Gene network inference applied to beef cattle breeding and genetics

The advent of cheaply available high-throughput genetic and genomic techniques has equipped animal geneticists with an unprecedented ability to generate massive amounts of molecular data. As a result, large lists of genes differentially expressed in many experimental conditions of interests have been reported. Similarly, the association of an ever growing number of DNA variants with phenotypes of importance is now a routine endeavour. Inspired by this wealth of information, systems biology aims to formally integrate seemingly disparate datasets allowing for a holistic view of the system as a whole, where the key properties emerge in a natural fashion. Anchored in the power of gene network inference, this talk will present two examples of rigorous ways of integrating molecular data of relevance to beef cattle breeding and genetics. In both examples, we advocate the use of PCIT, a network inference algorithm that exploits the dual concepts of partial correlation and information theory and highly regarded in the recent literature. The first example in concerned with the onset of puberty in beef cattle. The second example merges phenotype, metabolomic and genomic data to reveal biological processes contributing to genetic variability in feed efficiency.