add dockerfile for pgma

[rpetit3]

This PR adds a fork of pmga for serotyping, serotyping and MLST of all Neisseria species and Haemophilus influenzae.

It is originally a part of Bacterial Meningitis Genome Analysis Platform (BMGAP) which is available on the OAMD portal, but I wanted to be able to run it on the command-line.

This docker build includes the pre-built BLAST databases which makes the image slightly larger. If there is enough interest in the version of pmga, I would consider looking into making the pre-builds smaller (e.g. do we need 5k databases?)

• This comment contains a description of what is in the pull request.

• Build your own docker image using a Dockerfile
  • Directory structure should be name of the tool in lower case with special characters removed with a subdirectory of the version number (i.e. spades/3.12.0/Dockerfile)
  • Includes the recommended LABELS
• (Optional) Dockerfile is built with best practices and has been approved by a linter (such as https://hadolint.github.io/hadolint/)
• Edit main README.md
• Edit Program_Licenses.md
• Create a simple container-specific README.md in the same directory as the Dockerfile (i.e. spades/3.12.0/README.md)
• Write a GitHub actions workflow
  • Should be located in .github/workflows/ and named test-.yml (i.e. .github/workflows/test-spades.yml)
  • Any files required for building are located in the same directory as the Dockerfile (i.e. spades/3.12.0/my_spades_tests.sh)
  • Have successfully run the workflow "Test image" in your forked repository

[rpetit3]

Example run:

docker run --rm -u $(id -u):$(id -g) -v ${PWD}:/data pmga:latest pmga.py /data/GCF_900478275.fna --blast /pmga/blastdbs --force
2022-02-08 02:39:38:root:INFO - Found --force, removing existing /data/pmga
2022-02-08 02:39:39:root:INFO - Using Mash predicted species: Haemophilus influenzae
2022-02-08 02:39:39:root:INFO - Step 1. BLASTing against PubMLST DBs
2022-02-08 02:39:39:root:INFO - Blasting against pubMLST database with 1 workers for file /data/GCF_900478275.fna
2022-02-08 02:49:05:root:INFO - Completed BLAST for /data/GCF_900478275.fna
2022-02-08 02:49:06:root:INFO - Evaluating BLAST Results
2022-02-08 02:49:06:root:INFO - Step 2. Parsing BLAST results
2022-02-08 02:49:06:root:INFO - Parsing BLAST data to identify alleles to extract from pubMLST
2022-02-08 02:49:08:root:INFO - Compiling results
2022-02-08 02:49:08:root:INFO - Step 3. Writing outputs

[kapsakcj]

Thanks for the PR!

Docker image builds successfully, although it does take quite a long time to build.....not your fault. I'll likely build locally and push to dockerhub/quay manually since this one takes so long to build.

I was able to run pmga successfully on one N. gonnorhea genome and one H. influenzae genome.

$ docker run --rm -u $(id -u):$(id -g) -v ${PWD}:/data robert/pmga:latest pmga /data/GCF_021535585.1_ASM2153558v1_genomic.Ngonorrhoeae.fna --blast /pmga/blastdbs -t 4 -o /data/Ngonorrhoeae-pmga-test
2022-02-08 15:50:05:root:INFO - Using Mash predicted species: Neisseria gonorrhoeae
2022-02-08 15:50:05:root:INFO - Step 1. BLASTing against PubMLST DBs
2022-02-08 15:50:05:root:INFO - Blasting against pubMLST database with 4 workers for file /data/GCF_021535585.1_ASM2153558v1_genomic.Ngonorrhoeae.fna
2022-02-08 16:07:34:root:INFO - Completed BLAST for /data/GCF_021535585.1_ASM2153558v1_genomic.Ngonorrhoeae.fna
2022-02-08 16:07:59:root:INFO - Evaluating BLAST Results
2022-02-08 16:07:59:root:INFO - Step 2. Parsing BLAST results
2022-02-08 16:07:59:root:INFO - Parsing BLAST data to identify alleles to extract from pubMLST
2022-02-08 16:08:04:root:INFO - Compiling results
/usr/local/lib/python3.8/dist-packages/Bio/Seq.py:2979: BiopythonWarning: Partial codon, len(sequence) not a multiple of three. Explicitly trim the sequence or add trailing N before translation. This may become an error in future.
  warnings.warn(
2022-02-08 16:08:05:root:INFO - Step 3. Writing outputs

# roughly 18 minutes wall time


# H influenzae ran a bit faster than the Ngonnorhea
$ time docker run --rm -u $(id -u):$(id -g) -v ${PWD}:/data robert/pmga:latest pmga /data/GCF_900478275.1_34211_D02_genomic.H-influenzae.fna --blast /pmga/blastdbs -t 4 -o /data/Hinfluenzae-pmga-test
2022-02-08 16:12:30:root:INFO - Using Mash predicted species: Haemophilus influenzae
2022-02-08 16:12:30:root:INFO - Step 1. BLASTing against PubMLST DBs
2022-02-08 16:12:30:root:INFO - Blasting against pubMLST database with 4 workers for file /data/GCF_900478275.1_34211_D02_genomic.H-influenzae.fna
2022-02-08 16:16:22:root:INFO - Completed BLAST for /data/GCF_900478275.1_34211_D02_genomic.H-influenzae.fna
2022-02-08 16:16:24:root:INFO - Evaluating BLAST Results
2022-02-08 16:16:24:root:INFO - Step 2. Parsing BLAST results
2022-02-08 16:16:24:root:INFO - Parsing BLAST data to identify alleles to extract from pubMLST
2022-02-08 16:16:26:root:INFO - Compiling results
/usr/local/lib/python3.8/dist-packages/Bio/Seq.py:2979: BiopythonWarning: Partial codon, len(sequence) not a multiple of three. Explicitly trim the sequence or add trailing N before translation. This may become an error in future.
  warnings.warn(
2022-02-08 16:16:26:root:INFO - Step 3. Writing outputs

real    3m59.915s
user    0m0.049s
sys     0m0.028s

Generated the expected output files 👍

[rpetit3]

Thank you!

[kapsakcj]

pmga is now available on dockerhub and quay 🥳

https://hub.docker.com/r/staphb/pmga/tags

https://quay.io/repository/staphb/pmga?tab=tags