FUSION

Summary

FUSION (Functional Summary-based Imputation) is a software suite designed to perform transcriptome-wide and regulome-wide association studies (TWAS/RWAS) using functional expression models and GWAS summary statistics. It builds predictive models of the genetic component of expression or other molecular phenotypes using reference panels (e.g., GTEx or METSIM) and evaluates their association with complex traits. FUSION computes gene-level associations by imputing functional traits into GWAS cohorts, helping to identify potential causal genes and mechanisms behind genetic associations.

Methodology and Use Cases

FUSION operates by combining genotype-expression training data with large-scale GWAS summary statistics: - Expression Imputation: It trains multiple models (such as Elastic Net, Lasso, and Best Linear Unbiased Predictor/BLUP) to predict gene expression from cis-SNPs, and selects the model with the highest cross-validation accuracy. - Handling Missing SNPs: When SNPs are present in the eQTL prediction models but absent in the target GWAS, FUSION uses the ImpG-summary algorithm to impute the missing GWAS summary values, whereas other tools like sPrediXcan remove them. - Benchmark Findings: When evaluated for TWAS signature-matching for drug candidate prioritisation, FUSION led to a weaker, non-significant enrichment of first-line treatments (e.g., HMGCR inhibitors for LDL cholesterol) compared to sPrediXcan, potentially due to the deleterious impact of imputing missing GWAS summary values.

Citations

• Gusev, A., Ko, A., Shi, H., Bhatia, G., Chung, W., Penninx, B. W., ... & Pasaniuc, B. (2016). Integrative approaches for large-scale transcriptome-wide association studies. Nature Genetics, 48(3), 245–252.
• Chauquet, S., Jiang, J. C., Barker, L. F., Hunter, Z. L., Singh, G., Wray, N. R., McRae, A. F., & Shah, S. (2026). From GWAS to drug: A framework for drug candidate prioritisation using a gene expression signature matching approach. medRxiv. Source paper: 2026.04.22.26349470v2.full.pdf