From PileLine

Revision as of 18:56, 29 December 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/pilelinetools/

Main Features

PileLine commands and accepted input files.

1. Quick filtering and search within GP files without indexing steps.
2. GP files comparisons.
3. 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.
4. SIFT, PolyPhen-2 and Firestar compatible inputs to facilitate the biological interpretation of huge lists of variants.
5. Genotyping quality control functionality to estimate performance metrics (Harismendi et al. 2009) on detecting homo/heterozigote variants against a given gold standard genotype.
6. Modular design to facilitate the inclusion of new functionalities.

Getting started

New to PileLine? Please, follow our Quick Start.

PileLine Commands

Processing and Annotation Commands

• pileline-fastseek

Prints a given range of a GP file.

• pileline-sort

Sorts GP files by coordinate.

• pileline-fastjoin

Joins two SORTED GP files.

• pileline-rfilter

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-pileup2sift

Generates SIFT compatible infiles from pileup files.

• pileline-pileup2polyphen

Generates PolyPhen-2 compatible infiles from pileup files.

• pileline-pileup2firestar

Generates Firestar compatible infiles from GP files.

Analysis Commands

• pileline-2smc

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

Compares n samples reporting consistent variants.

• pileline-genotest

Calculates the NGS performance on genotyping, surveying a set of genomic positions whose genotype is known in the sample. This functionality requires a previous step using *pileline-genindex command for genome indexing.

Use Cases

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

• Perform 2 samples comparison

pileline-2smc.sh 
–a <file_A.pileup> –b <file_B.pileup>
–v <variants_file_A.pileup> –w <variants_file_B.pileup> 
–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>

• Generate column compatible to SIFT intput

pileline-pileup2sift.sh -i <file.pileup>

• Perform a genotyping test for quality control

Genotest metrics table description. It may be obtained by using --print-help-table argument.

# Warning: Check that your alleles in the <gold_genotype.sorted> file are expressed in the same strand as the 
#          reference genome sequence used in your NGS experiment. Typically forward (+) strand. 

## Step1.

#Create reference index <ref_genome.pileline> using pileline-genindex command.
pileline-genindex --index -i  <ref_genome.pileline> -g <ref_genome.fa>

## Step2.
#Create genotest file (required).
pileline-genotest --create-genotest-file <experiment.genotest> –p <GP_file.txt> –g <gold_genotype.sorted> -r <ref_genome.pileline>

## Step3. QC analysis.
 #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

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

PileLine GUI

General scheme of the PileLine GUI software.

PileLine GUI is a front-end of the PileLine toolkit, plus a genome browser. With this tool you can run via an intuitive graphical desktop application:

1. Processing commands of GP Files, like join, annotation, filtering.
2. Perform 2-samples and n-samples somatic mutation calling (via the PileLine 2smc and nsmc commands).
3. Browse GPFiles in a interactive local genome browser.

You can download PileLine GUI from here.