Genetics

Common meiosis-gene variants shape recombination and maternal aneuploidy risk

Retrospective analysis of 139,416 IVF embryos mapped 3,809,412 crossovers and 92,485 aneuploid chromosomes from PGT data. Aneuploid embryos had fewer crossovers than euploid embryos, underscoring the importance of recombination for proper chromosome segregation. A common SMC1B haplotype was associated with reduced crossover counts and increased maternal meiotic aneuploidy, likely via cis-regulatory effects. Variants in C14orf39, CCNB1IP1, and RNF212 were also implicated in meiotic aneuploidy and recombination phenotypes. Several aneuploidy-associated variants showed secondary associations with reproductive ageing traits, suggesting shared genetic control of fertility timing and chromosomal stability. 1

Functional saturation screens clarify PALB2 missense variant breast cancer risk

Site-saturation functional screens tested 84% of all possible missense variants across 11 PALB2 exons, using PARP inhibitor sensitivity as a homologous recombination readout. The targeted exons encode the coiled-coil and WD40 domains, which were identified as the minimal regions required for homologous recombination. Among 6,718 missense variants, 58% were classified functional, 36% intermediate, and 6% damaging by the assay. Burden analysis showed that carriers of damaging PALB2 missense variants had significantly increased breast cancer risk, similar to truncating variants. These data support reclassification of many PALB2 missense VUS and more confident risk management for carriers. 2

Schizophrenia risk loci refined using ancestrally diverse, especially African-ancestry, cohorts

Genomic data from the Million Veteran Program linked to electronic health records enabled large-scale schizophrenia genetics in African-ancestry populations. The study identified ancestry-independent associations in African-ancestry cohorts and expanded the schizophrenia risk locus catalogue by more than 100 regions. Statistical fine-mapping and integrative transcriptomic analyses refined signals to consensus genes with convergent neurobiological functions. Findings show substantial shared polygenic architecture across ancestries but highlight limitations of Eurocentric discovery datasets for global prediction. Results argue for routine inclusion of diverse ancestries to improve biological insight and equity in psychiatric genetic tools. 3