adeslatt / methylseqcaptureseq Goto Github PK

• for each command in bash script, make into a Nextflow process and test with test data (https://github.com/nf-core/test-datasets/tree/methylseq/testdata)

• piggy back from tools with code already written in Nextflow

• create new code in nextflow for biscuit and wgbs_tools

• create a new docker file and build image for wgbs_tools

• create Flowcraft components for each Nextflow process

• build a pipeline with Flowcraft

• run Nextflow pipeline on CloudOS (lifebit)

Starting with the bash scripts -- take them to build the required Nextflow pipeline for execution on Amazon or other cloud resources.

What is the best practice when we need to containerize another repository?

Such as wgbs_tools

We need to containerize the wgbs_tools for use in our nextflow pipeline for methylseq.

#Dependencies wgbs_tools

samtools
python 3+
c++ boost library

#How to download:
git clone https://github.com/nloyfer/wgbs_tools.git
cd wgbs_tools

#Compile the cpp files:
python3 setup.py

#used to be this:
g++ -std=c++11 src/pat2beta/stdin2beta.cpp -o src/pat2beta/stdin2beta
g++ -std=c++11 src/pat_sampler/sampler.cpp -o src/pat_sampler/pat_sample
g++ -std=c++11 wgbs_tools/pipeline_wgbs/patter.cpp -o wgbs_tools/pipeline_wgbs/patter
g++ -std=c++11 wgbs_tools/pipeline_wgbs/match_maker.cpp -o wgbs_tools/pipeline_wgbs/match_maker

#setup reference genome
python3 wgbs_tools.py init_genome /path/to/genome.fa GENOME_NAME

#could loop in reference
#ln -s /path/to/genome.fa .

#init.genome dependencies
import argparse
from utils_wgbs import delete_or_skip, validate_single_file, eprint, IllegalArgumentError, DIR
import re
import numpy as np
import pandas as pd
import os.path as op
import os
from itertools import groupby
import subprocess
from multiprocessing import Pool
import multiprocessing

#bam2pat
python3 wgbs_tools.py bam2pat BAM_PATH

#bam2pat dependencies
import os
import os.path as op
import argparse
import subprocess
from multiprocessing import Pool
from utils_wgbs import IllegalArgumentError, match_maker_tool, patter_tool, add_GR_args
from init_genome_ref_wgbs import chromosome_order
from pat2beta import pat2beta
from pipeline_wgbs.test import run_test
from genomic_region import GenomicRegion

divide repository into a nextflow pipeline methylSeqCaptureSeq-nf and methylSeqCaptureSeq. The first will contain the nextflow pipeline for processing and the latter will contain any jupyter notebooks that contain the analyses.

https://github.com/nloyfer/wgbs_tools

samtools version 1.9
trim_galore version 0.6.4_dev (dependency fastqc and cudadapt v. 2.8)
biscuit version 0.3.8.20180515
bamtools version 2.5.1
picard version 2.18.15-snapshot
wgbs_tools = github repository (dependencies python3, c++ boost library, samtools)
https://github.com/nloyfer/wgbs_tools