hmmer-cwl

This repository contains a few Common Workflow Language (CWL) tool and workflow descriptions for HMMER. The workflows follow the examples presented in the turorial section of HMMER user guide.

Install

Prerequisites

an implementation of the Common Workflow Language. Because the workflows presented here use the v1.0 CWL specifications, some features are not yet supported by most implementations. We recommend to use the reference implentation.
Docker as we run tools in Docker images.

Get the code

Grab the latest version of the documents:

git clone https://github.com/matthiasblum/hmmer-cwl.git

Get the data

Four input files are required to run the workflows. Three files are from the HMMER turorial:

fn3.sto: an example alignment of 106 fibronectin type III domains.
globins4.sto: an example alignment of four globin sequences.
Pkinase.sto: the Pfam 22.0 Pkinase seed alignment of protein kinase domains.

And the last one is the UniProt Swiss-Prot FASTA format flatfile.

To download these files, run the setup workflow:

cd hmmer-cwl
cwl-runner workflows/setup.cwl setup.yml

Workflows

Searching a protein sequence database with a single protein profile HMM

This 2-step workflow builds a profile HMM from a multiple-sequence alignment, and searchs a sequence database with the generated model.

cwl-runner workflows/search.cwl search.yml

Searching a profile HMM database with a query sequence

This 4-step workflow builds HMM profiles from several multiple-sequence alignments, merge the models to create an HMM database flatfile, index this database then search it.

cwl-runner workflows/scan.cwl scan.yml

Todo

Explain workflows' output parameters
Explain input paramaters contained in YAML job files
Add other examples from tutorial
Add "real world" workflow
Support all options (input parameters)