Refined DNase-seq protocol and data analysis reveals intrinsic bias in transcription factor footprint identification

Abstract

DNase-seq is a powerful technique for identifying cis-regulatory elements across the genome. We studied the key experimental parameters to optimize the performance of DNase-seq. We found that sequencing short 50-100bp fragments that accumulate in long inter-nucleosome linker regions is more efficient for identifying transcription factor binding sites than using longer fragments. We also assessed the potential of DNase-seq to predict transcription factor occupancy through the generation of nucleotide-resolution transcription factor footprints. In modeling the sequence-specific DNaseI cutting bias we found a surprisingly strong effect that varied over more than two orders of magnitude. This confounds DNaseI footprint analysis to the extent that the nucleotide resolution cleavage patterns at most transcription factor binding sites are derived from intrinsic DNaseI cleavage bias rather than from specific protein-DNA interactions. In contrast, quantitative comparison of DNaseI hypersensitivity between states can predict transcription factor occupancy associated with particular biological perturbations.

A synergistic DNA logic predicts genome-wide chromatin accessibility

“…Despite some variation at any individual loci, the SCM model captures the overall structure of DNase I accessibility over the held-out chromosome. DNase-seq is known to have an underlying sequence preference, resulting in the possibility that a SCM model would learn the inherent sequence bias of the DNase I enzyme rather than the relationship between DNA sequence and accessibility (He et al 2013;Lazarovici et al 2013). In order to account for this confounder, we validate our model on DNase I hypersensitive sites (Fig.…”

Modeling cis-regulation with a compendium of genome-wide histone H3K27ac profiles

“…Several genomic techniques, including ChIP-seq (Barski et al 2007;Johnson et al 2007;Mikkelsen et al 2007), DNaseseq (Crawford et al 2006;Hesselberth et al 2009;Boyle et al 2011;He et al 2014), and ATAC-seq (Buenrostro et al 2013), have been developed to experimentally identify cis-regulatory regions genome-wide. Attempts to use these data to understand gene expression have, however, been impeded by the following factors: Data for only a small subset of transcription factors (TFs) participating in any system can be generated in practice (Gerstein et al 2012); not all TF binding sites necessarily play roles in gene regulation; mapping between enhancers and genes is still an open question; the regulatory environment that controls a gene may depend on a complex interaction of many factors at the promoter and enhancers that may act cooperatively or antagonistically (Montavon et al 2011;Spitz and Furlong 2012); and technical biases in chromatin profiling data may obscure biologically relevant signal .…”

“…First, the transcription factor binding sites discovered in most ChIPseq experiments tend to fall within a set of genomic regions that are DNase I-hypersensitive (Hesselberth et al 2009;Neph et al 2012b;Thurman et al 2012;He et al 2014). The union of DNase-seq (UDHS) peaks across a broad array of human cell types can therefore be used to define a superset of transcription factor binding loci in most cell types.…”

Combining transcription factor binding affinities with open-chromatin data for accurate gene expression prediction

“…TFs 25 are known to have important roles in several diseases, e.g. a third of known human 26 developmental disorders are related to deregulated TFs [64]. 27 Several general approaches have been proposed to identify TFs acting as key players 28 in gene regulation depending on the available data: Coexpression analysis combined 29 with computational predictions of TF sequence binding can be used to identify key 30 TFs [18].…”