MetaPhlAn database that allows quantification of Bifidobacterium longum subspecies (Bifidobacterium longum subsp. infantis and Bifidobacterium longum subsp. longum) from metagenomic data.
This tool is based on the standard MetaPhlAn4 database. After running it MetaPhlAn4 can re ran as usual with the flags --index and --bowtie2db specifying the new custom database.
If you use this tool please cite:
Ennis, D., Shmorak, S., Jantscher-Krenn, E., & Yassour, M. (2024). Longitudinal quantification of Bifidobacterium longum subsp. infantis reveals late colonization in the infant gut independent of maternal milk HMO composition. Nature Communications, 15(1), 894.
- Clone the repository using
git clone. - Make sure you have the following dependencies (excpet for click they are all included in the MetaPhlAn conda environment):
- bowtie2
- python3 packages: click, pickle, bz2, subprocess, os
To add B. longum subspecies specific markers to the metaphlan database run
metaphlan_longum_markers.py --mpa-db-directory <PATH TO METAPHLAN ORIGINAL DATABASE>
Optional arguments:
--db_name The name of the original database without a suffix. Default :'mpa_vOct22_CHOCOPhlAnSGB_202212'
--output The directory in which to save the new database. Default: current directory
Note that since the MetaPhlAn bowtie database is inspected and rebuilt, this process needs time and memory (~3 hours, 32g, 12 CPUs)