Modeling interactions between adjacent nucleosomes improves genome-wide predictions of nucleosome occupancy

Theory of Active Chromatin Remodeling

Nucleosome positioning controls the accessible regions of chromatin and plays essential roles in DNA-templated processes. ATP driven remodeling enzymes are known to be crucial for its establishment in vivo, but their nonequilibrium nature has hindered the development of a unified theoretical framework for nucleosome positioning. Using a perturbation theory, we show that the effect of these enzymes can be well approximated by effective equilibrium models with rescaled temperatures and interactions. Numerical simulations support the accuracy of the theory in predicting both kinetic and steady-state quantities, including the effective temperature and the radial distribution function, in biologically relevant regimes. The energy landscape view emerging from our study provides an intuitive understanding for the impact of remodeling enzymes in either reinforcing or overwriting intrinsic signals for nucleosome positioning, and may help improve the accuracy of computational models for its prediction in silico.

A Modeling Framework for Generation of Positional and Temporal Simulations of Transcriptional Regulation

We present a modeling framework aimed at capturing both the positional and temporal behavior of transcriptional regulatory proteins in eukaryotic cells. There is growing evidence that transcriptional regulation is the complex behavior that emerges not solely from the individual components, but rather from their collective behavior, including competition and cooperation. Our framework describes individual regulatory components using generic action oriented descriptions of their biochemical interactions with a DNA sequence. All the possible actions are based on the current state of factors bound to the DNA. We developed a rule builder to automatically generate the complete set of biochemical interaction rules for any given DNA sequence. Off-the-shelf stochastic simulation engines can model the behavior of a system of rules and the resulting changes in the configuration of bound factors can be visualized. We compared our model to experimental data at well-studied loci in yeast, confirming that our model captures both the positional and temporal behavior of transcriptional regulation.

Nucleosome Presence at AML-1 Binding Sites Inversely Correlates with Ly49 Expression: Revelations from an Informatics Analysis of Nucleosomes and Immune Cell Transcription Factors

Beyond its role in genomic organization and compaction, the nucleosome is believed to participate in the regulation of gene transcription. Here, we report a computational method to evaluate the nucleosome sensitivity for a transcription factor over a given stretch of the genome. Sensitive factors are predicted to be those with binding sites preferentially contained within nucleosome boundaries and lacking 10 bp periodicity. Based on these criteria, the Acute Myeloid Leukemia-1a (AML-1a) transcription factor, a regulator of immune gene expression, was identified as potentially sensitive to nucleosomal regulation within the mouse Ly49 gene family. This result was confirmed in RMA, a cell line with natural expression of Ly49, using MNase-Seq to generate a nucleosome map of chromosome 6, where the Ly49 gene family is located. Analysis of this map revealed a specific depletion of nucleosomes at AML-1a binding sites in the expressed Ly49A when compared to the other, silent Ly49 genes. Our data suggest that nucleosome-based regulation contributes to the expression of Ly49 genes, and we propose that this method of predicting nucleosome sensitivity could aid in dissecting the regulatory role of nucleosomes in general.

The nucleosome—a large protein complex with DNA wound around it—is the fundamental unit of genomic organization in the eukaryotic cell. More than just a DNA organizer, however, nucleosomes may control gene expression by interfering with the cell’s ability to access the wound-up DNA, as shown by recent research. In this report, we demonstrate a computational method for predicting which elements of the genome are sensitive to regulation by nucleosomes. As a proof-of-concept, we identify AML-1a binding sites—important sequences in DNA regulation—as being specifically nucleosome sensitive. We then show that AML-1a sites are specifically depleted of nucleosomes when a gene is expressed, indicating the ability for nucleosomes to suppress the expression of that gene. This finding confirms that nucleosomes are likely involved in genome regulation, and provides a method for predicting which areas of the genome are probably affected most by nucleosomes. This paper also highlights the usefulness of the Ly49 gene family in testing computer-derived genomic predictions, and is of interest to anyone studying how gene expression is regulated from cell to cell.

Conditions for positioning of nucleosomes on DNA

Positioning of nucleosomes along a eukaryotic genome plays an important role in its organization and regulation. There are many different factors affecting the location of nucleosomes. Some can be viewed as preferential binding of a single nucleosome to different locations along the DNA and some as interactions between neighboring nucleosomes. In this study, we analyze positioning of nucleosomes and derive conditions for their good positioning. Using analytic and numerical approaches we find that, if the binding preferences are very weak, an interplay between the interactions and the binding preferences is essential for a good positioning of nucleosomes, especially on correlated energy landscapes. Analyzing the empirical energy landscape, we conclude that good positioning of nucleosomes in vivo is possible only if they strongly interact. In this case, our model, predicting long-length-scale fluctuations of nucleosomes' occupancy along the DNA, accounts well for the empirical observations.

Affinity, stoichiometry and cooperativity of heterochromatin protein 1 (HP1) binding to nucleosomal arrays

Heterochromatin protein 1 (HP1) participates in establishing and maintaining heterochromatin via its histone-modification-dependent chromatin interactions. In recent papers HP1 binding to nucleosomal arrays was measured in vitro and interpreted in terms of nearest-neighbour cooperative binding. This mode of chromatin interaction could lead to the spreading of HP1 along the nucleosome chain. Here, we reanalysed previous data by representing the nucleosome chain as a 1D binding lattice and showed how the experimental HP1 binding isotherms can be explained by a simpler model without cooperative interactions between neighboring HP1 dimers. Based on these calculations and spatial models of dinucleosomes and nucleosome chains, we propose that binding stoichiometry depends on the nucleosome repeat length (NRL) rather than protein interactions between HP1 dimers. According to our calculations, more open nucleosome arrays with long DNA linkers are characterized by a larger number of binding sites in comparison to chains with a short NRL. Furthermore, we demonstrate by Monte Carlo simulations that the NRL dependent folding of the nucleosome chain can induce allosteric changes of HP1 binding sites. Thus, HP1 chromatin interactions can be modulated by the change of binding stoichiometry and the type of binding to condensed (methylated) and non-condensed (unmethylated) nucleosome arrays in the absence of direct interactions between HP1 dimers.

Regulation of the nucleosome repeat length in vivo by the DNA sequence, protein concentrations and long-range interactions

The nucleosome repeat length (NRL) is an integral chromatin property important for its biological functions. Recent experiments revealed several conflicting trends of the NRL dependence on the concentrations of histones and other architectural chromatin proteins, both in vitro and in vivo, but a systematic theoretical description of NRL as a function of DNA sequence and epigenetic determinants is currently lacking. To address this problem, we have performed an integrative biophysical and bioinformatics analysis in species ranging from yeast to frog to mouse where NRL was studied as a function of various parameters. We show that in simple eukaryotes such as yeast, a lower limit for the NRL value exists, determined by internucleosome interactions and remodeler action. For higher eukaryotes, also the upper limit exists since NRL is an increasing but saturating function of the linker histone concentration. Counterintuitively, smaller H1 variants or non-histone architectural proteins can initiate larger effects on the NRL due to entropic reasons. Furthermore, we demonstrate that different regimes of the NRL dependence on histone concentrations exist depending on whether DNA sequence-specific effects dominate over boundary effects or vice versa. We consider several classes of genomic regions with apparently different regimes of the NRL variation. As one extreme, our analysis reveals that the period of oscillations of the nucleosome density around bound RNA polymerase coincides with the period of oscillations of positioning sites of the corresponding DNA sequence. At another extreme, we show that although mouse major satellite repeats intrinsically encode well-defined nucleosome preferences, they have no unique nucleosome arrangement and can undergo a switch between two distinct types of nucleosome positioning.

Quantifying the role of steric constraints in nucleosome positioning

Statistical positioning, the localization of nucleosomes packed against a fixed barrier, is conjectured to explain the array of well-positioned nucleosomes at the 5′ end of genes, but the extent and precise implications of statistical positioning in vivo are unclear. We examine this hypothesis quantitatively and generalize the idea to include moving barriers as well as nucleosomes actively packed against a barrier. Early experiments noted a similarity between the nucleosome profile aligned and averaged across genes and that predicted by statistical positioning; however, we demonstrate that aligning random nucleosomes also generates the same profile, calling the previous interpretation into question. New rigorous results reformulate statistical positioning as predictions on the variance structure of nucleosome locations in individual genes. In particular, a quantity termed the variance gradient, describing the change in variance between adjacent nucleosomes, is tested against recent high-throughput nucleosome sequencing data. Constant variance gradients provide support for generalized statistical positioning in ∼50% of long genes. Genes that deviate from predictions have high nucleosome turnover and cell-to-cell gene expression variability. The observed variance gradient suggests an effective nucleosome size of 158 bp, instead of the commonly perceived 147 bp. Our analyses thus clarify the role of statistical positioning in vivo.

Toward a unified physical model of nucleosome patterns flanking transcription start sites

Recent genome-wide maps of nucleosome positions in different eukaryotes revealed patterns around transcription start sites featuring a nucleosome-free region flanked by a periodic modulation of the nucleosome density. For Saccharomyces cerevisiae, the average in vivo pattern was previously shown to be quantitatively described by a "nucleosome gas" model based on the statistical positioning mechanism. However, this simple physical description is challenged by the fact that the pattern differs quantitatively between species and by recent experiments that appear incompatible with statistical positioning, indicating important roles for chromatin remodelers. We undertake a data-driven search for a unified physical model to describe the nucleosome patterns of 12 yeast species and also consider an extension of the model to capture remodeling effects. We are led to a nucleosome gas that takes into account nucleosome breathing, i.e., transient unwrapping of nucleosomal DNA segments. This known biophysical property of nucleosomes rationalizes a "pressure"-induced dependence of the effective nucleosome size that is suggested by the data. By fitting this model to the data, we find an average energy cost for DNA unwrapping consistent with previous biophysical experiments. Although the available data are not sufficient to reconstruct chromatin remodeling mechanisms, a minimal model extension by one mechanism yields an "active nucleosome gas" that can rationalize the behavior of systems with reduced histone-DNA ratio and remodeler knockouts. We therefore establish a basis for a physical description of nucleosome patterns that can serve as a null model for sequence-specific effects at individual genes and in models of transcription regulation.

Structural features based genome-wide characterization and prediction of nucleosome organization

Background

Nucleosome distribution along chromatin dictates genomic DNA accessibility and thus profoundly influences gene expression. However, the underlying mechanism of nucleosome formation remains elusive. Here, taking a structural perspective, we systematically explored nucleosome formation potential of genomic sequences and the effect on chromatin organization and gene expression in S. cerevisiae.

Results

We analyzed twelve structural features related to flexibility, curvature and energy of DNA sequences. The results showed that some structural features such as DNA denaturation, DNA-bending stiffness, Stacking energy, Z-DNA, Propeller twist and free energy, were highly correlated with in vitro and in vivo nucleosome occupancy. Specifically, they can be classified into two classes, one positively and the other negatively correlated with nucleosome occupancy. These two kinds of structural features facilitated nucleosome binding in centromere regions and repressed nucleosome formation in the promoter regions of protein-coding genes to mediate transcriptional regulation. Based on these analyses, we integrated all twelve structural features in a model to predict more accurately nucleosome occupancy in vivo than the existing methods that mainly depend on sequence compositional features. Furthermore, we developed a novel approach, named DLaNe, that located nucleosomes by detecting peaks of structural profiles, and built a meta predictor to integrate information from different structural features. As a comparison, we also constructed a hidden Markov model (HMM) to locate nucleosomes based on the profiles of these structural features. The result showed that the meta DLaNe and HMM-based method performed better than the existing methods, demonstrating the power of these structural features in predicting nucleosome positions.

Conclusions

Our analysis revealed that DNA structures significantly contribute to nucleosome organization and influence chromatin structure and gene expression regulation. The results indicated that our proposed methods are effective in predicting nucleosome occupancy and positions and that these structural features are highly predictive of nucleosome organization.

The implementation of our DLaNe method based on structural features is available online.

Nucleosome positioning in Saccharomyces cerevisiae

The DNA of eukaryotic cells is spooled around large histone protein complexes, forming nucleosomes that make up the basis for a high-order packaging structure called chromatin. Compared to naked DNA, nucleosomal DNA is less accessible to regulatory proteins and regulatory processes. The exact positions of nucleosomes therefore influence several cellular processes, including gene expression, chromosome segregation, recombination, replication, and DNA repair. Here, we review recent technological advances enabling the genome-wide mapping of nucleosome positions in the model eukaryote Saccharomyces cerevisiae. We discuss the various parameters that determine nucleosome positioning in vivo, including cis factors like AT content, variable tandem repeats, and poly(dA:dT) tracts that function as chromatin barriers and trans factors such as chromatin remodeling complexes, transcription factors, histone-modifying enzymes, and RNA polymerases. In the last section, we review the biological role of chromatin in gene transcription, the evolution of gene regulation, and epigenetic phenomena.

Statistical-mechanical lattice models for protein-DNA binding in chromatin

Statistical-mechanical lattice models for protein-DNA binding are well established as a method to describe complex ligand binding equilibria measured in vitro with purified DNA and protein components. Recently, a new field of applications has opened up for this approach since it has become possible to experimentally quantify genome-wide protein occupancies in relation to the DNA sequence. In particular, the organization of the eukaryotic genome by histone proteins into a nucleoprotein complex termed chromatin has been recognized as a key parameter that controls the access of transcription factors to the DNA sequence. New approaches have to be developed to derive statistical-mechanical lattice descriptions of chromatin-associated protein-DNA interactions. Here, we present the theoretical framework for lattice models of histone-DNA interactions in chromatin and investigate the (competitive) DNA binding of other chromosomal proteins and transcription factors. The results have a number of applications for quantitative models for the regulation of gene expression.

Learning a weighted sequence model of the nucleosome core and linker yields more accurate predictions in Saccharomyces cerevisiae and Homo sapiens

DNA in eukaryotes is packaged into a chromatin complex, the most basic element of which is the nucleosome. The precise positioning of the nucleosome cores allows for selective access to the DNA, and the mechanisms that control this positioning are important pieces of the gene expression puzzle. We describe a large-scale nucleosome pattern that jointly characterizes the nucleosome core and the adjacent linkers and is predominantly characterized by long-range oscillations in the mono, di- and tri-nucleotide content of the DNA sequence, and we show that this pattern can be used to predict nucleosome positions in both Homo sapiens and Saccharomyces cerevisiae more accurately than previously published methods. Surprisingly, in both H. sapiens and S. cerevisiae, the most informative individual features are the mono-nucleotide patterns, although the inclusion of di- and tri-nucleotide features results in improved performance. Our approach combines a much longer pattern than has been previously used to predict nucleosome positioning from sequence—301 base pairs, centered at the position to be scored—with a novel discriminative classification approach that selectively weights the contributions from each of the input features. The resulting scores are relatively insensitive to local AT-content and can be used to accurately discriminate putative dyad positions from adjacent linker regions without requiring an additional dynamic programming step and without the attendant edge effects and assumptions about linker length modeling and overall nucleosome density. Our approach produces the best dyad-linker classification results published to date in H. sapiens, and outperforms two recently published models on a large set of S. cerevisiae nucleosome positions. Our results suggest that in both genomes, a comparable and relatively small fraction of nucleosomes are well-positioned and that these positions are predictable based on sequence alone. We believe that the bulk of the remaining nucleosomes follow a statistical positioning model.

DNA in eukaryotes is packaged into a chromatin complex, the most basic element of which is the nucleosome. The precise positioning of the nucleosome cores allows for selective access to the DNA, and the mechanisms that control this positioning are important pieces of the gene expression puzzle. In this work, we describe a large-scale DNA sequence pattern that jointly characterizes the sequence preferences of the nucleosome core and the adjacent linkers. We show that this pattern can be used to predict nucleosome positions in both H. sapiens and S. cerevisiae more accurately than previously published methods. The model is most accurate in predicting the most stably positioned nucleosomes, and describes a sequence composition pattern that determines a locally optimal dyad (nucleosomal DNA mid-point) position. In contrast to some previous models, this model is not based primarily on excluding poly-A/T sequences, nor does the model prefer 10 bp periodicity. Our results suggest that local sequence composition is one of many factors that direct the positioning of nucleosomes, while dynamic processes such as transcriptional elongation and the actions of chromatin remodeling complexes also play a significant role in the overall chromatin landscape.

Predicting nucleosome positioning using a duration Hidden Markov Model

Background

The nucleosome is the fundamental packing unit of DNAs in eukaryotic cells. Its detailed positioning on the genome is closely related to chromosome functions. Increasing evidence has shown that genomic DNA sequence itself is highly predictive of nucleosome positioning genome-wide. Therefore a fast software tool for predicting nucleosome positioning can help understanding how a genome's nucleosome organization may facilitate genome function.

Results

We present a duration Hidden Markov model for nucleosome positioning prediction by explicitly modeling the linker DNA length. The nucleosome and linker models trained from yeast data are re-scaled when making predictions for other species to adjust for differences in base composition. A software tool named NuPoP is developed in three formats for free download.

Conclusions

Simulation studies show that modeling the linker length distribution and utilizing a base composition re-scaling method both improve the prediction of nucleosome positioning regarding sensitivity and false discovery rate. NuPoP provides a user-friendly software tool for predicting the nucleosome occupancy and the most probable nucleosome positioning map for genomic sequences of any size. When compared with two existing methods, NuPoP shows improved performance in sensitivity.