COWADAPT: genetic variants for cattle adaptability to harsh environments uncovered through a bovine multi-assembly graph

Abstract

The prevalence of large structural variations in the cattle genome as well as their sequence content and biological relevance remain largely unknown. In order to make this type of genetic variation amenable to genetic investigations, COWADAPT will construct a multi-assembly graph that integrates reference-quality assemblies from underrepresented breeds of indicine and taurine cattle. Using the multi-assembly graph, we will discover structural variants and investigate their sequence content to reveal so far neglected coding sequences. We will test if these previously unused variants are associated with the adaptability of indicine cattle to tropical environments. Our study will enable us to investigate the full spectrum of sequence diversity from a bovine multi-assembly graph. COWADAPT main objectives are: 1) obtain de novo assemblies of reference-quality genomes from taurine and indicine cattle breeds; 2) establish the pangenome of taurine and indicine cattle through multi-assembly graph; 3) detect and characterize structural variation in the genome of indicine cattle; 4) investigate the role of structural variation in cattle adaptation to tropical conditions. In order to address the research objectives of COWADAPT, we compiled four Work Packages (WP). In WP1, we will obtain long DNA fragments for de novo assembly of haplotype-resolved genomes from different breeds of taurine (n=4) and indicine cattle (n=3). To provide an overview of the prevalence of structural variants in an indicine cattle breed, we will sequence the genomes of n=20 prioritized Nellore cattle with long sequencing reads. The haplotype-resolved genome assemblies will be integrated into a multi-assembly graph in WP2. The analysis of the sequence content of the multi-assembly graph will reveal sequences that are private to indicine cattle. WP2 will also establish a framework for sequence variant analysis from multi-assembly graphs. Structural variants in the genome of indicine cattle will be screened and characterized in WP3. Already available short sequencing reads of Nellore cattle will be mapped against a Nellore-augmented reference genome, aiming to identify structural variations that are not accessible from the standard reference genome. Then, long sequencing reads from Nellore cattle (from WP1) will be aligned to the multi-assembly graph to compile a reference panel to accurately genotype structural variants in short-read sequenced cattle. The Nellore aligned long-read sequences will be used as a reference to impute genotypes for large structural variants into the larger cohort of short-read sequenced samples, and subsequently, into a mapping cohort of > 20,000 genotyped Nellore cattle that also have phenotypes for traits relevant for adaptation. In WP4, we will use the mapping cohort to detect structural variants associated with the adaptability of indicine cattle to harsh environmental conditions. In conclusion, COWADAPT will exploit a multi-assembly graph to unravel variants associated with adaptability to harsh environments. (AU)