At 11:00 AM on Tuesday, September 28, Dr. Hugh Williams of the Department
of Computer Science at the Royal Melbourne Institute of Technology will
present a CBB seminar in the 8th floor conference room.  The title and
abstract of his talk are:

                  Problems in Indexing Genomic Data
 
   Efficiency in searching databases of DNA and amino-acid sequences is
   essential as some servers process over 40,000 queries per day, and
   resolution of each query requires comparison to over two gigabytes of
   genomic sequence data.  While existing systems, which compare each
   database sequence to a query sequence, are practical at present, they
   may become prohibitively expensive as genomic databases continue to
   double in size every 14 or 15 months, and as user numbers and query
   rates grow.
 
   One possible approach to reduce query evaluation costs is to apply
   indexing techniques used in text databases to genomic sequence data.
   However, indexing genomic databases is a difficult problem that requires
   trade-offs and careful selection of heuristics.  In this talk, I detail
   some of the techniques used in our `Cafe'' indexed genomic search system,
   discuss the problems of indexing genomic databases, and present results
   of a comparison of `Cafe'' to the popular BLAST and FastA search systems.

Would anyone interested in speaking with Dr. Williams on Tuesday the 28th
please let me know as soon as possible, so that I may arrange a schedule
for his visit to NCBI that day.

Brief bio of speaker:
 
Since 1994, Hugh has been a lecturer at RMIT University in the Department
of Computer Science.  His research interests include bioinformatics,
high-performance information retrieval, compression, databases, and web
systems.  Hugh has a PhD in Computer Science that investigated the area
of his talk, as well as related work combining the fields of text database
indexing, string matching, compression, and bioinformatics.

-Stephen Altschul