Quick Start
This guide is intended to quickly get you up and running with PGAP. If you have any questions please read the FAQs, watch this webinar or look over the rest of the documentation.
To run the PGAP pipeline you will need:
- Python (version 3.6 or higher),
- the ability to run Docker (see https://docs.docker.com/install/ if it is not already installed), Singularity, or Podman
- about 100GB of storage for the supplemental data and working space,
- and 2GB-4GB of memory available per CPU used by your container.
- The CPU must have SSE 4.2 support (released in 2008).
Our software development and the bulk of our testing is conducted in Docker containers on single 8 CPU 32 GB RAM and 16 CPU 64 GB RAM Linux CentOS 7 machines. We have limited experience executing in non-Docker containers (Singularity or Podman) or on Mac and Windows machines. We do not have the resources to help troubleshoot issues with these platforms, or with running PGAP on distributed compute clusters.
Before opening an Issue, please test your installation with the Mycoplasma genitalium genome distributed with the software (MG37), as described below, to verify that your platform is configured correctly. If this test doesn't succeed, try reinstalling fresh. Please also consult the FAQs.
Download the file using either
$ curl -OL https://github.com/ncbi/pgap/raw/prod/scripts/pgap.pyor
$ wget -O pgap.py https://github.com/ncbi/pgap/raw/prod/scripts/pgap.pydepending upon which utility your system has installed. If one does not work, try the other.
Install the pipeline. By default it will install in $HOME/.pgap, but this location can be changed by setting environmental variable PGAP_INPUT_DIR.
$ chmod +x pgap.py
$ ./pgap.py --update # required files are downloaded and extractedRun the pipeline on the Mycoplasmoides genitalium genome provided with the installation:
$ ./pgap.py -r -o mg37_results -g $HOME/.pgap/test_genomes/MG37/ASM2732v1.annotation.nucleotide.1.fasta -s 'Mycoplasmoides genitalium'Output will be located in the mg37_results subdirectory as specified by the -o flag.
To run this pipeline using your own genomes, you will need, at a minimum, the multifasta file for the genome, and the associated organism name (genus or genus species).
$ ./pgap.py -r -o <results> -g <fasta> -s '<organism_name>'
To produce an annotation that is suitable for submission to GenBank, more information is needed, and you will need to provide three input files, all in the same directory. Instructions for preparing your data are in the Input Files section.
- The multifasta file for the genome
- A YAML file containing metadata
- A YAML file that describes the pipeline inputs, including the above two files, <generic.yaml>
$ ./pgap.py -r -o <results> <generic.yaml>
To get a complete list of options, use the -h flag. However, here are some notable options.
| Command | Description |
|---|---|
-g <path>, --genome <path> |
Path to genomic fasta |
-s 'organism', --organism 'organism' |
Genus, or genus species |
-r, --report-usage-true |
Report anonymized usage metadata to NCBI |
-n, --report-usage-false |
Do not report anonymized usage metadata to NCBI |
-o <path>, --output <path> |
Output directory to be created, which may include a full path |
--ignore-all-errors |
Ignore errors from quality control analysis, in order to obtain a draft annotation |
--no-internet |
Disable internet access for all programs in pipeline |
-D <path>, --docker <path> |
Docker-compatible executable (e.g. docker, podman, singularity), which may include a full path like /usr/bin/docker |
--taxcheck |
Also calculate the Average Nucleotide Identity to type assemblies |
--taxcheck-only |
Only calculate the Average Nucleotide Identity to type assemblies, do not run PGAP |
--auto-correct-tax |
Override the organism provided in the input YAML file, if the taxcheck predicts a different organism with high confidence. Use in combination with the --taxcheck flag |
-d, --debug |
Debug mode. Retain intermediate files needed for investigating failures |