srdsam / exomepipe

This is a nextflow-based pipeline built to identify causative variants in subjects disease phenotype, based on exome and HPO encoded data.

Exome Pipe

Introduction

This is a nextflow-based pipeline built to analyze exome data. There are two main pathways which the pipe will run:

• Unrelated samples

• Trio samples (although technically there is no limit, there could be numerous related samples)

Structure

Nextflow script containing pre-processing and instructions for running sarek and Exomiser on FastQ data.

Pipeline Flow:

Stage 1:

1. cleanBeds: Cleans up the BED file for the sample group

2. renameFastQs: Groups fastQ files by pairs and renames them

3. runPipe: Runs sarek pipeline

Stage 2 (differs for multisample and triosample):

3. produceHpoString: Produces a short string from hpo file using Proband as key

4. produceExomiserYAML: Produces a YAML analysis file specifically for Proband

5. produceExomiserBatch: Produce a batch.txt file which lists the analysis YAML files (only in multiSample pipe)

6. runExomiser: Runs exomiser using batch analysis file

Use of pipe.nf

Set alias nextflow=/efs/sam/bin/nextflow

Then copy and paste pipe.nf into Haggis (vi pipe.nf - i to insert and then :wq to save and quit)

-c: Configuration file

--bed: BED file to use with samples (is always SureSelect_v6.bed)

--fastq: Directory containing .fastq files

--hpo: Directory containing hpo files for unrelated samples || Hpo file for proband in trio sample

--ped: .ped file containing pedigree.

--pipe: multiSample || trioSample

Example Commands:

nextflow run pipe.nf -profile slurm \
-c /efs/sam/configScripts/slurm.config \
--bed /efs/sam/Macrogen_HN00115050/SureSelect_v6.bed \
--fastq /efs/sam/Macrogen_HN00115050/fastq \
--hpo /efs/sam/Macrogen_HN00115050/hpo \
--ped /efs/sam/Macrogen_HN00115050/ped \
--pipe multiSample

OR

nextflow run pipe.nf -profile slurm \
-c /efs/sam/configScripts/slurm.config \
--bed /efs/sam/Macrogen_HN00115050/SureSelect_v6.bed \
--fastq /efs/sam/trio_example/fastq \
--hpo /efs/sam/trio_example/hpo/141641.hpo \
--ped /efs/sam/trio_example/ped \
--pipe trioSample

Runtime will be roughly 6-7 hours.

Input Assumptions

• Samples are named in the format proband_1.fastq.gz
• .bed files contain chr prefix and 2 lines of unneccessary headers.

slurm.config

Configuration parameters for slurm when running pipe.nf and nf-core/sarek.

DEPRECATED: pipe_multiSample.nf/pipe_trioSample.nf

Each contains the test pipelines for analyzing trio or multi-sample data. Replaced by --pipe tag in pipe.nf.

DEPRECATED: exomiser-template.yml

Template for exomiser analysis.

Ideas for future development

• Add in function which automatically uploads data to database