A flexible simulation toolkit for designing and evaluating ChIP-sequencing experiments

Abstract

A major challenge in evaluating quantitative ChIP-seq analyses, such as peak calling and differential binding, is a lack of reliable ground truth data. We present Tulip, a toolkit for rapidly simulating ChIP-seq data using statistical models of the experimental steps. Tulip may be used for a range of applications, including power analysis for experimental design, benchmarking of analysis tools, and modeling effects of processes such as replication on ChIP-seq signals.