Faba bean populations already contain the inbreds needed for breeding

Abstract Key message Highly inbred individuals are naturally present in faba bean populations and should be used for rapid and cost-efficient development of line and synthetic cultivars. Abstract Producing inbred lines in faba bean ( Vicia faba L.) breeding is, although crucial, a long, costly and laborious process due to the partial allogamy of the crop. However, in any genetically diverse and open pollinating faba bean population, this particular reproduction system leads to the so-called identity disequilibrium, resulting in a significant proportion of individuals being highly inbred. This study presents a cost-efficient approach to identify these highly inbred individuals from populations with a limited number of KASP markers. Using simulations, the most accurate inbreeding coefficient estimator from limited number of markers (ten to 100) was found to be the so-called estimator F L (which is based on a maximum likelihood approach). In the “Göttingen Winter Bean Population”, 8.4% of individuals (61 out of 728) were classified as highly inbred (F L  ≥  $$0.8\overline{3 }$$ 0.8 3 ¯ ). We evaluated the performance of classifying highly inbred individuals using a limited number of markers with both simulated and empirical data, allowing to identify strategies for optimizing this approach. Using 50 markers and a threshold of F L  ≥ 0.9, we identified 42 individuals in our population, including 36 truly highly inbred individuals (out of 61) and six insufficiently inbred individuals (false positives). Furthermore, applying a preliminary step with a low number of markers to eliminate insufficiently inbred individuals successfully permitted a marked reduction in genotyping cost (up to 46%). Faba bean breeders can greatly benefit from the presented approach by identifying highly inbred individuals fast, at low cost and with limited labour, even though no DH technology is available.