Prediction of optimal crosses based on haplotype-based QTL-allele systems for simultaneous improvement of 100-kernel weight and 100-pod weight in peanut

The 100-kernel weight (HKW) and 100-pod weight (HPW) are pivotal yield components with complex genetic regulation in cultivated peanut ( Arachis hypogaea L.), which directly determine peanut yield level and commercial value. Exploring the genetic basis and favorable allele resources underlying HKW and HPW is essential for molecular improvement and high-yield breeding of peanut varieties. Based on high-quality SNP data from 353 global peanut core germplasm accessions, we constructed 51,669 SNP linkage disequilibrium blocks and systematically dissected the QTL-allele systems underlying HKW and HPW using a multi-locus genome-wide association model. A total of 29 main-effect QTLs containing 192 alleles were identified for HKW, which were distributed on 13 chromosomes and collectively explained 84.55% of the phenotypic variation, with the largest-effect QTL located on chromosome Arahy.16 ( R 2 = 36.95%) and chromosome Arahy.12 harboring the largest number of QTLs (seven). For HPW, 28 QTLs with 208 alleles were detected on 13 chromosomes, accounting for 85.85% of the phenotypic variation, with QTL clusters on chromosomes Arahy.12, Arahy.14 and Arahy.19 (four QTLs each) and the largest-effect QTL located on chromosome Arahy.14 ( R 2 = 18.67%). Notably, a shared pleiotropic QTL on chromosome Arahy.05 was detected for both traits, explaining 29.58% of the phenotypic variation for HKW and 15.63% for HPW. Based on the established QTL-allele matrices, three optimal cross selection strategies (HKW-priority, HPW-priority, and HKW-HPW-balanced), were applied to screen elite crosses for simultaneous improvement of both traits, yielding 30 optimal crosses whose predicted recombination potentials reached 162.35–170.21 g for HKW (HKW-priority), 422.17–432.21 g for HPW (HPW-priority), and 136.00–154.03 g for HKW together with 371.09–413.21 g for HPW (balanced strategy), substantially exceeding the best observed values in the population for the corresponding trait(s). This study systematically clarifies the QTL-allele genetic architecture of peanut HKW and HPW. The QTL-allele matrices, the elite-allele donor accessions, and the predicted optimal crosses together constitute valuable genetic, allele, and germplasm resources, and the “QTL-allele system, in silico progeny simulation, and multi-strategy cross selection” pipeline demonstrated here provides an actionable technical route and theoretical references for molecular design breeding and high-yield genetic improvement in peanut.