We provide a unified framework that combines total and allele-specific (at the gene level) read counts and is scalable to large studies with thousands of samples, such as GTEx.

We develop computationally efficient methods that integrate allele-specific and total read counts to improve cisQTL discovery, fine-mapping, and genetic prediction of the transcriptome. Our log-linear approximation allows scaling-up of computations well-beyond a few hundreds of samples, providing an alternative to current total read count-based methods.

This repository hosts the R implementation of mixQTL, mixFine, and mixPred.

The manuscript describing the details of the approach can be found here

Find detailed instructions on how to run mixQTL here via the command line tool or R functions. A computationally efficient GPU-based implementation of mixQTL is embedded in the tensorQTL software here