BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20111116T011500Z
DTEND:20111116T030000Z
LOCATION:WSCC North Galleria 2nd/3rd Floors
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: We addresses the problem of performing faster read alignment on GPU devices. This task is extremely computationally intensive as constant progress in sequencing technology leads to everincreasing amounts of sequence data. One of possible solutions for this problem is using the extreme parallel capacities of modern GPU devices.=0A=0AWe found memory limitations to be main performance issue in previous solutions. Attempts have been made to decrease memory consumption of the matching algorithm, and we suggest to make another step in this direction by building a much smaller index using Burrows-Wheeler Transform.  =0A =0AAt the same time we  use suffix array on host to have constant size output from each kernel and exploit different computational characteristics of both GPU and CPU. We reduced index size 12 times and achieved 3-4 time performance improvement.
SUMMARY:Fast GPU Read Alignment with Burrows Wheeler Transform Based Index
PRIORITY:3
END:VEVENT
END:VCALENDAR