We have developed Docker for 2 pipelines:
- Whole Genome Sequencing (WGS)
- Gene Regulatory Network (SCENIC+)
For WGS, we have developed the nextflow pipeline
- FASTQC
- GATK
- PICARD
- BWA
- ANNOVAR
- SAMTOOLS
singularity pull docker://abhinavjain1993/wgs:amd64
singularity exec -bind /diazlab/data3/.abhinav/tools/singularity/data/:/data --contain-all wgs_amd64.sif /bin/bash
### Download the data
fasterq-dump --split-3 DRR001913 -O ./ -t ./tmp/
head -10000 DRR001913_2.fastq > DRR001913_2_10000.fastq
head -10000 DRR001913_1.fastq > DRR001913_1_10000.fastq
# java -jar /usr/local/bin/trimmomatic-0.39.jar PE -threads 4 DRR001913_1_10000.fastq DRR001913_2_10000.fastq trim_DRR001913_1_10000.fastq unpaired_DRR001913_1_10000.fastq trim_DRR001913_2_10000.fastq unpaired_DRR001913_2_10000.fastq AVGQUAL:20 SLIDINGWINDOW:5:20 MINLEN:50
wget https://ftp.ensembl.org/pub/release-107/fasta/homo_sapiens/dna/Homo_sapiens.GRCh38.dna.primary_assembly.fa.gz
### Run it using Nextflow
nextflow run exome_seq_nextflow.nf --with-apptainer /diazlab/data3/.abhinav/tools/singularity/wgs_amd64.sif
it is install in the Conda environment scenicplus:amd64 is being designed for the linux environment
docker pull abhinavjain1993/wgs:amd64
docker pull abhinavjain1993/scenicplus:amd64
docker pull abhinavjain1993/scenicplus_docker:latest
source activate scenicplus
singularity pull docker://abhinavjain1993/scenicplus_docker:amd64
singularity exec --bind <Mounted_directory>:/data --contain-all scenicplus_docker_latest.sif /bin/bash
Note While Running the singularity please keep all files in the same folder where you are working.