Methylation information of cell-free DNA in cerebrospinal fluid (CSF), plasma and other body fluids has been utilized to diagnose early-stage diseases and predict therapy response. However, in common clinical settings only a very low amount of cell-free DNA can be purified, usually in the range of a few dozen nanograms. Even worse is that the cfDNA is fragmented and peaked around 120 bases. Whole genome bisulfite sequencing (WGBS), the gold standard to measure DNA methylation, on such a low amount of fragmented DNA molecules introduces a critical data analysis challenge, which is the low mapping ratio. This, in turn, generated low sequencing depth of each CpG and low coverage of genome-wide CpGs sites. The problem of insufficient informative CpGs became the bottleneck of the clinical application of cell-free DNA WGBS assays. Hence, we developed LiBis, a novel method for low input bisulfite sequencing data alignment. By dynamically clipping initially unmapped reads and remapping clipped fragments, we conservatively rescued those reads and uniquely aligned them to the genome. With much improved mapping ratio, LiBis as an integrative toolkit increases the number of informative CpGs and the precision of methylation status of each CpG. High sensitivity and cost efficiency achieved by LiBis for low input samples allow discoveries of genetic and epigenetic features for downstream analysis and biomarker identification from liquid biopsy.
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.
Install Python3, conda and pip Python3 and conda can be downloaded and installed from https://www.anaconda.com/distribution/ Please make sure than Python version is at least 3.6
Please run following command to make sure python, pip and conda are correctly installed.
python --version
pip --version
conda --version
conda config --add channels defaults
conda config --add channels bioconda
conda config --add channels conda-forge
conda install libis
LiBis integrates FastQC, Bedtools, Trim-galore, moabs and samtools. These packages can be installed independently or by conda.
conda install -c bioconda fastqc
conda install -c bioconda bedtools
conda install -c bioconda cutadapt
conda install -c bioconda trim-galore
conda install -c bioconda moabs
conda install -c bioconda samtools=1.1
Check the installation of integrated tools
fastqc --version
bedtools --version
trim_galore --version
moabs
samtools --version
Install LiBis by pip
pip install LiBis
LiBis --help
The parameters of LiBis will be printed to the screen if install successfully.
LiBis also supports the Docker installation.
wget https://github.com/Dangertrip/LiBis/archive/master.zip
unzip master.zip
cd LiBis-master/
docker build --tag=libis ./
docker run libis LiBis --help
git clone https://github.com/Dangertrip/LiBis.git
cd LiBis/Example/
# Run LiBis from begining:
LiBis -n sample1_mate1.fq.gz,sample1_mate2.fq.gz sample2_mate1.fq.gz,sample2_mate2.fq.gz -r /PATH_TO_FASTA_REFERENCE
docker run -v /path/to/yourdata:/data/ libis LiBis -f 0 -n name1.fq.gz name2.fq.gz -r hg19.fa
Yue Yin, Jia Li, Jin Li, Minjung Lee, Sibo Zhao, Linlang Guo, Jianfang Li, Mutian Zhang, Yun Huang, Xiao-Nan Li, Deqiang Sun
This project is licensed under the MIT License.
show this help message and exit
Required. Enter a number, 0 means using parameter to set up, 1 means using text file to set up
Setting txt file name. Ignore if -f is 0
Required. Fastq file name.
Clip mode. 0 means close. 1 means open. default=1
Labels for samples
Genome the reference belong to.(Use for plotting) hg38/hg19/mm10/mm9 and so on. Plotting script will not avaliable if leave it blank. default: 'hg38'.
Window length for clipping mode, default=40
Step size for clipping mode, default=5
Process using for one pipeline. Normally bsmap will cost 8 cpu number. So total will be 8p.
Required. Reference genome file name.
Do(1) quality control or not(0)
Do(1) trimming or not(0). Don't need to do trimming if you use clip mode.
Plot setting. Set the bin size for averaging methylation ratio among samples, default=1000k
Minimal length for recombined reads, default=46
Processed bam file for the first step. If bam files are offered here, the first step of bsmap will be skipped. BAM files can only be generated by BSMAP. Different files should be seperated by ','. If there's no bam file for part of the samples, leave the space blank. For example: a.bam,b.bam,,,,f.bam
Run mcall for mapped bams or not
Generate the final report or not
Run LiBis as MOABS module. Please only apply this when integrating LiBis to MOABS.
Skip the checking step for result folders. Using this parameter may rewrite the previous results.
Given processed bam file contains unmapped reads. Only use with -bam is open.
Temporary files are in gz format.
Keep all temp files.