EPIK: Precise and scalable evolutionary placement with informative k-mers

N. Romashchenko, B. Linard, F. Pardi, E. Rivals.
Bioinformatics, in press
DOI: 10.1093/bioinformatics/btad692
EPIK is a program dedicated to "Phylogenetic Placement" (PP) of metagenomic or metabarcoding reads on a reference tree.

It is similar in spirit and technically the successor of RAPPAS (Linard et al. 2020). EPIK achieves identical or slightly better accuracy than RAPPAS and outperforms it in speed and flexibility. In many aspects the documentation of RAPPAS remains valid.

EPIK takes as input a file containing database of phylo-k-mers built with IPK and a set of reads to place on the related phylogeny. It works for nucleotidic and amino-acid query sequences.
EPIK can filter the database to load only the most informative phylo-k-mers, which reduces the memory usage.
EPIK can also run in parallel.

Keywords

Phylogenetic placement, metabarcoding, taxonomic identification, NGS, software
ipk-epik_overview

The workflow combining IPK, to precompute the index, and then EPIK, to perform the placement of the reads.