Genetic control of the transcriptional response to active tuberculosis disease and treatment

The functional impact of genomic regulatory variation during active TB disease and anti-TB treatment (ATT) remains poorly understood. Here, using 240 paired whole-blood RNA-seq samples from n = 48 active TB patients who underwent ATT, we call and impute 1,506,948 genome-wide variants from these transcriptomes to map cis -expression quantitative trait loci (eQTL), response-eQTL, and cell type interaction eQTL (ieQTL) by computationally deconvolving the bulk RNA-seq data. We show that the whole-blood transcriptome is substantially perturbed during ATT by identifying 7,246 differentially expressed genes. We characterise 5,129 cis -eQTL and 587 response-eQTL and show that genes associated with reQTL are significantly enriched in pathways impacting the host response to mycobacterial challenge and xenobiotic metabolism. We highlight significant changes in cell type proportions due to ATT, notably for NK cells and classical monocytes. We characterise 921 ieQTL, including an active-TB classical monocyte-specific ieQTL for the gene NOD2 . Our work sheds light on the immunogenetics of TB disease and treatment, while providing a framework for integrative genomics studies using only RNA-seq data.