Africa GENetic Evaluation Service (AGENES): A pipeline for routine genetic and genomic evaluations

Genomic selection has significantly transformed livestock breeding programs, particularly in developed countries where genetic evaluation and selection systems are well-established for dairy, beef, and sheep. However, such advancements have yet to be widely adopted in the Global South, largely due to: lack of appropriate pipelines for routine genetic and genomic evaluations adapted to local production systems, the absence of standardised data collection, lack of expertise in developing theoretical and statistical frameworks for data management, and practical challenges in linking genetic variation to desirable traits. Consequently, livestock improvement has relied on conventional breeding practices, often using imported genetic material that is not well-suited to local conditions and lacks proper on-farm breeding infrastructure and inputs. To date, no comprehensive genetic and genomic evaluation pipeline has been fully developed or implemented in developing countries. Recently, initiatives like the Africa Dairy Genetic Gain (ADGG) program, led by the International Livestock Research Institute (ILRI) in collaboration with the Centre for Tropical Livestock Genetics and Health (CTLGH) in Edinburgh, UK, have made significant strides in addressing these challenges. ADGG has established a robust data collection platform necessitating the integration of genetic evaluation pipelines for routine evaluation. As part of a peer-to-peer collaboration in Africa, the Edinburgh Genetic Evaluation Service (EGENES) team at SRUC, in partnership with ILRI and CTLGH, has developed genetic and genomic evaluation pipeline for dairy cattle. It focuses on selected milk production and fertility traits while leveraging performance and pedigree data collected through the ADGG data platform. These data are sourced typically from registered smallholder farmers, research institutes, breeders, and AI providers across various African countries. Genomic data for bulls and cows are generated by ADGG, curated for quality assurance and integrated with the platform. Data providers upload their datasets to a centralized proprietary cloud platform connected to a dedicated computing system for processing and analysis. Access to the cloud platform is managed through role-based permissions, ensuring tiered data governance and usage. Once consolidated, quality control measures are applied, and the genetic evaluation is conducted by researchers using dedicated compute locally. The results are then made available back to the ownCloud platform, enabling users, including smallholder farmers, to access the results and make informed breeding decisions and identify top-performing bulls and cows. The current paper is a description of a streamlined, all-in-one cloud-native genetic evaluation pipeline that has been established, delivering actionable insights to users. In summary, the pipeline is a significant forward step in enhancing livestock breeding programs in Africa and beyond.