In smallholder, low-input farming systems diffused in the Global South, farmers select and propagate crop varieties based on their traditional knowledge and experience. A quantitative integration of their knowledge into breeding pipelines may support the sustainable intensification of local farming. This data entry combines genomics with socioeconomics to tap into traditional knowledge in smallholder farming systems, focusing on durum wheat (Triticum durum Desf.). Data refer to a large nested association mapping (EtNAM) population that we developed by recombining elite international breeding line with Ethiopian traditional varieties maintained by local farmers. This entry carries also molecular and phenotypic data produced on a diversity panel (DP) of Ethiopian landraces previously characterized in four year-location combinations and published in Mengistu et al 2016 (https://doi.org/10.1111/pbi.12538). 

EtNAM lines and DP genotypes were evaluated for agronomic performances and farmers’ appreciation in multiple locations, reveailing that gender and location can influence farmers’ preference and that women and men farmers can consistently identify the best durum wheat genotypes. We used this data to train a genomic selection (GS) model with farmer scores to show that their prediction accuracy over grain yield was higher than that of the benchmark GS model trained on grain yield. The data was also used in a genome wide association mapping (GWAS) and quantitative trait locus (QTL) mapping to identify genetic determinants of agronomic traits and farmer scores.  Our data shows that farmers’ traditional knowledge can be integrated in a quantitative framework to increase genetic gain in pre-breeding programs, supporting genomics-driven breeding for local adaptation.

The Rdata files contain phenotypic and molecular characterization data for 1,200 recombinant inbred lines (RILs) deriving from the EtNAM and phenotypic and molecular characterization data for 400 durum wheat genotypes in the Ethiopian DP.

Phenotypic data includes agronomic traits and participatory variety selection data collected in muiltiple locations in Ethiopia. Molecular charatcterization data is based on the wheat SNP array provided by TraitGenetics. Data has been processed for basic QC. See README for details 

Data included in this entry shoud be analyzed with the open source program R (https://cran.r-project.org/)