ABS filter for allele-specific analysis of ChIP-seq data

R package - ABS filter

Identification and removal of low-complexity sites in allele-specific analysis of ChIP-seq data

Sebastian M. Waszak^1,3, Helena Kilpinen^2,3,4, Andreas Gschwind^3,5, Andrea Orioli⁵, Sunil K. Raghav¹, Robert M. Witwicki⁵, Eugenia Migliavacca^3,5, Alisa Yurovsky^2,3,4, Tuuli Lappalainen^2,3,4, Nouria Hernandez⁵, Alexandre Reymond⁵, Emmanouli T. Dermitzakis^2,3,4, and Bart Deplancke^1,3

¹Institute of Bioengineering, School of Life Sciences, Ecole Polytechnique Federale de Lausanne (EPFL), Lausanne, Switzerland
²Department of Genetic Medicine and Development, University of Geneva Medical School, Geneva, Switzerland
³Swiss Institute of Bioinformatics, Lausanne, Switzerland
⁴Institute of Genetics and Genomics in Geneva, University of Geneva, Geneva, Switzerland
⁵Center for Integrative Genomics, Faculty of Biology and Medicine, University of Lausanne, Lausanne, Switzerland

Bioinformatics 2013 Nov 18, doi:10.1093/bioinformatics/btt667

Abstract

The development of high-throughput sequencing technologies has enabled the genome-wide analysis of the impact of genetic variation on molecular phenotypes with a single base pair resolution. However, while powerful, these technologies can also introduce unexpected artifacts into the data leading to significant biases in downstream analyses. We investigated the impact of library amplification bias on the identification of allele-specific molecular events from high-throughput sequencing data derived from chromatin immunoprecipitation assays (ChIP-seq). Using RNA polymerase II ChIP-seq data from six cell lines from the 1000 Genomes Project, we identified putative allele-specific (AS) DNA binding events. We found that many of the sites showing a significant AS effect suffered from an amplification bias, as evidenced by a larger number of clonal reads carrying one of the two alleles. To eliminate such false positive allele-specific DNA binding signals, we devised an amplification-bias detection strategy, which filters out sites with low read complexity as well as sites featuring a significant excess of clonal reads. This method should prove useful for allele-specific analyses involving ChIP-seq and other functional sequencing applications.

ABS filter 1.0

1. Requirements

Unix OS
R >=2.15 (available here)
R package intervals >=0.13.3 (available here)
SAMtools 0.1.19 (available here)

2. Download

R package absfilter (available here)
ABS simulation files (available for 36 bp, 40 bp reads)

3. Usage

Download the R package and install it on the terminal with R CMD INSTALL absfilter_1.0.tar.gz

Open R and load the package with: library(absfilter)

Type ?absfilter and see the 'Examples' section for more details on the usage of the ABS filter.

4. Notes

ChIP-seq reads must be stored in the BAM file format.
SNPs must be stored in a tab-delimited file and the first four columns must correspond to:

Please visit the ABS google group to get news on updates and also feel free to post bugs and questions that might be of general interest for other users. For further information please contact Sebastian Waszak or Bart Deplancke.

License: GNU General Public License (Version 3.0). Last update: 11.12.2013