CAW

Nextflow Cancer Analysis Workflow Prototype developed at the National Genomics Infastructure at SciLifeLab Stockholm, Sweden.

Version

0.8.3

Authors

• Sebastian DiLorenzo (@Sebastian-D)
• Jesper Eisfeldt (@J35P312)
• Maxime Garcia (@MaxUlysse)
• Szilveszter Juhos (@szilvajuhos)
• Max Käller (@gulfshores)
• Malin Larsson (@malinlarsson)
• Björn Nystedt (@bjornnystedt)
• Pall Olason (@pallolason)
• Pelin Sahlén (@pelinakan)

Installation and first execution

See the Nextflow documentation from SciLifeLab

See the workflow installation documentation

Usage

I would recommand to run Nextflow within a screen session (cf help on screen).

nextflow run SciLifeLab/CAW --sample <file.tsv> [--steps STEP[,STEP]]

All variables and parameters are specified in the config (cf configuration options) and the sample files.

Steps

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)
• recalibrate (will start workflow with non-recalibrated BAM files)
• skipPreprocessing (will start workflow with recalibrated BAM files)
• 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)

Verbose

To have more information about files being processed, you can use the verbose option

nextflow run SciLifeLab/CAW --sample mysample.tsv --steps preprocessing --verbose

Nextflow parameters

config

More informations on Nextflow documentation

-c <file.config>

If no config file is specified, Nextflow will look for one in Nextflow intallation $NXF_HOME/config of for one in the current directory nextflow.config.

The config file provided as an example is a config file specific to Swedish UPPMAX milou cluster, but can be easily modified to suit any clusters.

You can use this file as an example to make your own config file. And you can even if needed make several config files (for example if you want to have a config file for each UPPMAX project identifier).

clean

Use nextflow clean -f to remove everything contained in the work directory. Do not worry, non-recalibrated bam, indexes and recalibration tables as well as recalibrated bams and index are stored respectively in the Preprocessing/NonRecalibrated and Preprocessing/Recalibrated directories. And variant calling files are stored in the VariantCalling directory.

nextflow clean -f

resume

Use -resume to restart the workflow where it last failed.

nextflow run SciLifeLab/CAW --sample mysample.tsv --steps preprocessing -resume

info

Use info to get information about the workflow.

nextflow info SciLifeLab/CAW

pull

Use pull to update the workflow.

nextflow pull SciLifeLab/CAW

Nextflow processes

Several processes are run within Nextflow We divide them for the moment into 2 main steps

Preprocessing:

• Mapping - Map reads with BWA
• MergeBam - Merge BAMs if multilane samples
• RenameSingleBam - Rename BAM if non-multilane sample
• MarkDuplicates - using Picard
• CreateIntervals - using GATK
• Realign - using GATK
• CreateRecalibrationTable - using GATK
• RecalibrateBam - using GATK

Variant Calling:

• RunMutect2 - using MuTect2 shipped in GATK v3.6
• concatFiles - merge MuTect2 results
• VarDict - run VarDict on multiple intervals
• VarDictCollatedVCF - merge Vardict results
• RunStrelka - using Strelka 1.0.15
• Manta - run Manta 1.0.0

TSV file for sample

It's a Tab Separated Value file, based on: subject status sample lane fastq1 fastq2, subject status sample bam bai recal or subject status sample bam bai Quite straight-forward:

• subject is the ID of the Patient
• status is the status of the Patient, (0 for Normal and 1 for Tumor)
• sample is the Sample ID (It is possible to have more than one tumor sample for each patient)
• lane is used when the sample is multiplexed on several lanes
• fastq1 is the path to the first pair of the fastq file
• fastq2 is the path to the second pair of the fastq file
• bam is the realigned bam file
• bai is the index
• recal is the recalibration table

Example of TSV files

See the workflow TSV example documentation

Tools and dependencies

• nextflow 0.17.3
• bwa 0.7.8
• samtools 1.3
• picard 1.118
• GATK 3.6
• R 3.2.3
• gcc 4.9.2
• perl 5.18.4
• strelka 1.0.15
• manta 1.0.0

Use cases

See the workflow use cases documentation