The current pipeline analyze the exon-seq data generated from either of Regeneron’s own pipeline (SPB) or Functionally
Equivalent (FE) piplines from UKbiobank. Initially, the raw input data (.bed,.fam, .bai) are filtered and converted to vcf files using two plink2 processes. Then, the variants (vcf file) are annotated in the third process, using the ensembleVariant Effect Predictor (vep) tool. The corresponding result is then processed using an in-house tool called SEAK. The pipeline has a total of four processes. The tools used for all the four processes are containerized in the docker image
git clone https://github.com/HealthML/Nextflow-pipeline.git
cd Nextflow-pipeline
git checkout dev_nf
make install
To install the docker image for all the process tools using Docker, run the Makefile command in the container directory.
cd container
make docker-build
In order to run the pipeline for the data generated from Regeneron’s own pipeline (SPB) or Functionally
Equivalent (FE) pipleine from UKbiobank using the VEP's cache references, please use the following command. For example, if you wanna run the samples from FE pipeline try the follwoing command on the terminal.
./nextflow run main.nf -resume --samples ukb_FE_50k_exome_seq
The pipeline downloads automatically hg38 fasta file. However, for the current pipeline I am using reference genome (.fa), the annoation file (.gtf) and their corresponding indexed files (.fai & .tbi files). For runing the pipeline using these references, please run the following command on the terminal.
./nextflow run main.nf --ref_fa /home/Alva.Rani/UKbiobank/derived/projects/kernels_VEP/Homo_sapiens.GRCh38.dna.primary_assembly.fa --gtf /home/Alva.Rani/data/reference/Homo_sapiens.GRCh38.97.gtf.gz --gtf_tbi /home/Alva.Rani/data/reference/Homo_sapiens.GRCh38.97.gtf.gz.tbi
If you can access the VM server and the above mentioned folder, there is index for the reference genome.
Otherwise, you can also run the whole pipeline by using the following one liner,
./nextflow run main.nf