Plant & Animal Genomes XVII Conference

January 10-14, 2009
Town & Country Convention Center
San Diego, CA

Comparative in-silico Analysis And Functional Classification Of Putative And Verified Imprinted Bovine Genes

I.G. Imumorin¹ , J.M. Jeff¹ , P. Dubose¹ , M. De Donato^1,2 , C. Bates¹ , G.R. Hansen³

¹  Department of Biology, Spelman College, Atlanta, GA 30314. USA

²  Laboratory Genetica Molecular, IIBCA, Universidad de Oriente, Cumana, Venezuela

³  Dept of Animal Science, North Carolina State University, Plymouth, NC 27962. USA

It has been estimated that the number of imprinted mammalian genes may be as high as 600 in the mouse using in-silico prediction methods, although in reality very few have been identified and experimentally verified in most mammalian species. We conducted a census of putative and verified imprinted genes across the bovine genome using a list compiled from human and mouse genes. Sequences of about 100 genes were analyzed for putative imprinting signatures and assigned to functional classes by presenting the genes to the gene ontology analysis database. In comparison to human counterparts, bovine NESP55 and SGCE showed the highest level of sequence identity for known imprinted and putative imprinted genes respectively while the rest varied fairly widely with average between 80 and 87%. A cluster of three imprinted genes: NDN, MKRN3 and MAGEL2 was found on BTA2, in a similar cluster and orientation as seen on HSA15 and MMU7. The results support a possible strong conservation of imprinting status and the presence of imprinting control regions known in such clusters in eukaryotic genomes. Functionally, almost half of the genes were classified as growth factors, followed by involvement in cell metabolism, transcription factors, cell matrix proteins and finally RNAs with unknown functions. This appears consistent with involvement of many imprinted genes in embryonic and fetal growth and development. Improved in-silico identification and analysis will benefit from more computationally rigorous comparative approaches including identification of common motifs, promoter usage, and 3-D structure prediction, although experimental verification will still be required.