Whole genome sequencing to identify host genetic risk factors for severe outcomes of hepatitis a virus infection

Abstract

Acute liver failure is a severe, but rare, outcome of hepatitis A virus infection. Unusual presentations of prevalent infections have often been attributed to pathogen-specific immune deficits that exhibit Mendelian inheritance. Genome-wide resequencing of unrelated cases has proven to be a powerful approach for identifying highly penetrant risk alleles that underlie such syndromes. Rare mutations likely to affect protein expression or function can be identified from sequence data, and their association with a similarly rare phenotype rests on their existence in multiple affected individuals. A rare or novel sequence variant that is enriched to a significant degree in a genetically diverse cohort suggests a candidate susceptibility allele. Whole genome sequencing of ten individuals from ethnically diverse backgrounds with HAV-associated acute liver failure was performed. A set of rational filtering criteria was used to identify genetic variants that are rare in the population, but enriched in this cohort. Single nucleotide polymorphisms, insertions, and deletions were considered and autosomal dominant, autosomal recessive, and polygenic models were applied. Analysis of the protein-coding exome identified no single gene with putatively deleterious mutations shared by multiple individuals, arguing against a simple Mendelian model of inheritance. A number of rare variants were significantly enriched in this cohort, consistent with a complex and genetically heterogeneous trait. Several of the variants identified in this genome-wide study lie within genes important to hepatic pathophysiology and are candidate susceptibility alleles for hepatitis A virus infection.

Keywords: acute liver failure; genome-wide; hepatitis A; host genetics; immunity; viral hepatitis.

© 2014 Wiley Periodicals, Inc.

Figures

Figure 1

Noncoding variants shared among individuals by mode of inheritance. Shown are number of single nucleotide variants present at each allele frequency assuming either an autosomal recessive (top) or autosomal dominant (bottom) model of inheritance in each of the ten individuals sequenced. Grey bars indicate the number of variants identified using basic frequency cutoffs described in text and Supplementary Methods. Black bars indicate the number of variants remaining after filtering out those present in any of the three control genomes.