microtan0902/erds-pe

#ERDS-pe #Introduction ERDS-pe is a tool designed to detect detect CNVs from whole-exome sequencing (WES) data. ERDS-pe employs RPKM and principal component analysis to normalize WES data and incorporates RD and single-nucleotide variation information together as a hybrid signal into a paired hidden Markov model to infer CNVs from WES data. #Installation ERDS-pe is easy to run. you just need:

1. Python 2.7+ (Python 3 is not yet supported)

2. Numpy, PyVCF, Pysam libraries installed.

3. Target(Probe) files in bed format.

#Running

usage: python erds_pe.py [OPTIONS]

1. Getting RD and transformming to RPKM format

   This command is to extract the read depth (RD) signal from the BAM files and to calculate RPKM values for all samples.

    usage: python erds_pe.py rpkm [OPTIONS] 
   
    --target  <FILE>  Target region file in bed format (required)
   
    --input   <FILE>  A list of bam files list. eg. bam_list_example.txt (required)
   
    --output  <FILE>  Directory for RPKM files (required)
   

2. Merge single sample RPKM files to a union data matrix

    usage: python erds_pe.py merge_rpkm [OPTIONS] 
   
    --rpkm_dir  <FILE>  Giving the RPKM files directory for taking data (required)
   
    --target   <FILE>  Target region file in bed format (required)
   
    --output  <FILE>  Output the data matrix file (optional)
   

3. Normalization

   This command is to normalize RPKM data matrix using principal component analysis.

    usage: python erds_pe.py svd [OPTIONS] 
   
    --rpkm_matrix  <FILE>  Giving the RPKM files PCA normalization (required)
   
    --output  <FILE>  Output the normalized data matrix file (optional)
   

4. Calling CNVs

   This command is to call CNVs from pooled whole-exome sequencing samples.

    usage: python erds_pe.py discover [OPTIONS]  
   
    --params  <FILE>  Parameters file for HMM (required)
   
    --datafile  <FILE>  Normalized data matrix file (required)
    
    --output  <FILE>  Output the normalized data matrix file (optional)
    
    --sample  <STRING>  Giving a specific sample for calling (optional)
   
    --vcf  <FILE>  Taking SNV information from vcf file (optional)
    
    	--hetsnp  <BOOL>  Using or not take heterogenous SNV information into HMM (optional, default FALSE)
    
    	--tagsnp  <BOOL>  Using or not take tagSNP-copy number polymorphism information into HMM (optional, default FALSE)
   
    --tagsnp_file  <FILE>  A file records the linkage disequilibrium information between tagSNP and copy number polymorphism (optional)
   

#File Instruction

1. bam list file (three columns)

   Column 1: Sample_name Column 2: Path of bam files Column 3: The population of the corresponding sample. e.g CEU, YRI, CHB etc.

   Example:

    NA06984	/data/rjtan/1000GP/exome/NA06984.mapped.ILLUMINA.bwa.CEU.exome.20120522.bam	CEU
    NA06985	/data/rjtan/1000GP/exome/NA06985.mapped.ILLUMINA.bwa.CEU.exome.20130415.bam	CEU
    NA06994	/data/rjtan/1000GP/exome/NA06994.mapped.ILLUMINA.bwa.CEU.exome.20120522.bam	CEU
    NA07000	/data/rjtan/1000GP/exome/NA07000.mapped.ILLUMINA.bwa.CEU.exome.20130415.bam	CEU
    ...
   

#Quick Start & a example

1. Step 1:

    python erds_pe.py rpkm
    --input $bamlist_file \
    --target $target_file \
    --output $rpkm_files
   

2. Step 2:

    python erds_pe.py merge_rpkm
    --rpkm_dir $rpkm_files \
    --target $target_file
   

3. Step 3:

    python erds_pe.py svd
    --rpkm_matrix $RPKM_matrix.raw.filtered
   

4. Step 4:

    python erds_exome.py discover
    --params params.txt \
    --datafile $RPKM_matrix.raw.filtered.SVD \
    --sample NA12878 \
    --vcf=$snv_vcf_file \
    --hetsnp True \
    --tagsnp Ture \
    --tagsnp_file $tagsnp_file \
    --output NA12878.pooled.Het.Tag.cnv
   

#Contact [email protected]