While many dont associate the Institute of Electrical and Electronics Engineers with bioinformatics, theres a growing kernel of interest within the IEEE Computer Society, which hosted its second International Symposium on Bioinformatics and Bioengineering (BIBE 2001) November 4-5 in Bethesda, Md.
Attendance was lower than expected due to travel restrictions imposed by many organizations in light of the events of September 11, but the intimate gathering of around 40 afforded ample opportunity for in-depth discussions and lively debate. With the group split into even smaller components on the second day of the symposium for parallel sessions on bioinformatics and bioengineering, the atmosphere was, as one attendee described it, more like a workshop than a conference.
Nikolaos Bourbakis, director of the Information Technology Research Institute at Wright State University and general chair of BIBE 2001, said the meeting had its roots in the mid-1990s, when bioinformatics began to firmly establish itself as a subset of computer science. Since bioinformatics ultimately grapples with the data generated by the fruits of bioengineering, Bourbakis said its important for BIBE to address both sides of the equation.
The first days talks addressed interoperability of biological databases, phylogenetic trees, biological information sources, and protein structure prediction. Cathy Wu of the National Biomedical Research Foundation at Georgetown University kicked off the meeting with a discussion of the Protein Information Resource (http://pir.georgetown.edu). Additional highlights included a graphical query tool for phylogenetic databases presented by Giovanni Modica of Mississippi State University, and prototype software to automate database annotation presented by Philip Lord of the University of Manchester. A talk on the GIMS (Genome Information Management System) data warehouse by Mike Cornell, also of the University of Manchester, spurred a follow-on discussion among attendees on the need for better audit trails for databases and data warehouses derived from primary NCBI or EMBL data. Once the data is updated, several participants noted, it is often impossible to find information that appeared in a previous search.
The second days bioinformatics track covered gene expression and gene regulation, biological data mining and knowledge discovery, and sequence search and alignment. Arvind Bansal of Kent State University discussed his use of pair-wise genome comparisons as the foundation for automating reconstruction of microbial metabolic pathways. Aaron Davidson of the University of Alberta presented a fast-pruning algorithm for sequence alignment that he said improves upon the Needleman-Wunsch full-matrix dynamic programming method. Several attendees questioned the algorithms ability to perform global sequence alignment, and Davidson noted it is most effective for similar sequences of roughly equal lengths.
Proceedings from the symposium are available from the IEEE at http://computer.org or http://shop. ieee.org/store.