Add the decoy Sanger tool to the library

[ypriverol]

We need to add the decoy Sanger tool to the library. The tool is the following:

https://www.sanger.ac.uk/science/tools/decoypyrat

[ypriverol]

@yafeng @husensofteng I have added the decoy tool. The way to test it is using the command:

proteomicsdevmbpr1:pypgatk yperez$ python3.7 pypgatk_cli.py generate-decoy -h 
Usage: pypgatk_cli.py generate-decoy [OPTIONS]

Options:
  -c, --config_file TEXT          Configuration file for the protein database
                                  decoy generation
  -o, --output TEXT               Output file for decoy database
  -i, --input TEXT                FASTA file of target proteins sequences for
                                  which to create decoys (*.fasta|*.fa)
  -s, --cleavage_sites TEXT       A list of amino acids at which to cleave
                                  during digestion. Default = KR
  -a, --anti_cleavage_sites TEXT  A list of amino acids at which not to cleave
                                  if following cleavage site ie. Proline.
                                  Default = none
  -p, --cleavage_position TEXT    Set cleavage to be c or n terminal of
                                  specified cleavage sites. Options [c, n],
                                  Default = c
  -l, --min_peptide_length INTEGER
                                  Set minimum length of peptides to compare
                                  between target and decoy. Default = 5
  -n, --max_iterations INTEGER    Set maximum number of times to shuffle a
                                  peptide to make it non-target before
                                  failing. Default=100
  -x, --do_not_shuffle TEXT       Turn OFF shuffling of decoy peptides that
                                  are in the target database. Default=false
  -w, --do_not_switch TEXT        Turn OFF switching of cleavage site with
                                  preceding amino acid. Default=false
  -d, --decoy_prefix TEXT         Set accession prefix for decoy proteins in
                                  output. Default=DECOY_
  -t, --temp_file TEXT            Set temporary file to write decoys prior to
                                  shuffling. Default=protein-decoy.fa
  -b, --no_isobaric TEXT          Do not make decoy peptides isobaric.
                                  Default=false
  -m, --memory_save TEXT          Slower but uses less memory (does not store
                                  decoy peptide list). Default=false
  -h, --help                      Show this message and exit.

[ypriverol]

@yafeng We need to take a decision about the DECOY Ids. The way the tool generate the ids is by creating a new protein with the following accession DECOY_1 , DECOY_2... Can you check this works in our pipelines? The second problem I see is that we lost all the information related with the proteins, My guess is that we will recover that information in the re-mapping after the protein identification step?

[yafeng]

@ypriverol I have modified the script so that the output decoy sequences contain original protein ID, this is required to distinguish decoys from different classes.

[ypriverol]

I will chec kthe code because it doesn't look like it will work now. LEt me check

[ypriverol]

@yafeng I just updated the code and it works fine now. I removed also the from the concatenation, that means that the decoy prefix should contain the `` itself. Can you test now the current version? If you are happy with that I can close this issue.

[yafeng]

@ypriverol this is first time I tried, i got this error. Do I forget to setup something?

Traceback (most recent call last): File "pypgatk/pypgatk_cli.py", line 11, in from pypgatk.commands import ensembl_downloader as ensembl_downloader_cmd ModuleNotFoundError: No module named 'pypgatk'

[husensofteng]

It works here. @yafeng that is weird may be you should run the setup.py script with install command

[ypriverol]

The method to use the library now is you need to install it first, as:

python3.7 setup.py install

The current tool is more a library with a commandline tool than simple scripts that is why you need to install it. @enriquea if you succeed can you improve the README.

[yafeng]

Please update the README, the following python packages needs to be installed.

pip install click
pip install PyVCF
pip install gffutils
pip install pyyaml
pip install biopython

[ypriverol]

This packages are added in the requirements.txt you should be able to install then using pip.

[yafeng]

pip install -r requirements.txt did not give any error.

[ypriverol]

The protocol to build the package should be:

1 -

pip install -r requirements.txt 

2-

python3.7 setup.py install

then you should be able to run the script.

[ypriverol]

@yafeng @enriquea did you manage to build the package ?

[enriquea]

I'm getting the following error:

enrique$ python3.6 pypgatk_cli.py --help
Traceback (most recent call last):
  File "pypgatk_cli.py", line 14, in <module>
    from pypgatk.commands import cosmic_to_proteindb as cosmic_to_proteindb_cmd
  File "/anaconda3/lib/python3.6/site-packages/pypgatk-0.0.1-py3.6.egg/pypgatk/commands/cosmic_to_proteindb.py", line 3, in <module>
    from pypgatk.cgenomes.cgenomes_proteindb import CancerGenomesService
  File "/anaconda3/lib/python3.6/site-packages/pypgatk-0.0.1-py3.6.egg/pypgatk/cgenomes/cgenomes_proteindb.py", line 3, in <module>
    from Bio import SeqIO
ModuleNotFoundError: No module named 'Bio'

[husensofteng]

@enriquea you need to install the biopython package as noted above I will update the docs to list the requirements.

[ypriverol]

@yafeng If the decoy tool work for you please close this issue.