Efficient Pangenome Construction through Alignment-Free Residue Pangenome Analysis (ARPA)

Protein sequences can be transformed into vectors composed of counts for each amino acid (vector of Residue Counts; vRC) that are mathematically tractable and retain information about homology. We use vRCs to perform alignment-free, residue-based, pangenome analysis (ARPA; https://github.com/Arnavlal/ARPA). ARPA is 70-90 times faster at identifying homologous gene clusters compared to standard techniques, and offers rapid calculation, visualization, and novel phylogenetic approaches for pangenomes.