Preprint presents PEC algorithm for correcting pedigree errors using haplotype and SNP-chip data

A preprint on bioRxiv describes PEC (Pedigree Error Correction), a new algorithm for identifying and correcting errors in pedigree records by cross-referencing them with SNP-chip genotype data. Pedigree errors — where recorded parent–offspring or full-sibling relationships do not reflect true genetic relationships — are common in livestock breeding populations that rely on long-term manual record-keeping, and they reduce the accuracy of genomic estimated breeding values and the efficiency of selection programmes.

PEC uses linkage disequilibrium (LD) block structure and haplotype fragment matching between candidate parents and offspring to assess relationship likelihood, and then applies Mendelian conflict detection to flag and resolve inconsistencies. The authors report that PEC outperforms existing pedigree-correction tools on accuracy benchmarks while requiring substantially lower computational resources.

While the immediate application described is livestock genomics — specifically dairy cattle breeding programmes — the underlying methodological problem of reconciling recorded pedigrees against molecular marker data is directly relevant to human pedigree-modelling research and to genealogical genomics. The Evagene platform's pedigree-modelling work addresses analogous challenges. The preprint has not yet been peer reviewed.