A practical comparison of methods for detecting transcription factor binding sites in ChIP-seq experiments

Chromatin immunoprecipitation coupled with massively parallel sequencing (ChIP-seq) is increasingly being applied to study transcriptional regulation on a genome-wide scale. While numerous algorithms have recently been proposed for analysing the large ChIP-seq datasets, their relative merits and pot...

Full description

Saved in:
Bibliographic Details
Published in:BMC genomics 2009-12, Vol.10 (1), p.618-618, Article 618
Main Authors: Laajala, Teemu D, Raghav, Sunil, Tuomela, Soile, Lahesmaa, Riitta, Aittokallio, Tero, Elo, Laura L
Format: Article
Language:English
Subjects:
Algorithms
Binding Sites
Binding sites (Biochemistry)
Chromatin
Chromatin Immunoprecipitation - methods
Computational Biology - methods
Computer applications
Data processing
DNA sequencing
Experiments
Gene regulation
Genetic algorithms
Genetic aspects
Genomes
Genomics
Immunoprecipitation
Methods
Nucleotide sequencing
Physiological aspects
Proteins
R&D
Research & development
Research article
Sequence Analysis, DNA - methods
Software
Studies
Transcription factors
Transcription Factors - metabolism
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Chromatin immunoprecipitation coupled with massively parallel sequencing (ChIP-seq) is increasingly being applied to study transcriptional regulation on a genome-wide scale. While numerous algorithms have recently been proposed for analysing the large ChIP-seq datasets, their relative merits and potential limitations remain unclear in practical applications. The present study compares the state-of-the-art algorithms for detecting transcription factor binding sites in four diverse ChIP-seq datasets under a variety of practical research settings. First, we demonstrate how the biological conclusions may change dramatically when the different algorithms are applied. The reproducibility across biological replicates is then investigated as an internal validation of the detections. Finally, the predicted binding sites with each method are compared to high-scoring binding motifs as well as binding regions confirmed in independent qPCR experiments. In general, our results indicate that the optimal choice of the computational approach depends heavily on the dataset under analysis. In addition to revealing valuable information to the users of this technology about the characteristics of the binding site detection approaches, the systematic evaluation framework provides also a useful reference to the developers of improved algorithms for ChIP-seq data.
ISSN:1471-2164
1471-2164
DOI:10.1186/1471-2164-10-618