dcc2 is a tool for compiling core set data for h1s using predicted orthologs from OMA, both OMA-browser and OMA-standalone. Outputs of this tool are 3 (optional 4) folders required for a HaMStR run, including (1) core_orthologs (comprises of OMA orthologous group - OG, or OMA pairs - OP. Each OG/OP has its own directory, where a multiple fasta file and a corresponding profile HMM can be found), (2) genome_dir (contains gene sets of taxa, from which the orthologs are originated), (3) blast_dir (holds the blast databases of those gene sets within genome_dir), and an optional (4) weight_dir (contains feature architecure annotations of all gene sets).
dcc2 is distributed as a python package called dcc2. It is compatible with Python ≥ v3.7.
You can install dcc2 using pip
:
python3 -m pip install dcc2
or, in case you do not have admin rights, and don't use package systems like Anaconda to manage environments you need to use the --user
option:
python3 -m pip install --user dcc2
and then add the following line to the end of your ~/.bashrc or ~/.bash_profile file, restart the current terminal to apply the change (or type source ~/.bashrc
):
export PATH=$HOME/.local/bin:$PATH
After installing dcc2, you need to run the prepare script to download and parse required OMA browser data.
You can do it by running this command
prepareDcc -o /output/path/for/oma/data
For parsing OMA orthologs by using an OMA group ID:
parseOmaById -g 1 -n HUMAN,THEAM,DESM0 -o /output/path/ -j jobName --cpus 8
Or using list of OMA taxa:
parseOmaBySpec -n HUMAN,ECOLI,YEAST -o /output/path/ -j jobName --annoFas --cpus 8
If only 2 OMA taxa are given, you can choose to use OMA pairs instead of OMA groups:
parseOmaBySpec -n HUMAN,ECOLI -t pair -o /output/path/ -j jobName --annoFAS --cpus 8
For parsing an output from OMA-standalone, dcc2 requires:
- the output orthoXML file from OMA,
- a taxon mapping file in tab-delimited format containing 3 columns
<NCBI taxon ID> <Original taxon name> <Abbr. taxon name>
, where original taxon name is the name that is written in the orthoXML input file and abbr. taxon name is its abbreviation spcide code (for example: HOMSA for Homo sapiens) - protein set of included taxa. This can be either given as a folder, or automatically downloaded from OMA database
parseOrthoxml -i input.orthoxml -m mapping_file.txt -g /path/to/gene/set -o /output/path/ -j jobName --annoFas --cpus 8
Any bug reports or comments, suggestions are highly appreciated. Please open an issue on GitHub or be in touch via email.
For further support or bug reports please contact: [email protected]