Skip to main content
ARS Home » Plains Area » Lubbock, Texas » Cropping Systems Research Laboratory » Livestock Issues Research » Research » Publications at this Location » Publication #164541


item Dowd, Scot

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 6/20/2004
Publication Date: 6/21/2004
Citation: Dowd, S.E. 2004. Livestock issues addressed with distributed bioinformatics applications: Gene Ontology, S-BLAST, and W.ND-BLAST [abstract]. Southwest Grid Computing Conference

Interpretive Summary:

Technical Abstract: Toward the development of targeted microarray technology that is able to efficiently and economically be used to evaluate livestock health and well-being, we have found it necessary to develop bioinformatics protocols capable of handling extremely large datasets. Gene Ontology (GO) is a controlled vocabulary that can be applied to define and describe gene products and attributes. Defining genetic messages derived from cattle and swine with GO, we can quickly group messages into functional associations allowing rapid designing of system targeted microarrays. Such targeted microarrays will allow for efficient and cost effective monitoring of hundreds of system specific genes expressed during live animal experiments. Because of the huge amount of genetic information being processed, we have found it necessary to develop distributed Basic Local Alignment Search Tool protocols that allow these important and time consuming database searches (some queries taking up to 1 month to complete) to be conducted in a fraction of the time. Using standard BLASTx (translated nt query against a database), 10,000 sequences were processed against a 750 MB database in 98.5 hours. Using either of the two distributed BLAST 1. S-BLAST (Sorcer-BLAST) or W.ND-BLAST (Windows .NET Distributed BLAST) and 8 computers connected by Gigabit network, this same search can be completed in just over 11 hours. These applications provide fault tolerance to worker node collapse and recovery from master node collapse. All include GUI interfaces for the front end and high throughput output processing with a separate GUI. Each is easy to install with minimal networking experience and has been designed for ease of use by non-computer savvy biologists. By completing these database searches faster, we are able to speed progress toward development of microarray technologies that can more efficiently and more rapidly help us develop better management practices for our livestock.