Large-scale evaluation characterises the 'colocalisation gap' between GWAS loci and eQTL datasets

A study published in PLOS Genetics (8 May 2026) by Guillermo Reales, Jeffrey M. Pullin, and colleagues at the MRC Biostatistics Unit (Chris Wallace's group) provides a systematic characterisation of why eQTL–GWAS colocalisation analyses leave a large proportion of association signals unexplained — a phenomenon the authors term the 'colocalisation gap'.

Colocalisation analysis is a widely used bioinformatic approach that integrates GWAS results with expression quantitative trait loci (eQTL) data to identify candidate effector genes at disease-associated loci. The study analysed over 1.3 million colocalisation tests from the Open Targets Genetics resource and an additional large-scale evaluation, examining what proportion of GWAS signals colocalise with an eQTL in a given tissue and identifying the factors — including tissue specificity, cell-type composition, sample size, and linkage disequilibrium structure — that explain variation in colocalisation success rates.

The authors conclude that a substantial fraction of GWAS loci act through mechanisms not captured by bulk tissue eQTL datasets, including cell-type-specific regulation, splicing QTLs, and effects visible only in specific developmental contexts or environmental conditions. The paper provides practical guidance for researchers designing future eQTL studies and for those interpreting negative colocalisation results. It is directly relevant to statistical geneticists, functional genomicists, and those working on translating GWAS findings into mechanistic understanding.