Methodology article

ChIP on Chip: surprising results are often artifacts

Torsten Waldminghaus and Kirsten Skarstad^*

• * Corresponding author: Kirsten Skarstad kirsten.skarstad@rr-research.no

Author Affiliations

Department of Cell Biology, Institute for Cancer Research, The Norwegian Radium Hospital, Oslo University Hospital and University of Oslo, 0310 Oslo, Norway

For all author emails, please log on.

BMC Genomics 2010, 11:414 doi:10.1186/1471-2164-11-414

Published: 5 July 2010

Abstract

Background

The method of chromatin immunoprecipitation combined with microarrays (ChIP-Chip) is a powerful tool for genome-wide analysis of protein binding. However, a high background signal is a common phenomenon.

Results

Reinvestigation of the chromatin immunoprecipitation procedure led us to discover four causes of high background: i) non-unique sequences, ii) incomplete reversion of crosslinks, iii) retention of protein in spin-columns and iv) insufficient RNase treatment. The chromatin immunoprecipitation method was modified and applied to analyze genome-wide binding of SeqA and σ³² in Escherichia coli.

Conclusions

False positive findings originating from these shortcomings of the method could explain surprising and contradictory findings in published ChIP-Chip studies. We present a modified chromatin immunoprecipitation method greatly reducing the background signal.