cfDNA-kit

Introduction

cfDNA-kit is a software use to detect cfDNA feature using WPS signal.

Require：

python2.7
bx-python=0.7.4
pysam=0.7.6
scipy=1.2.3

python3.8
yaml
pickle
numpy
matplotlib
sklearn
scipy
seaborn
pandas
tqdm

Install

git clone https://github.com/lysovosyl/cfDNA-kit.git
cd cfDNA-kit
conda create -n cfDNA-kit python==3.8
conda activate cfDNA-kit
pip install yaml pickle tqdm numpy matplotlib sklearn scipy seaborn pandas

conda create -n wps-kit python==2.7
conda activate wps-kit
pip install bx-python pysam scipy
conda install r-base

After you install those software, you should modify those software path in config file, including ./config/config.yaml and ./prepare_bam/test/config.txt

Usage

Step by step

Fq to bam

1. First you should build a sample file (samples.lst), just like ./prepare_bam/test/samples.lst
2. Create a new folder and ensure that samples.lst and config.txt are located in the same folder.
3. Note! You should make sure the softwares in config.txt have been installed and the path of those softwares are correct.
4. Run command.

perl ./prepare_bam/bin/prepare_cfdna_bam.pl config.txt
cd out/[sample_name]
make

Analyse WPS signal of cfDNA

1. You should build a sample file (sample.lst), just like ./sample.lst
2. This file including four rows, [group_name,group_type,file_name,bam_path]
3. Run command.

python ./makeshell.py -sample_file ./your_path/sample.lst -save_dir /your_save_path -tmp_dir /your_temp_path -shell_path /your_shell_path -config ./config/config.yaml -thread 70

4. You will get two shell files in your_save_path, including preprocess.sh and summary.sh
5. Then run that two shell files, you should run preprocess.sh at first

cd /your_save_path
sh preprocess.sh && sh summary.sh