PgmNr M291: Allelic imbalance is a prevalent and tissue-specific feature of autosomal and X-linked genes in F1 hybrid mice.

Authors:
S. F. Pinter 1,2,3,7 ; D. Colognori 2,3,4,7 ; B. J. Beliveau 3 ; R. I. Sadreyev 2,3 ; B. Payer 2,3,5 ; E. Yildirim 2,3,6 ; C. T. Wu 3 ; J. T. Lee 2,3,4,8


Institutes
1) UConn Health, Institute for Systems Genomics, Farmington, CT; 2) Massachusetts General Hospital, Boston, MA; 3) Harvard Medical School, Boston, MA; 4) Howard Hughes Medical Institute, Chevy Chase, MD; 5) Centre for Genomic Regulation, Barcelona, Spain; 6) Duke University, School of Medicine, Durham, NC; 7) co-first authors; 8) corresponding author.


Abstract:

In mammals, several classes of monoallelic genes have been identified, including those subject to X-chromosome inactivation (XCI), genomic imprinting, and random monoallelic expression (RMAE). However, the extent to which these epigenetic phenomena are influenced by underlying genetic variation is unknown. We performed a systematic classification of allelic imbalance in mouse hybrids derived from reciprocal crosses of divergent strains. We observe that deviation from balanced biallelic expression is common, occurring in ∼20% of the mouse transcriptome in a given tissue. Allelic imbalance attributed to genotypic variation is by far the most prevalent class and typically is tissue-specific. However, some genotype-based imbalance is maintained across tissues and is associated with greater genetic variation, especially in 5′ and 3′ termini of transcripts. We further identify novel random monoallelic and imprinted genes and find that genotype can modify penetrance of parental origin even in the setting of large imprinted regions. Examination of nascent transcripts in single cells from inbred parental strains reveals that genes showing genotype-based imbalance in hybrids can also exhibit monoallelic expression in isogenic backgrounds. This surprising observation may suggest a competition between alleles and/or reflect the combined impact of cis- and trans-acting variation on expression of a given gene. We also identify tissue-specific genes that escape XCI, and share some surprising and speculative insights from integrating these allele-specific expression results with other epigenomic data.