MAGSplitter

Script downstream of a community MetaPathways run, to split results by metagenomic bin. Used for metagenomic bin-specific pathway inference from Pathway Tools.

Background Information

---

Currently,the Metapathways pipeline outputs are created in the form of ePGDBs readable for pathway tools. In other words, the outputs of metapathways signify entire metagenomes, rather than individual MAGs. If the end user were to want to look at metabolic pathway information for a single MAG, they would need to redo the metapathways pipeline for each single MAG. This is inefficient in terms of time and resources. Therefore, this tool alleviates this inefficiency by taking the outputs from the WGS pipeline and the metapathways pipeline, and splitting the metabolic ORF information into individual MAGs.

However, currently the file for ORF metabolic reactions within the ePGDB file has duplicate reactions previously removed. This is because the metapathways pipeline automatically removes duplicate pathways since frequency information is not utilized in pathway tools. Therefore, a step to bring these ORFs are required beforehand.

Once the removed ORF reactions are added to our ePGDB file, the ORFs are then mapped to their respective metagenome contigs and MAGs. Following, the reactions are split into individual MAGs, and new 0.pf files are created for each MAG. Unbinned ORFs are placed into a directory named "non_binned_metagenome"

Approach

---

• The ORF metabolic information outputted by metapathways is converted from PGDB format to a dataframe
• A list of missing ORFs are used to de-remove the removal of ORFs from the ePGDB file.
• A mapping file between ORFs and metagenome contigs are converted into a dataframe
• A mapping file between metagenome contigs and MAGs are converted into a dataframe
• Mapping files are then used to group the ORFs by MAGs.

Inputs

---

• 0.pf: pathway tools initial input from metapathways outputs (found in /ptools/)

• <sample_name>.ORF_annotation_tabel.txt: map of ORFS and contigs for an environment metagenome, outputted by metapathways (found in /results/annotation_table/)

• <sample_name>.mapping.txt: map of original input contigs IDs to internal contig IDs for an environment metagenome, outputted by metapathways (found in /preprocessed/)

• orf_map.txt: map of duplicated ORFS within a metagenome, with the first ORF being the intact ORF, and the rest being the duplicates (found in /results/annotation_table/)

• config_info.tsv: map of contigs to MAGs created through the WGS pipeline binning process.
  (found in /binning/results/greedy)

Usage

---

MAGSplitter is meant to be used after the WGS and the metapathways pipeline. There are two options to run MAGsplitter

Option 1:

To use this script pass the file locations for the input files as arguments. A "results" folder will be created in the current working directory

options:
  -h, --help            show this help message and exit
  -p PF_FILE, --pf_file PF_FILE
                        metapathways ePGDB output file location (typically
                        ptools/0.pf)
  -r ORF_MAPPING, --orf_mapping ORF_MAPPING
                        metapathways duplicate ORF mapping file location
                        (typically ptools/orf_map.txt)
  -c ORF_CONTIG_MAP_FILE, --orf_contig_map_file ORF_CONTIG_MAP_FILE
                        metapathways orf to contig mapping file location
                        (typically results/annotation_table/<samplename>.ORF_anno
                        tation_table.txt)
  -m CONTIG_MAG_MAP, --contig_mag_map CONTIG_MAG_MAP
                        wgs pipeline contig contig to mag mapping file location
                        (typically binning/results/greedy/config_info.tsv
  -i CONTIG_CONTIG_MAP_FILE, --contig_contig_map_file CONTIG_CONTIG_MAP_FILE
                        metapathways contig to contig mapping file location
                        (typically preprocessed/<samplename>.mapping.txt

example usage:
python ./magsplitter/main.py -p example/0.pf -r example/orf_map.txt -c example/GAPP-5498e568-6918-4000-b27e-dbeff35eeee7.ORF_annotation_table.txt -m example/contig_info.tsv  -i example/GAPP-5498e568-6918-4000-b27e-dbeff35eeee7.mapping.txt

Option 2:

You can also run this as a full package. First install using pip install:

pip install -e .

Then run with a command line entry point to main, with the same argument options as listed above.

example usage:
magsplitter ./magsplitter/main.py -pf example/0.pf -orf example/orf_map.txt -contig example/GAPP-5498e568-6918-4000-b27e-dbeff35eeee7.ORF_annotation_table.txt -mag example/contig_info.tsv  -i example/GAPP-5498e568-6918-4000-b27e-dbeff35eeee7.mapping.txt

Mock community run

We can also call the program to run on a mock metagenome dataset by not supplying arguments to the program This can be called with either

python ./magsplitter/main.py

or if the package is installed:

magsplitter