Back

Expanded map of genomic imprinting reveals insight into human disease

Smail, C.; Cheung, W. A.; Koseva, B.; Johnson, A. F.; Bi, C.; Schreck, C. F.; Lydic, M.; Holoch, K.; Repnikova, E.; Herriges, J.; Marsh, C.; Thiffault, I.; Pastinen, T.; Grundberg, E.

2025-09-16 genetic and genomic medicine
10.1101/2025.09.15.25335770 medRxiv
Show abstract

Genomic imprinting involves parent-of-origin effect (POE) of regulatory element activity, often measured through methylation of CpG (5-mC) dinucleotides. While a dozen clinical syndromes are linked to defective imprinting, the extent this epigenetic phenomenon is linked to phenotypic variation and disease susceptibility remains undetermined. We show long-read HiFi genome sequencing for single-molecular profiling of 5-mC, together with pedigree-based phasing in early developmental tissue, provides critical insight into previously uncharted loci in the human genome. Using this approach in 75 samples from 25 trios, we develop a 10-fold enhanced map of human imprinting during development. The majority of POE was maternal (90%) with germ cell hypermethylation was confirmed at most loci (72%) showing signature of paternal imprinting. Integrating summary statistics from population GWAS finds enrichment of common (birthweight) and rare (congenital anomalies) disease loci in newly identified imprinting regions. Accessing pedigree-based rare disease cohorts, we show preponderance of paternal inheritance of pathogenic variants mapping to autosomal dominant OMIM genes with a maternal POE-bin identifying two genes (BNC2, DNMT1) as novel candidate imprinting disorder loci. Our enhanced human map of POE of 5-mC significantly extends the current "imprintome" and uncovers previously underappreciated genes and variants that appear crucial for human development and disease.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.