Mammalian Genome Genetics

Conserved SINE and lineage-variable LINE-1 depletion at neurodevelopmental disorder promoters

Sevilmiş, Can
Springer Nature · 2026-10-22 · DOI 10.1007/s00335-026-10285-6

Abstract

Developmental genes lie in transposon-poor regions, but whether that deficit is a conserved property of the promoters themselves remains unmeasured. Transposable element frequency was compared at neurodevelopmental disorder (NDD) promoters (TSS ±2 kb) in seven primates and two outgroups, contrasting a curated NDD gene set (SFARI and ClinGen; $$n=1020$$ ) with brain-expressed housekeeping genes ( $$n=1679$$ ). Alu depletion was significant in every primate ( $$r = -0.147$$ to $$-0.384$$ ), including the mouse lemur at 70 Mya. Mouse B1 and B2, SINEs amplified independently of primate Alu, each showed the deficit; the AT-biased Can-SINEs of dog did not. LINE-1 depletion was weak against housekeeping promoters but comparable to or deeper than the SINEs against the genome baseline, and more lineage-variable. The signal survived permutation testing, CpG and GC stratification, and gene-density and recombination controls. It persisted for alignment-defined rather than annotation-defined promoter windows, and against each genome’s own baseline in seven of nine, marmoset and dog excepted; the squirrel monkey showed no such exception. Against human controls matched on constraint, brain expression and GC it narrowed but persisted ( $$r = -0.106$$ ); Alu density does not otherwise track constraint. Windows away from the start site separate two components: the surrounding 250 kb is itself Alu-poor, and a promoter-proximal deficit survives normalisation to it. Depleted promoters coincided with active cis-regulatory elements, preferentially fetal-brain ones. Low promoter SINE content is thus an old feature of NDD-associated loci; whether the elements were never inserted or inserted and later lost is unresolved.

Read at publisher →