A genome-to-genome analysis of associations between human genetic variation, HIV-1 sequence diversity, and viral control

Abstract

HIV-1 sequence diversity is affected by selection pressures arising from host genomic factors. Using paired human and viral data from 1071 individuals, we ran >3000 genome-wide scans, testing for associations between host DNA polymorphisms, HIV-1 sequence variation and plasma viral load (VL), while considering human and viral population structure. We observed significant human SNP associations to a total of 48 HIV-1 amino acid variants (p<2.4 × 10(-12)). All associated SNPs mapped to the HLA class I region. Clinical relevance of host and pathogen variation was assessed using VL results. We identified two critical advantages to the use of viral variation for identifying host factors: (1) association signals are much stronger for HIV-1 sequence variants than VL, reflecting the 'intermediate phenotype' nature of viral variation; (2) association testing can be run without any clinical data. The proposed genome-to-genome approach highlights sites of genomic conflict and is a strategy generally applicable to studies of host-pathogen interaction. DOI:http://dx.doi.org/10.7554/eLife.01123.001.

Keywords: HIV; Human; human genomics; viral mutations.

Figure 1.. A triangle of association testing.

The following association analyses were performed: [Study A] human SNPs vs plasma viral load (1 GWAS); [Study B] human SNPs vs variable HIV-1 amino acids (3007 GWAS); and [Study C] variable HIV-1 amino acids vs plasma viral load (1 proteome-wide association study). DOI: http://dx.doi.org/10.7554/eLife.01123.003

Figure 2.. Results of the genome-wide association analyses.

(A) Associations between human SNPs and HIV-1 plasma viral load. The dotted line shows the Bonferroni-corrected significance threshold (p-value < 7.25 × 10⁻⁹). (B) Associations between human SNPs and HIV-1 amino acid variants, with 3007 GWAS collapsed in a single Manhattan plot. The dotted line shows the Bonferroni-corrected significance threshold (p-value < 2.4 × 10⁻¹²). (C) Schematic representation of the HLA class I genes and of the SNPs associated with HIV-1 amino acid variants in the region. (D) Same association results as in panel B, projected on the HIV-1 proteome. Only the strongest association is shown for each amino acid. Significant associations are indicated by a blue dot. The gp120 part of the HIV-1 proteome was not tested. The colored bar below the plot area shows the positions of the optimally defined CD8+ T cell epitopes. An interactive version of this figure can be found at http://g2g.labtelenti.org (which is also available to download from Zenodo, http://dx.doi.org/10.5281/zenodo.7138). DOI: http://dx.doi.org/10.7554/eLife.01123.004

Figure 3.. Association of HIV-1 amino acid variants with plasma viral load.

(A) Changes in VL (slope coefficients from the univariate regression model and standard error, log₁₀ copies/ml) for the 48 HIV-1 amino acids that are associated with host SNPs in the genome-to-genome analysis. (B) rs2395029, a marker of HLA-B*57:01 is associated with a 0.38 log₁₀ copies/ml lower VL (black bar) in comparison to the population mean. Gray bars represent changes in VL for amino acid variants associated with rs2395029 (p<0.001). In case of multiallelic positions, the change in VL is shown for all minor amino acids combined vs the major amino acid (e.g., GAG147 not I). DOI: http://dx.doi.org/10.7554/eLife.01123.006