Whole Transcriptome Value Index (WTVI): A Methodology for Integrating Functional Sequences from RNA-Seq Data into Animal Selection

Dramatic genetic progress resulting from animal genomic selection has become possible using high density SNP panels but challenges remain such as high genotyping errors, computational complexities of implementation, poor ascertainment in cattle breeds not included in SNP discovery used in standard SNP panels and not taking account of non-additive effects such as epistasis and dominance between SNPs. Transcriptome analysis provides information about differential expression of candidate genes that can potentially affect traits of interest. RNA-Seq has become an important tool to generate large scale exome sequences that can be mapped back to the genome to identify new candidate genes through simultaneous availability of both sequence and expression data. Due to the plummeting cost of this technology, we propose computational strategies for direct integration of exomic sequence from RNA-Seq data into genomic selection. Our method uses simulated data to develop whole transcriptome value index (WTVI) to rank animals, which will be tested on real data to ascertain response to genomic selection based on exome sequences captured in the WTVI. Our integrated algorithm exploits two main features in these data, that is, SNP genotyping from direct alignment of sequences located within functional genes, and gene expression quantification from the number of reads. These would be mapped to gene regulatory networks by selecting SNPs located in more important nodes of such networks that will likely elicit bigger phenotypic effects. We will assess efficiency, accuracy, cost effectiveness and simplicity of computational implementation of whole transcriptomic selection using these SNPs in cattle. Our rationale is that the use of SNPs located in transcripts and/or likely expressed functional regions may potentially account for higher genetic gain with less potential errors, thereby increasing the reliability of genomic selection.