Beschreibung:
<jats:title>Abstract</jats:title><jats:sec>
<jats:title>Background</jats:title>
<jats:p>Current sequencing technologies have provided for a more comprehensive genome-wide assessment and have increased genotyping accuracy of rare variants. Scan statistic approaches have previously been adapted to genetic sequencing data. Unlike currently-employed association tests, scan-statistic-based approaches can both localize clusters of disease-related variants and, subsequently, examine the phenotype association within the resulting cluster. In this study, we present a novel Quantitative Phenotype Scan Statistic (QPSS) that extends an approach for dichotomous phenotypes to continuous outcomes in order to identify genomic regions where rare quantitative-phenotype-associated variants cluster.</jats:p>
</jats:sec><jats:sec>
<jats:title>Results</jats:title>
<jats:p>We demonstrate the performance and practicality of QPSS with extensive simulations and an application to a whole-genome sequencing (WGS) study of cerebrospinal fluid (CSF) biomarkers from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Using QPSS, we identify regions of rare variant enrichment associated with levels of AD-related proteins, CSF Aβ<jats:sub>1–42</jats:sub> and p-tau<jats:sub>181P</jats:sub>.</jats:p>
</jats:sec><jats:sec>
<jats:title>Conclusions</jats:title>
<jats:p>QPSS is implemented under the assumption that causal variants within a window have the same direction of effect. Typical self-contained tests employ a null hypothesis of no association between the target variant set and the phenotype. Therefore, an advantage of the proposed competitive test is that it is possible to refine a known region of interest to localize disease-associated clusters. The definition of clusters can be easily adapted based on variant function or annotation.</jats:p>
</jats:sec>