Oral Presentation GENEMAPPERS 2026

Genetic regulation of isoform expression at single-cell resolution using long-read sequencing (#2)

Therese Johansson 1 2 , Walter Muskovic 2 , Agota Tuzesi 2 , Eszter Sagi-Zsigmond 2 , Scarlett Marx 2 , Rachael Zekanovic 2 , Joseph Powell 2 3 4
  1. Gene technology, KTH Royal Institute of Technology, Stockholm, Sweden
  2. Garvan-Weizmann Centre for Cellular Genomics, Garvan Institute of Medical Research, Sydney, New South Wales, Australia
  3. Faculty of Medicine and Health, University of New South Wales, Sydney, NSW, Australia
  4. UNSW Cellular Genomics Futures Institute, University of New South Wales, Sydney, NSW, Australia

Single-cell expression quantitative trait loci (sc-eQTL) studies have advanced our understanding of how genetic variation influences gene expression in a cell-type-specific manner. However, to date, these have focused on gene-level signals for each cell. Alternative splicing, which generates multiple RNA isoforms from a single gene, is one such mechanism that can further modulate RNA products and thus protein sequences. Genetic effects on splicing (isoform-specific expression) can thus offer critical insights into genotype–phenotype relationships by revealing regulatory links invisible to standard gene-level eQTL analyses. 

As part of the TenK10K project, we generated long-read single-cell RNA-seq data from >400,000 immune cells from 372 individuals, along with matched whole genome sequencing and short-read single-cell RNA data. Through the implementation of a novel splice-QTL method, developed on top of SAIGE-QTL, we were able to map isoform- and splice-specific eQTL across 28 cell types and transient cell states. We identified a total of 4,515 isoform-specific eQTLs, of which 38 showed isoform-specific regulatory effects in opposite directions, underscoring a complexity missed by gene-level analyses. For instance, we identified a cell-type specific sQTL in the PMF1 gene, where the variant 1:156236330:G>A in central memory CD4+ T cells drives opposing expression of two isoforms: upregulation of ENST00000368279 and downregulation of the canonical ENST00000368277. These isoforms encode structurally distinct proteins, the latter producing a full-length transcriptional coactivator, while the former results in a truncated version lacking key functional domains. This splicing event has been previously implicated in COVID-19 susceptibility, pointing to relevance in immune-related diseases. Notably, this regulatory effect is not detectable at the gene level, underscoring the value of isoform-resolved QTL mapping. This work underscores the value of long-read transcriptomic profiling in capturing the full complexity of genetic regulation, ultimately enhancing our understanding of how genetic variation shapes complex human traits.