Please use the latest version of the Snakemake pipeline maintained by GIS RPD team.
This pipeline is based on Bpipe, and is now routinely used by the group to process (pair-end) shotgun metagenomics samples.
- Bpipe 0.9.9 beta 1 (sge_fixes branch) or Snakemake 3.7.1
- MetaPhlAn 1.7.7 (http://huttenhower.sph.harvard.edu/metaphlan)
- Kraken 0.10.5 beta (https://ccb.jhu.edu/software/kraken/)
- BWA 0.7.10 (http://bio-bwa.sourceforge.net/)
- famas 0.0.10 (https://github.com/andreas-wilm/famas.git)
- HUMAnN 2 (https://bitbucket.org/biobakery/humann2/wiki/Home)
- Customized scipts:
- split_metaphlan_table.sh
- merge_tables.py (CSB5/misc_scripts/merge_tables.py)
- decont.py (CSB5/decont)
- Using snakemake
$ snakemake_pipeline/create_config.sh FILE [FILE2 FILE3 ...]
$ snakemake_pipeline/run_pipeline.sh
Note: You can modify the paths variables in the conf.yaml created by "create_config.sh".
- Using bpipe (this is no longer in use)
$ bpipe test -n200 run_metagenomics_pipeline.bpipe Sample_Folder1 <Sample_Folder2 Sample_Folder3 ...>
$ bpipe run -n200 run_metagenomics_pipeline.bpipe Sample_Folder1 <Sample_Folder2 Sample_Folder3 ...>