Introduction
The West African Ebola epidemic was the largest Ebola epidemic to date with over 28,000 cases. The large number of cases permitted assessment of different disease phenotypes and outcomes of Ebola virus disease (EVD). Given the variety of disease phenotypes in EVD, a genetic predisposition to disease phenotype and outcome was hypothesised.

Methods
Samples from 325 deceased patients and 174 surviving patients were provided through the Sierra Leone Ministry of Health-Public Health England Ebola Biobank. Additionally, 1021 household contacts, 1004 community controls and 504 Ebola survivors were recruited in Sierra Leone. Participants provided a saliva sample for DNA extraction and an oral fluid sample for anti-EBOV IgG antibodies. Exome sequencing was undertaken on 250 extreme phenotype cases and genome wide genotyping was undertaken on 2153 Ebola patients, household contacts and community controls. Data analysis of the exome data included within family segregation studies, gene burden testing and pathway analysis. The genotyped data was interrogated through a genome wide association study comparing deceased and surviving cases.

Results
Of the household contacts, 3.5% were positive for anti-EBOV IgG. Seropositivity correlated with risk exposure level, with the highest risk level demonstrating seropositivity rates of 15.6%. Ebola survivors with more severe acute disease demonstrated lower levels of anti-EBOV IgG antibodies (p=0.01), as did those with more severe post-Ebola syndrome, although this was not significant. Exome sequencing revealed multiple protective mutations within cholesterol metabolism pathways. A key finding was a protective variant in the PCSK9 gene (p=0.002). Preliminary GWAS analysis of deceased versus surviving Ebola patients identified a genome wide significant (p=2.9x10-8) SNP in the Carbonic Anhydrase 5a gene.

Conclusions
The study established different extreme phenotypes of EVD, including highly exposed antibody negative and asymptomatic antibody positive individuals. Genetic factors affect both susceptibility to and severity of Ebola virus disease; with rare deleterious mutations in genes within cholesterol metabolism pathways, and common polymorphisms determining outcome of exposure to Ebola virus.