- beg-sysbiol@psb.ugent.be - maillist.psb.ugent.be

[Beg-sysbiol] Interessant? Cross-species analysis of biological networks by Bayesian alignment
by Roeland Merks 17 Jul '06

17 Jul '06

Cross-species analysis of biological networks by Bayesian alignment Johannes Berg and Michael Lässig PNAS published 11 July 2006, 10.1073/pnas.0602294103 http://www.pnas.org/cgi/content/abstract/0602294103v1 <http://www.pnas.org/cgi/content/abstract/0602294103v1?etoc> ------------------------------------------------------------------------ -- Roeland Merks Department of Plant Systems Biology VIB / Ghent University Technologiepark 927 B-9052 Ghent, Belgium http://www.psb.ugent.be Tel. +32-(0)9-33 13825 email: post(a)roelandmerks.nl http://www.roelandmerks.nl weblog: http://biolog.volkskrantblog.nl

1 0

[Beg-sysbiol] [Fwd: scripts etc Syntren]
by Tom Michoel 07 Jul '06

07 Jul '06

I put those scripts in the "java/timvdb" subdirectory of the segal project. Tom -------- Original Message -------- Subject: scripts etc Syntren Date: Fri, 07 Jul 2006 18:01:18 +0200 From: Tim Van den Bulcke <Tim.VandenBulcke(a)esat.kuleuven.be> To: tom.michoel(a)psb.ugent.be Hey Tom, hierbijgevoegd al de scripts die we overlopen hebben woensdag. Ik kon niet aan de gewijzigde genomica op de computercluster, dus ik zal Koen contacteren van zodra hij terug in Belgie is (maandag of dinsdag) daarvoor. De files: inference_Genomica_submitter.py zal jullie het meeste interesseren waarschijnlijk. Dit start een reeks genomica runs op de cluster. ./expressiondata/ bevat enkele scripts om expressiedata te genereren, misschien nog niet relevant op dit ogenblik, maar er zitten misschien een paar stukken code in die herbruikbaar zijn om dingen op een computercluster te draaien. in ./analyse/ zitten een reeks python scripts om data te verzamelen en te analyseren (ook om de output van genomica te converteren). Syntren zelf kan je afhalen op de website: http://homes.esat.kuleuven.be/~kmarchal/SynTReN/index.html Prettig weekend! Tim. -- _______________________________________________________ ir. Tim Van den Bulcke research assistant ESAT-SCD K.U. Leuven Kasteelpark Arenberg 10 B-3001 Leuven-Heverlee Belgium room: 03.11/14 tel: +32 16 32.86.43 fax: +32 16 32.19.70 email: tim.vandenbulcke(a)esat.kuleuven.be homepage: http://homes.esat.kuleuven.be/~vdbulcke/ personal homepage: http://timvandenbulcke.objectis.net _______________________________________________________

1 0

[Beg-sysbiol] Tim VdB will visit us on wednesday
by Tom Michoel 03 Jul '06

03 Jul '06

time to be determined. tom

1 0

[Beg-sysbiol] [Fwd: confidentieel]
by Steven Maere 30 Jun '06

30 Jun '06

-- ================================================================== Steven Maere, PhD Dept. of Plant Systems Biology Bioinformatics & Evolutionary Genomics research group Ghent University / VIB Technologiepark 927 B-9052 Gent BELGIUM tel: +32 9 331 3692 fax: +32 9 331 3809 email: stmae(a)psb.UGent.be http://www.psb.UGent.be http://bioinformatics.psb.ugent.be/ ==================================================================

1 0

[Beg-sysbiol] syntren paper
by Steven Maere 30 Jun '06

30 Jun '06

Dear all, As some of you know already, we have an opportunity to publish some of our Segal work in an article that Tim and Kathleen were invited to submit to BMC Bioinformatics about their data generator SynTReN. Following discussion with Tom, we came up with a tentative plan for the content of the paper (it should somehow be linked to SynTReN, of course, see attachment), which is due by the end of September. Only one problem: not everybody of our sysbiol group can be featured on the paper due to the fact that there's already 5 authors from Leuven and Antwerp. Let's discuss tomorrow (when exactly, Yves ?)... Cheers, Steven -- ================================================================== Steven Maere, PhD Dept. of Plant Systems Biology Bioinformatics & Evolutionary Genomics research group Ghent University / VIB Technologiepark 927 B-9052 Gent BELGIUM tel: +32 9 331 3692 fax: +32 9 331 3809 email: stmae(a)psb.UGent.be http://www.psb.UGent.be http://bioinformatics.psb.ugent.be/ ==================================================================

4 3

[Beg-sysbiol] regulation programs test
by Eric Bonnet 14 Jun '06

14 Jun '06

Hi there, I did a test of the regulation programs (segal, biocomp) using the Harbison data set. What about a small get-together tomorrow to discuss the results ? Eric

3 3

[Beg-sysbiol] Iterative Reconstruction of Transcriptional Regulatory Networks: An Algorithmic Approach
by Roeland Merks 12 Jun '06

12 Jun '06

Iterative Reconstruction of Transcriptional Regulatory Networks: An Algorithmic Approach http://compbiol.plosjournals.org/perlserv/?request=get-document&doi=10.1371… -- Roeland Merks Department of Plant Systems Biology VIB / Ghent University Technologiepark 927 B-9052 Ghent, Belgium http://www.psb.ugent.be Tel. +32-(0)9-33 13825 email: post(a)roelandmerks.nl http://www.roelandmerks.nl weblog: http://biolog.volkskrantblog.nl

1 0

[Beg-sysbiol] Integrated biclustering of heterogeneous genome-wide datasets for the inference of global regulatory networks
by Tom Michoel 12 Jun '06

12 Jun '06

http://www.biomedcentral.com/1471-2105/7/280

1 0

[Beg-sysbiol] Entrez PubMed
by Yves Van de Peer 11 Jun '06

11 Jun '06

http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&dopt=Abstract&db…

1 0

[Beg-sysbiol] module testing
by Eric Bonnet 08 Jun '06

08 Jun '06

hi there, I wrote a program to test the functional coherence of the modules as described in Segal's PhD. The test is simply looking for any significant overrepresentation of a given GO category in each module, measured with the hypergeometric test with a bonferroni correction. For each module, the result is counted as the percentage of genes associated with a significant GO category (p<0.05) / number of genes tested for this module. biocomp modnet with HC ---------------------- Mean 52.48 Std Dev 38.92 Num. Modules > 0% 40 Num. modules > 50% 27 Segal ----- Mean 50.47 Std Dev 34.43 Num. Modules > 0% 43 Num. modules > 50% 26 As you can see our results are similar to those of segal, even slightly better. 40 modules /50 have a least 1 "significant" gene and 26 modules have more than 50% of the gene than are "significant". The average percentage is 52% per module, value higher compared to Segal. Besides that, the computing with the Hierarchical clustering approach is amazingly fast: less than 5 minutes for the Yeast data (2355 genes x 173 experimenents and 321 regulators), compared to several hours for the classical greedy approach. That would be a big advantage for applying the technique to big datasets like Arabidopsis or Human.. I'll try to use the data contained in the paper send by Steven to test the value of the regulation programs. Eric PS: for those interested, the script: /nas/biocomp/projects/segal/java/erbon/gotest.pl perl gotest.pl gene_association.sgd segal_mod.tab

2 2