v2.0.2

• Fixes #39
• Fixes #40
• Fixes #36
• Fixes #38
  • Removed internet access
  • Added new wheels for installing Python packages
• Fixes #37
  • Removed Sophie Ratkai from the developers
  • Use all available cores when downloading inputs

v2.0.1

Other dockers, that have custom modules, can now be processed by eggd_multiQC

v2.0.0

New release of a DNAnexus app to run the MultiQC tool

New in this release:

• Docker image of the MultiQC tool is an input and can be changed without having to change the app (should use a image.tar.gz that is signed off from 001_Reference)
• Input files to the app/report are downloaded based on their location (project:/output/workflow/) and their filename extension, which are specified in the "dx_sp" section of the config.yaml --> This change means that all config.yaml needs to be updated to include a dx_sp section (example config files are available in this repo).
• A list of input file IDs is saved to a text file and uploaded as an output of the app for better tracking.
• Percentage coverage is calculated at depths parsed from the user input field (comma separated values)
• A new custom metric (% usable unique bases on-target) is also calculated in the same script from HSmetrics files when the 'calculate custom metrics' input is set to True.

Configuration

• requires a config.yaml file which has a "dx_sp" section specifying file name extension for input files, list of modules to be displayed and conditional formatting for data display in the report.
• all supported modules are available with this app.

v1.1.2

now includes Somalier by downloading files from the project:output/single/multi/relate2multiqc folder.
small improvement in speed by parallelising the download of sentieon metric files and upload of outputs.
READme now includes a template command for running the applet.

v1.1.1

New features:

• new module added: bcl2fastq. The input file Stats.json is downloaded from the input project
• option to run app on a single folder (running on a workflow output dir with or without multi-sample workflow is still supported)
• option to include custom coverage values calculated at 200, 250, 300, 500 and 1000x
  optional inputs default to 'false'
  Further information about the app/tool and how to incorporate into workflows can be found here

v1.1.0

Use dockerised MultiQC_egg, which is based on ewels/MultiQC:v1,9 with modified happy and added sentieon modules.
App now takes a project name and either a single folder or a pair of single and multi sample workflows to extract QC tool output files and feed them in a single input folder to MultiQC. Input folder is renamed to project-single_sample without '002_' and '_clinicalgenetics' and is also shown on the report for easier identification of the data in the report.

v1.0.6

Added requirements.txt to pip install specific versions of dependencies

v1.0.5

Update app to include wildcards in app output directory names to make it more flexible for future app updates.

Split Happy values into separate tables

This app release uses a specific commit of the forked MultiQC tool with sentieon and modified happy module. Should be used with a config file v1.0.3 or later as the happy table column names have changed (for conditional formatting).

v1.0.3

changed: download happy output files into separate folder and create snp.csv and indel.csv into inp folder on which MultiQC is ran. Along with the updated gemini config file v1.0.2, html reports can have different conditional formatting rules as snp and indel values are shown in different tables.
It is also more robust in that it trims whitespace before and after user input strings and can handle multiple happy output files.