DoAggregate: a pipeline for aggregation testing for rare-variant on a gene and pathway level

Motivation Aggregation tests overcome the classical limitation of rare variant analysis in cohort study design. Choosing the right statistical test remains one of the biggest challenges. Existing tools fall short in offering a comprehensive solution that includes automatic data processing, testing across all genes, as well as exploration of broader genetic regions such as pathways and Gene Ontology (GO) terms. Most aggregation tests only report p-values and basic statistics, lacking in the presentation of cohort level statistics and comparison of results across whole genome or exome scales. Results We introduce DoAggregate, an innovative, scalable pipeline designed to address these challenges. DoAggregate enables users to perform aggregation tests across genes, pathways, and GO terms. It leverages Excalibur, an ensemble approach combining 36 distinct aggregation methods. The pipeline enhances data analysis by presenting over-representation analysis and aggregation results in detailed Excel sheets, complete with annotations and plots to facilitate nuanced interpretation and comparison. Applied to public exome data from the 1000 Genomes Project, DoAggregate tested 10,447 genes, 49 over-represented pathways and 36 over-represented GO terms in a single run, returning annotated, cohort-level results for each region. Availability and implementation DoAggregate source code, documentation and installation instructions as well as test data sets and results are freely available at https://github.com/boutrys/DoAggregate and released under the MIT license. The repository contains the serial implementation (R package), the parallel implementation for Slurm-based computer clusters and the web application.