We have clear and consistent evidence that the presence of a host gene, namely KIR2DL2, enhances both protective and detrimental HLA class I associations in two unrelated chronic viral infections. What is much less clear is the underlying mechanism of action. Epistatic HLA-KIR associations are typically attributed to direct NK-mediated lysis of virus-infected cells. We believe direct NK killing is not compatible with our observations and instead hypothesise that the downstream effectors of KIR2DL2-dependent enhancement are CD8+ T cells. We wish to test this hypothesis and to explore the generality of this KIR2DL2-enhancement by investigating whether KIR2DL2 also impacts on the probability of developing virus-associated malignancy. Importantly, we suggest that KIR2DL2 consistently enhances the efficiency of the CD8+ T cell response; this does not imply that KIR2DL2 is consistently beneficial to human health as KIR2DL2 also enhances the efficiency of CD8+ T cells restricted by detrimental HLA alleles. Approach. We will take an integrated theoretical and experimental approach. We will study 3 patient cohorts. 1) HCV: 60 HCV-HIV coinfected individuals with acute HCV. 2) HTLV-1 (UK): 30 individuals chronically infected with HTLV-1. 3) HTLV-1 (Japan): 240 adult T cell leukemia/lymphoma (ATLL) cases and 200 HTLV-I-infected healthy controls. To compare the strength of the CD8+ T cell response between KIR2DL2+ and KIR2DL2- individuals we will use next generation sequencing of the HCV genome (Illumina HiSeq), mathematical modelling of host-pathogen dynamics including parameter inference and model selection, ex vivo CTL and NK killing assays and HLA class I epitope prediction. To assess the impact of KIR2DL2 in adult T cell leukemia/lymphoma we will use HLA and KIR genotyping, epitope prediction, logistic regression and Monte Carlo methods in a case-control cohort.