Summary ChIP‐chip (or ChIP‐on‐chip) is a technology for isolation and identification of genomic sites occupied by specific DNA‐binding proteins in living cells. The ChIP‐chip signals can be obtained over the whole genome by tiling arrays, where a peak shape is generally observed around a protein‐binding site. In this article, we describe the ChIP‐chip process and present a probability model for ChIP‐chip data. We then propose a model‐based method for recognizing the peak shapes for the purpose of detecting protein‐binding sites. We also investigate the issue of bandwidth in nonparametric kernel smoothing method.