Identification of causal factors for recessive lethals in dairy cattle with special focus on large chromosomal deletions

Mesbah is originally from Bengladesh. He was selected as a PhD candidate by the ITN Marie Curie EGS-ABG program to do a PhD with Aarhus University and AgroParisTech. He therefore shared time between Foulum and Jouy-en-josas, co-advised by Goutam Sahana and Bernt Guldbrandtsen (Aarhus), Didier Boichard and Aurélien Capitan (GABI). His dissertation was entitled, "Identification of causal factors for recessive lethals in dairy cattle with special focus on large chromosomal deletions". His defense jury was presided by Just Jensen included four rapporteurs, two physically present, Claus Boettcher Jorgensen (univ Copenhague) and Goran Andersson (Uppsala University), and two who provided a written report only, Alessandra Stella (CNR Milan) and George Thaller (univ Kiel).

This thesis was focused on a complex trait for genetic improvement in cattle due to its low heritability but its strong variability, and its opposition with dairy production. His work was aimed at identifying causal genetic variants implicated in femal fertility determinism using different approaches.

A first study published in the Journal of Dairy Science describes a systematic approach to map recessive lethal variants in French Normandy cattle based on the detection of homozygous haplotype deficiency. The results shows that sample size, quality of genotype phasing into haplotypes, frequency and haplotype age, significance levels…) affect the efficacy of breeding schemes and that the approaches used for fine mapping, annotation between species of candidate variants and large-scale genotyping are important to validate or invalidate the initial mutations. A lethal mutation in the CAD (carbamoyl-phosphate synthetase 2, aspartate transcarbamylase, and dihydroorotase) gene that has a 3% frequency in the population was identified.

Another part of the dissertation examines deletion variants, which can induce deleterious effects. An article published in DNA Research, describes the high-throughput mapping of large chromosomal deletions in a population of 175 animals from three bovine breeds and the mechanisms that lead to the formation of deletions based on the characteristics of the DNA sequences at the break-points. This study provides insights for future work on deletions. Another paper published in the Journal of Dairy Science reports an original approach based on a Gaussian mixture model to estimate genotypes at a known deletion locus from read-depth data from the variant call format (VCF), which allows access to a larger population. It presents a pipeline for the joint imputation of SNP variants and large chromosomal deletions. Since deletions were identified or predicted in a large population, QTL for fertility could be determined. A GWAS study of this trait was performed in three bovine breeds using SNPs and large deletions, which revealed three new QTL and confirmed previously identified QTL.

 A pangenomic association study of fertility was performed on three bovine breeds using SNP and large deletions. This study presents several loci for quantitative traites and confirms others. Finally, the value of such new variants (deletions or SNPs in the QTL) for predicting the genomic values of fertility was tested. Our findings show that these two variants, in particular the deletions, have a high predictive power for improving the precision of prediction compared to neutral markers.