A generalized linear model for peak calling in ChIP-Seq data

Chromatin immunoprecipitation followed by massively parallel sequencing (ChIP-Seq) has become a routine for detecting genome-wide protein-DNA interaction. The success of ChIP-Seq data analysis highly depends on the quality of peak calling (i.e., to detect peaks of tag counts at a genomic location and evaluate if the peak corresponds to a real protein-DNA interaction event). The challenges in peak calling include (1) how to combine the forward and the reverse strand tag data to improve the power of peak calling and (2) how to account for the variation of tag data observed across different genomic locations. We introduce a new peak calling method based on the generalized linear model (GLMNB) that utilizes negative binomial distribution to model the tag count data and account for the variation of background tags that may randomly bind to the DNA sequence at varying levels due to local genomic structures and sequence contents. We allow local shifting of peaks observed on the forward and the reverse stands, such that at each potential binding site, a binding profile representing the pattern of a real peak signal is fitted to best explain the observed tag data with maximum likelihood. Our method can also detect multiple peaks within a local region if there are multiple binding sites in the region.     

Shape-based peak identification for ChIP-Seq

Background  

The identification of binding targets for proteins using ChIP-Seq has gained popularity as an alternative to ChIP-chip. Sequencing can, in principle, eliminate artifacts associated with microarrays, and cheap sequencing offers the ability to sequence deeply and obtain a comprehensive survey of binding. A number of algorithms have been developed to call "peaks" representing bound regions from mapped reads. Most current algorithms incorporate multiple heuristics, and despite much work it remains difficult to accurately determine individual peaks corresponding to distinct binding events.     

Impact of Chromatin Structures on DNA Processing for Genomic Analyses

Chromatin has an impact on recombination, repair, replication, and evolution of DNA. Here we report that chromatin structure also affects laboratory DNA manipulation in ways that distort the results of chromatin immunoprecipitation (ChIP) experiments. We initially discovered this effect at the Saccharomyces cerevisiae HMR locus, where we found that silenced chromatin was refractory to shearing, relative to euchromatin. Using input samples from ChIP-Seq studies, we detected a similar bias throughout the heterochromatic portions of the yeast genome. We also observed significant chromatin-related effects at telomeres, protein binding sites, and genes, reflected in the variation of input-Seq coverage. Experimental tests of candidate regions showed that chromatin influenced shearing at some loci, and that chromatin could also lead to enriched or depleted DNA levels in prepared samples, independently of shearing effects. Our results suggested that assays relying on immunoprecipitation of chromatin will be biased by intrinsic differences between regions packaged into different chromatin structures - biases which have been largely ignored to date. These results established the pervasiveness of this bias genome-wide, and suggested that this bias can be used to detect differences in chromatin structures across the genome.     

Whole-genome sequencing and variant discovery in C. elegans

Massively parallel sequencing instruments enable rapid and inexpensive DNA sequence data production. Because these instruments are new, their data require characterization with respect to accuracy and utility. To address this, we sequenced a Caernohabditis elegans N2 Bristol strain isolate using the Solexa Sequence Analyzer, and compared the reads to the reference genome to characterize the data and to evaluate coverage and representation. Massively parallel sequencing facilitates strain-to-reference comparison for genome-wide sequence variant discovery. Owing to the short-read-length sequences produced, we developed a revised approach to determine the regions of the genome to which short reads could be uniquely mapped. We then aligned Solexa reads from C. elegans strain CB4858 to the reference, and screened for single-nucleotide polymorphisms (SNPs) and small indels. This study demonstrates the utility of massively parallel short read sequencing for whole genome resequencing and for accurate discovery of genome-wide polymorphisms.     

Investigating repetitively matching short sequencing reads: The enigmatic nature of H3K9me3

Most histone modifications can easily be characterized as either activating or repressive. For example, histone3, lysine 4 trimethylation (H3K4me3) is generally considered a distinct sign of actively transcribed promoters while H3K27me3 is generally found at repressed genes. This is not the case for H3K9me3, the subject of this communication, which is a modification that has traditionally been considered a mark of constitutive heterochromatin, but has also been found in significant levels in expressed genes. We therefore sought to use new high-throughput genome-wide maps of H3K9me3 localization to investigate the conflicting hypotheses concerning the nature of this modification. Before we could accurately analyze the locations of H3K9me3 along the genome, and especially in repetitive locations, we developed a method for accurately utilizing short sequencing reads that do not map uniquely to a location in the genome. Investigating the locations of H3K9me3 along the genome allowed us to determine that, while there are high levels of H3K9me3 outside of genes, this modification is not absent from genes. Therefore, we suggest that H3K9me3 may have a role in chromatin organization rather than being directly related to gene expression. In addition, we have found that there is a need to include repetitively matching reads in any high-throughput sequencing experiment.     

Picking ChIP-seq peak detectors for analyzing chromatin modification experiments

Numerous algorithms have been developed to analyze ChIP-Seq data. However, the complexity of analyzing diverse patterns of ChIP-Seq signals, especially for epigenetic marks, still calls for the development of new algorithms and objective comparisons of existing methods. We developed Qeseq, an algorithm to detect regions of increased ChIP read density relative to background. Qeseq employs critical novel elements, such as iterative recalibration and neighbor joining of reads to identify enriched regions of any length. To objectively assess its performance relative to other 14 ChIP-Seq peak finders, we designed a novel protocol based on Validation Discriminant Analysis (VDA) to optimally select validation sites and generated two validation datasets, which are the most comprehensive to date for algorithmic benchmarking of key epigenetic marks. In addition, we systematically explored a total of 315 diverse parameter configurations from these algorithms and found that typically optimal parameters in one dataset do not generalize to other datasets. Nevertheless, default parameters show the most stable performance, suggesting that they should be used. This study also provides a reproducible and generalizable methodology for unbiased comparative analysis of high-throughput sequencing tools that can facilitate future algorithmic development.