{ C O N T E N T S }
Volume 8, Issue 3

President's Letter

Rost To Become
Next ISCB President

Rocky'05 Bioinformatics
Conference Colorado
in December

Update on
PLoS Computational Biology

Putting Students
in the Spotlight

Call for 2006
Awards Nominations

ISMB 2006 in Brazil

New ISCB Membership Site
Now Open

ISMB 2006
Call for Tutorials

RECOMB 2006 Registration Opening This Week

ISMB 2005 SIGs Recap

Israeli Bioinformatics Symposium 2005
A Prelude to ECCB’06

Advertising & Corporate Membership Opportunities

Post Your News, Events
and Degree Programs
on ISCB Website

Upcoming Events
& Conferences

News from the Field

Acknowledgements and Contributors



Copyright © 2005 International Society for Computational Biology.
All rights reserved

ISMB 2005 Special Interest Groups Meeting Recaps

Seven Special Interest Groups (SIGs) met in conjunction with ISMB 2005 in Detroit, Michigan, this past summer. Each year ISCB invites select SIGs to share a summary of their meetings with our readers as a means of potentially expanding awareness of their work both during the conference and throughout the year. Below is a recap of three such SIGs from 2005.

Automated Function Prediction 2005 SIG Summary

Automated function prediction is maturing into a sub-field of its own within computational biology. On the one hand, with all the data coming in from various genomics projects, there is a dire need for annotation of genes and gene products. On the other hand, the computational biology community has just begun to recognize the need for bona fide automated annotators: algorithms that are capable of receiving sequence or structure data and producing a succinct function prediction, and doing so not only by homology based methods. Furthermore, with the emerging predictors, comes the need for assessment: how well are the function predictors doing?

The first Automated Function prediction SIG meeting was set up as focal point for the various research groups involved in this new and exciting field. The 17 talks that were given discussed the various aspects of computationally predicting function from sequence, function from protein 3D structure, function from genomic context, and the ways and means of assessing the quality of function prediction programs. The response from leading researchers was extraordinary, with Michael Sternberg from Imperial College delivering a keynote talk, and Olivier Lichtarge (Baylor College of Medicine), Russ Altman (Stanford University), Patricia Babbitt (UC San Francisco) and Adam Godzik (The Burnham Institute) leading plenary sessions on predicting function from structure, context based predictions, analysis of function site, functional diversity, and challenging the servers. Server challenge was done by Martin Jambon and Iddo Friedberg from The Burnham Institute, and reported by Iddo Friedberg. A poster session and lively discussions on the various aspects of this emerging field served to enhance this exciting day.

With 89 registrants and a head count of over 100 attendees, the one day meeting was considered extremely successful. It was agreed that function prediction is indeed a distinct field, and that there is a scientific community which is interested in meeting regularly to discuss and develop it. Excerpts from the meeting will be published in Protein Science in early 2006. More information can be found at: http://ffas.burnham.org/AFP.

BioLINK 2005 SIG Summary

BioLINK SIG is concerned with the rapidly maturing field of biomedical text mining: using techniques from natural language processing, information extraction and information retrieval to automate knowledge discovery from the biomedical literature and free text.

This was the fifth consecutive BioLINK SIG at ISMB, and the first time that it was held jointly with the Workshop on Linking Biological Literature, Ontologies and Databases of the Association for Computational Linguistics (ACL), whose annual conference took place in parallel to ISMB in neighboring Ann Arbor. The joint meeting was the largest BioLINK SIG so far (about 100 people in the audience), and attracted many excellent full-length and short papers, posters and talks. All papers, both long and short, went through a full peer-review process. Eight full-length papers were accepted out of 18 full-paper submissions; these were presented and published in the ACL Workshop proceedings. Additionally, 9 short papers were accepted out of 21 short submissions and included in the SIG handouts; 5 of them were selected for presentation during the workshop as short talks, for a total of 13 research talks.

The keynote speaker was Dr. Judith A. Blake, who is one of the leaders of the Mouse Genome Informatics Group at the Jackson Laboratories and a co-founder of the Gene Ontology Consortium. She presented a fascinating talk about the current challenges of integrating biomedical literature with bioinformatics and genomics information, to support curation of model organism databases. This successful meeting concluded with a lively poster session of 22 selected posters, followed by dinner at Detroit’s Greek Town. We would like to thank the many people who helped in organization and review, and particularly Steven Leard for his outstanding organizational support for the SIGs. We are looking forward to next year’s meeting in Brazil.

Bioinformatics Open-Source Conference SIG Summary

The Bioinformatics Open-Source Conference (BOSC) is held annually as a Special Interest Group with the Intelligent Systems for Molecular Biology (ISMB) conference, hosted by the Open Bioinformatics Foundation (www.open-bio.org). The aims of the conference are: to showcase the work of projects devoted to creating open-source bioinformatics software, or open standards of communication and information management; to facilitate communication and interaction between developers working within projects, and between project members and members of the wider community interested in their projects; and educate the research community on the value of open-source software and open standards. The meeting consists of keynote speakers invited by the BOSC organizing committee for outstanding achievements in Bioinformatics, Open-Source development, or overall research skill; twenty-minute presentations selected from a competetive pool of abstracts submitted for consideration; five-minute lightning-talks and software demonstrations; and Birds of a Feather meetings on specific projects or areas of interest.

BOSC 2005 included two keynote speakers: Jason Stajich, a key developer of the very popular BioPerl software package, and Hilmar Lapp, speaking on behalf of the Open Bioinformatics Foundation. Mr. Stajich's talk, 'Building Bioperl: lessons for Open-Source and Bioinformatics', presented an historical summary of the development of the BioPerl system, demonstrating the evolution of typical open-source software projects, and the challenges they face. Dr. Lapp presented the new organizational structure of the Open Bioinformatics Foundation as a not-for-profit organization with memberships open to all those who support their mission at no charge. All participants of BOSC 2005 were given the opportunity to become
members, and participate in shaping the future direction of the foundation.

The full length talk lineup included 15 speakers whose abstracts were chosen from a very competitive group. They demonstrate the diversity and quality of work being done in the open-source bioinformatics community. Highlights include talks on a variety of new tools for managing bioinformatics analyses, and result visualization; discussions of a number of powerful data-mining systems; and many tools designed to use or manage an ever expanding list of web services for bioinformatics. The schedule for BOSC 2005, including links to abstracts, images, and slides, can be accessed at http://open-bio.org/bosc2005/finalProgram.

Preparation for BOSC 2006 in Brazil is already underway. Check the OBF website (www.open-bio.org) for more details.