I would recommand to run Nextflow within a screen session (cf help on screen). It is recommanded to run only one instance of CAW for one patient in the same directory. Meaning there should be only one patient analysed in one directory. The typical command line is:
nextflow run SciLifeLab/CAW --sample <file.tsv>
All variables and parameters are specified in the config (cf configuration documentation).
All samples are specified in the TSV files (cf TSV documentation).
Steps are used to configure which processes will be runned or skipped in the workflow. Different steps to be separated by commas. Possible values are:
- preprocessing (default, will start workflow with FASTQ files)
- realign (will start workflow with BAM files (with T/N BAMs that were not realigned together))
- skipPreprocessing (will skip entire preprocessing (Only with T/N BAMs that were realigned together))
- MuTect1 (use MuTect1 for VC)
- MuTect2 (use MuTect2 for VC)
- VarDict (use VarDict for VC)
- Strelka (use Strelka for VC)
- HaplotypeCaller (use HaplotypeCaller for normal bams VC)
- Manta (use Manta for SV)
- Ascat (use ascat for CNV)
- MultiQC (Make a QC report)
To specify your UPPMAX project number ID. It can also be specified in your config
file (cf configuration documentation).
nextflow run SciLifeLab/CAW --sample <file.tsv> --project <UPPMAX_Project>
To have more information about files being processed, you can use the verbose option:
nextflow run SciLifeLab/CAW --sample mysample.tsv --verbose
To test CAW and run it on smaller dataset, use one of the following option:
--test
will test steppreprocessing
on tiny test data.--testRealign
will test steprealign
on tiny test data. Need to runnextflow run SciLifeLab/CAW --test
before.--testCoreVC
will test stepsskipPreprocessing
,MuTect1
,Strelka
andHaplotypeCaller
on tiny test data. Need to runnextflow run SciLifeLab/CAW --test
before.--testSideVC
will test stepsskipPreprocessing
,Ascat
,Manta
andHaplotypeCaller
on test downsampled set.
See the options documentation
More informations on the SciLifeLab Nextflow documentation. The default profile is standard
. If you want you can use your own profile:
nextflow run SciLifeLab/CAW --sample mysample.tsv -profile myprofile
A standard profile is defined in nextflow.config
. You can use the config/milou.config
file as a base to make a new config
file that you can specify directly (or add as a profile):
nextflow run SciLifeLab/CAW --sample mysample.tsv -c config/milou.config
To update workflow to the latest version use:
nextflow pull SciLifeLab/CAW
If there is a feature or bugfix you want to use in a resumed or re-analyzed run, you have to update the workflow to the latest version. By default it is not updated automatically, so use something like:
nextflow run -latest SciLifeLab/CAW --sample mysample.tsv -resume