome_sequenced.html . Same goes for recently sequenced Platypus gnome ( http://www.nature.com/nature/journal/v453/n7192/abs/nature06936.html , http://www.eurekalert.org/pub_releases/2008-05/embl-pgs050608.php ) and the Papaya genome draft ( http://www.nature.com/nature/journal/v452/n7190/abs/nature06856.html , http://www.biologynews.net/archives/2008/05/14/scientists_decipher_fruit_tree_genome_for_the_first_time.html ) but given these new cheaper sequencing technologies and better algorithms, we are bound to see cooler changes in Tree-of-Life ( http://www.boston.com/news/science/articles/2008/04/28/tree_of_life_continues_to_evolve/ ) !

clipped from www.ncbi.nlm.nih.gov

The Genome database provides views for a variety of genomes, complete chromosomes,
sequence maps with contigs, and integrated genetic and physical maps.
The database is organized in six major organism groups: Archaea,
Bacteria, Eukaryotae,
Viruses, Viroids, and
Plasmids and includes complete chromosomes, organelles and plasmids as well as draft
genome assemblies.

Predicting what you thinking ...

2008-05-31T18:48:00.001+05:30

Mitchell [ machine learning guru, author of the most lucid book on the subject, http://www.cs.cmu.edu/%7Etom/mlbook.html ] et al. in the article titled "Predicting Human Brain Activity Associated with the Meanings of Nouns" [ http://www.sciencemag.org/cgi/content/abstract/320/5880/1191 ] have developed a computational model based on fMRI scans predicting what you thinking ... more at http://www.nature.com/news/2008/080529/full/news.2008.864.html .

clipped from www.nature.com

A computer model has been developed that can predict what word you are thinking of. The model may help to resolve questions about how the brain processes words and language, and might even lead to techniques for decoding people’s thoughts.

The team then used functional magnetic resonance imaging (fMRI) to scan the brains of 9 volunteers as they looked at images of the nouns. The researchers then fed the model 58 of the 60 nouns to train it. For each noun, the model sorted through a trillion-word body of text to find how it was related to the 25 verbs, and how that related to the activation pattern.

Music from DNA string

2008-05-23T01:33:00.001+05:30

Had this thought of converting DNA strings to MIDI music files using Sean [ http://groups.google.com/groups/profile?enc_user=N0KQ5w8AAAD0VD270uUpCbIPP1utz--x ] 's MIDI module for PERL way back [ http://groups.google.com/group/DNA-Music/about ] in JNU days when I was working on improving fourier transform based coding region prediction [ http://www.ncbi.nlm.nih.gov/pubmed/11927773 ] .
Recently I got this opportunity to participate in YSR [ http://www.sars.no/seminars/Retreat_2008_invitation.pdf ] where I finally hacked together these ideas of converting the DNA strings to MIDI music. I am just coming back from there. The evening was full of serious biology talks from mostly experimental biologist [ http://www.uib.no/People/ash022/BYSR08_program_HR_GB.pdf ] where I hope, I lived up to the reputations of being a Bioinformatician, for not just having fun with computers but justifying out salary by plugging in the biological data somehow
The presentation tries to explain the flowchart in 4 slides [slideshare: http://www.slideshare.net/sharma_animesh/ysr-presentation-animesh-rev/ , pdf: http://sharma.animesh.googlepages.com/YSR_Presentation_Animesh_Rev.pdf ].
The code [ http://sharma.animesh.googlepages.com/create_music_from_string.pl ] depends on Sean's MIDI module for PERL which can be downloaded from CPAN [ http://search.cpan.org/~sburke/MIDI-Perl-0.81/lib/MIDI/Simple.pm ] takes in sequence in fasta format [ eg: http://sharma.animesh.googlepages.com/telo.fas , it does accept multiple sequences ] and outputs a MIDI file [ eg: http://sharma.animesh.googlepages.com/telo.fas.0.midi ] and DFT values [ eg: http://sharma.animesh.googlepages.com/telo.fas.0.fft , NOTE: for multiple sequences, the counter starts from 0 for 1st sequence and goes to N-1 for Nth sequence].
To play the midi files [ eg: http://sharma.animesh.googlepages.com/FLJ20436.fas.0.midi generated from http://sharma.animesh.googlepages.com/FLJ20436.fas ] in Linux, I would recommend to install timidity [ http://timidity.sourceforge.net/ ].
Lots need to be done to make this thing do something practical, like I need to change the code to to Fast Fourier transform, employ a window based strategy to walk through the whole genome and create region based music ...
By the way, there is a good use to look at the other generated file *.ff [ eg: http://sharma.animesh.googlepages.com/telo.fas.0.fft ] as well, it says a lot about the repeat length hidden in the sequence.
Like the sequence ( http://sharma.animesh.googlepages.com/telo.fas ) I used initially is:
>teloseq
AGGGTTAGGGTTAGGGTTAGG
which has telomere repeat signal TTAGGG of length 6, now when we look at its fft file ( http://sharma.animesh.googlepages.com/telo.fas.0.fft ) and plot it ( http://sharma.animesh.googlepages.com/telo.jpg ), we observe a peak (defined in the original paper as more then signal of 4) around 6 and 3.
Coding regions generally show a peak at 3, hypothesis being the codon usage bias in the coding strand, while non-coding regions lack this property (not useful in eucaryotic sequences as they have introns too)... so in a way if you input a sequence from coding region to this code, you might find a peak at 3 in the *.fft generated.
To read more about MIDI encoding, I would suggest "The Musician's Guide to MIDI by Christian Braut" [ http://www.amazon.com/Musicians-Guide-Sybex-Macintosh-library/dp/0782112854 ].
Now I need to train my ears to listen to lots of coding and non-coding region music and then given some unknown sequence, predict if it belongs to either of these regions just by listening to its midi file

clipped from www.sars.no

Young Scientist's Retreat - Bergen 2008, Invitation
Open to all phd students and postdocs at MBI, Dept. of Informatics,
Dept of Biomedicine, Sars Centre and CBU.
To register contact Ståle Ellingsen.

May 22

Sydneshaugen Skole 12:00-18:00
Host: MCB Research School

Faster way to screen nonribosomal peptides

2008-05-20T22:39:00.001+05:30

Bandeira et al. in the paper "De Novo Sequencing of Nonribosomal Peptides" ( http://www.springerlink.com/content/n532545722750915/ ) outline a strategy based on MS alignment leading to faster nonribosomal peptides (example penicillin) screening. This has potential to hasten the process of finding new drugs (better say candidates?).

clipped from www.genengnews.com

If you imagine the structure of an NRP as a cyclic string of beads, then the new algorithms both decipher the mass of each bead based on the mass spectrometry and determine the order of the beads within the ring crucial pieces of information for uncovering both the structure of the molecule and its pharmacological activities.

Rosendal Meet

2008-05-12T00:50:00.006+05:30

I was attending the Rosendal meet (from 6-8 th May, 2008) between members of CBU (Computational Biology Unit ... and Department of Informatics, University of Bergen, Norway) and MB-NIMR (Division of Mathematical Biology at National Institute for Medical Research, MRC, London group, UK).

Thanks to Xianjun for posting the pictures from the trip at http://picasaweb.google.com/sterding/Rosendal2008 (the beautiful pic above is coming from there) and Anders at http://flickr.com/photos/lanzen/sets/72157605097913531/ . The ones from my camera are available at http://picasaweb.google.com/sharma.animesh/Rosendal_meet?authkey=wFgJVxkGwTU .

Workshop: BREW 2008

2008-05-05T03:58:00.001+05:30

I was in Bielefeld, Germany for BREW2008 workshop. I was quite suspicious of this place as people (http://www.cebitec.uni-bielefeld.de/index.php?option=com_cudbhome&task=show&cudbname=sniemeie , http://www.cebitec.uni-bielefeld.de/index.php?option=com_cudbhome&task=show&cudbname=twittkop ) informed me that Bielefeld does not exist ( http://en.wikipedia.org/wiki/Bielefeld_Conspiracy )! So I went out and took loads of snaps ( http://picasaweb.google.com/sharma.animesh/BREW2008?authkey=ranwczoE2-Q ) to prove its existence ... well on second thought map is not the territory ( http://en.wikipedia.org/wiki/Thomas_Kuhn ).
Coming back to the workshop, this is a nice event with a great concept ( http://bib.oxfordjournals.org/cgi/content/abstract/9/3/250 ) organised by and for people from:
* International Graduate School in Bioinformatics and Genome Research, University of Bielefeld, Germany ( http://www.cebitec.uni-bielefeld.de/gradschool/ )
* Molecular and Computational Biology Research School, University of Bergen, Norway ( http://www.mcb.uib.no/ )
* Max Planck Research School for Computational Biology and Scientific Computing (IMPRS-CBSC), Berlin, Germany ( http://www.imprs-cbsc.mpg.de/ )
* EBI PhD Program, European Bioinformatics Institute, England ( http://www.ebi.ac.uk/ )
* ComBi Program, University of Helsinki, Finland ( http://www.cs.helsinki.fi/ )
I had a great time there and would recommend people from the above list to go for BREW 2009 which is tentatively scheduled after a year at http://www.cs.helsinki.fi/ ... (?)

clipped from www.techfak.uni-bielefeld.de

BREW 2008

Bioinformatics Research and Education Workshop, Bielefeld, Germany April 28-30 2008

March 1th -- Abstract submission deadline

March 7th -- Paper submission deadline

March 20th -- Review deadline

April 11th -- Deadline for final version of paper

April 28th -- Meeting in Bielefeld

About BREW

BREW aims to give an introduction to scientific conferences, including submission, peer review
and presentation of scientific papers. All participants must submit a paper
and take part in reviewing papers from other participants in order to attend the workshop.
After the review deadline, you have a chance to make alterations to your paper
and deliver a final version. All participants will present their paper at the meeting in April.

BREW is held in an around-robin fashion each year at the participating PhD programs:
Bergen (2007),
Cambridge (2006),
Berlin (2005),
Helsinki (2004),
Bielefeld (2003), and
Hinxton (2002).