Plant & Animal Genome VII Conference

W39

UTILIZATION OF BIOINFORMATIC APPROACHES FOR THE EVALUATION OF A BOVINE EXPRESSED SEQUENCE TAG (EST) DATABASE.

¹Monsanto Company, AA3C, 700 Chesterfield Parkway, St Louis, MO 63198 USA

²USDA, Meat Animal Research Center, Box 166, Clay Center, NE 68933 USA

In mammals whole genome and EST sequencing is ongoing in few target species. Bioinformatics is starting to enable biologists to organize, query and disseminate this information. For bovine, good genetic linkage maps exist, although a paucity of genes (ESTs) are available for study. We have targeted specific tissues and created non normalized cDNA libraries for EST sequencing. Total 5í EST sequences generated thus far are 19753 within which 10365 (89 %) singletons and 1302 (11 %) clusters are found to create a 11667 bovine unigene database. Following annotation against a non-redundant amino acid and nucleic acid database, for singletons 52% were unique and the remaining 48% were annotated, while for clusters 21 % were unique and 79 % annotated. Once this bovine unigene database was created three data mining approaches were employed 1) pathway mapping against the yeast genome, 2) single nucleotide polymorphism (SNP) identification and 3) simple sequence repeat (SSR) identification. Bovine ESTs orthologous to yeast genes at a significance level of P < 1E -12 allow segregation by metabolic pathway. The KEGG database (URL: http://www.genome.ad.jp/kegg/kegg.html) was used to superimpose these bovine genes. For the seven yeast metabolic pathways examined the average representation of bovine genes within each was 41 %. For SNP identification, clusters of bovine ESTs (n > 10) were examined following alignment by the PHRAP program using criteria that at least two differences must be found relative to the consensus nucleotide at that position to call a SNP. A total of 130 potential SNPs were found by these criteria with average SNP representation of 1 in 583 nucleotides. Finally, SSR markers found were few and the majority were CA repeats in the 3í untranslated region of an EST. A subset of both SSR and SNP markers will be mapped on the USDA reference family panel. In summary, these bioinformatic approaches are a start to understanding bovine gene polymorphism rate and sequence identity relative to other species.