alnsl

A nextflow pipeline for alignment of short WGS reads.

Preparing index sequences

• prepare reference index for bwa-mem2

   bwa-mem2 index ref.fa
  

• prepare elprep files using the folowing commands:

  #reference file
  elprep fasta-to-elfasta ref.fa ref.fa.elfasta
  #variats files GTAK4_Bunddle for variant calibration
  elprep vcf-to-elsites <vcf-file> <elsites-file>
  

Conda Environment

We will build a micromamba environment with the needed software

micromamba create -f conda.yml 

Param files

Here we provide an example to map the short reads to hg38, the context of the params files is the following

dbsnp: /mnt/beegfs/labs/DiGenomaLab/databases/references/human/GATK_Bundle/Homo_sapiens_assembly38.dbsnp138.elsites
dbindel: /mnt/beegfs/labs/DiGenomaLab/databases/references/human/GATK_Bundle/Mills_and_1000G_gold_standard.indels.hg38.elsites
ref: /mnt/beegfs/labs/DiGenomaLab/databases/references/human/bwa2/hs38DH.fa
elpre_ref: /mnt/beegfs/labs/DiGenomaLab/databases/references/human/hs38DH.fa.elfasta
alt_js: /mnt/beegfs/home/adigenova/micromamba/envs/aln/bin/bwa-postalt.js
bqsr: true

Save the above content in a file (i.e) : aln-params.yml

Read file

provide a csv file wiht the following information:

sampleId,read1,read2
test2,./test_reads/test2.R1.fq.gz,./test_reads/test2.R2.fq.gz
test3,./test_reads/test3.R1.fq.gz,./test_reads/test3.R2.fq.gz
test,./test_reads/test.R1.fq.gz,./test_reads/test.R2.fq.gz

Currently if a sample is split into several files is necesary to merge the reads before runing the pipeline.

Save the above content in a file (i.e) : reads.csv

runnig the pipeline

nextflow run main.nf --csv reads.cvs -profile uoh -params-file aln-params.yml

in case of failure use:

nextflow run main.nf --csv reads.cvs -profile uoh -params-file aln-params.yml -resume

that will generate a directory called results

Creating an aggregated report

To create an aggregated report across all the samples, is possible to run multiqc on the result directory:

• load the environment

  micromamba activate aln
  

• run multiqc

 multiqc .