From PileLine

Revision as of 08:51, 24 June 2010

Welcome to PileLine Wiki

PileLine is a flexible command-line toolkit for efficient handling, filtering, and comparison of genomic position (GP) files produced by next-generation sequencing experiments (i.e. pileup, BED,GFF, or VCF files). PileLine is designed to be memory efficient by performing on-disk operations over sorted GP files directly.

PileLine is available for downloading at: http://sourceforge.net/projects/pileline

Main Features

1. Filtering and comparison of GP files.
2. Full annotation of GP files with human dbSNP, HGNC Gene Symbol and Ensembl IDs. Custom annotations are also allowed and may be supplied through standard .BED or .GFF files.
3. SIFT and PolyPhen-2 compatible outputs to facilitate the biological interpretation of huge lists of variants.
4. Genotyping quality control functionality to estimate performance metrics (Harismendi et al. 2009) on detecting homo/heterozigote variants against a given gold standard genotype.

PileLine Commands

Processing Commands

• pileline-fastseek.sh

Prints a given range of a GP file.

• pileline-fastsjoin.sh

Joins two positional files.

• pileline-rfilter.sh

Filters (or annotates) a positional file with range-based annotations (in bed format). Each position that is inside of a specific range is annotated.

• pileline-sort.sh

Sorts GP files by coordinate.

• pileline-genindex.sh

Indexes fasta genome and then can perform range based queries in that genome.

Analysis Commands

• pileline-2smc.sh

Looks for discrepancies in genotypes of two samples (i.e.: case vs control). It also can annotate each output position with a user provided BED file containing custom annotations.

• pileline-nsmc.sh

Takes the output of several 2smc comparisons commands to reports where variants are reproduced.

• pileline-genotest.sh

Calculates the NGS performance on genotyping, surveying a set of genomic positions whose genotype is known in the sample.

• pileline-pileup2sift.sh

Generates SIFT compatible outfiles from pileup files.

• pileline-pileup2polyphen.sh'

Generates PolyPhen-2 compatible outfiles from pileup files.

Use Cases

PileLine coupled to SAMtools facilitating pileup handling. NS: non-synonymous

• Perform 2 samples comparison

pileline-2smc.sh 
–a <GPfile_A.txt> –b <GPfile_B.txt>
–v <variants_GPfile_A.txt> –w <variants_GPfile_B.txt> 
–o <out.txt> -d <min_depth>

• Perform n samples comparison

pileline-nsmc.sh
--a-samples<GPfile_a1>,<GPfile_a2>,<GPfile_a3> 
--b-samples <GPfile_b1>,<GPfile_b2>,<GPfile_b3>

• Sort GP files

pileline-sort.sh -i <input_GP_file.txt> -o <outfile.sorted.txt>

• Annotate a GP file with dbSNP

pileline-fastjoin.sh –a <GP_file.txt> -b dbSNP130.txt --left-outer-join

• Annotate a GP file with genes

pileline-rfilter.sh --annotate –A <GP_file.txt> –b <genes.bed>

• Filter pileup to exon loci

pileline-rfilter.sh –A <GP_file.txt> –b <exons.bed>

• Perform a genotyping test for quality control

## Step1. 

#Create genotest file (required).
pileline-genotest --create-genotest-file <experiment.genotest> –p <GP_file.txt> –g <gold_genotype.sorted> -r <ref_genome.pileline>

## Step2. QC analysis.

#Generate a metrics table of performance at a given threshold.
pileline-genotest -a <experiment.genotest> -t <snpq_treshold>

#Generate all performance metrics for several thresholds
pileline-genotest -a <experiment.genotest> --batch-t 0,255,1

#Generate values for ROC curve plot (outfile compatible to ROCR R package)
pileline-genotest -a <experiment.genotest> --roc