Genomic selection is a recent breeding method that involves the prediction of phenotypic performance of selection candidates from observed variations in their DNA code. This ability to identify superior accessions from DNA has catalyzed the fast-paced adoption of genomic selection in most plant breeding programs. Genotype imputation is a term used for predicting missing values in genotyping results. It is an essential component of the data processing pipeline that precedes genomic selection.

The goal of the PREDIMP project is to develop, benchmark and showcase a unified computational framework for genotype imputation and genomic prediction. This integrated approach allows to substantially reduce genotyping costs whilst preserving and possibly improving the genomic prediction accuracy. The resulting increase in selection efficiency and overall simplification of the analysis pipeline are two major advantages that will set a new standard for genomic selection in plant breeding.