ABSTRACT: BACKGROUND: The detection of enriched DNA or RNA fragments by tiling microarrays has become more and more popular. These microarrays contain a high number of small probes covering genomic loci. However, to achieve high coverage the probe sequences cannot be selected for their hybridization properties. The affinity of the probes towards their targets varies in a sequence-dependent manner. In order to remove this bias a number of approaches have been developed and shown to increase the detection of enriched DNA or RNA fragments. However, these approaches also employ a peak detection algorithm that is different from the one used previously. Thus, it seems possible that the enhancement of detection is due to the peak detection algorithm rather than the sequence-dependent normalization. RESULTS: We compared three different sequence-dependent probe level normalization procedures to a naive sequence-independent normalization technique. In order to achieve maximal comparability, we used the normalized intensity values as input to a single peak detection algorithm. A so-called "spike-in" data set served as benchmark for the performance. We will show that the sequence-dependent normalization procedures do not perform better than the naive approach, suggesting that the benefit of using these normalization approaches is limited. Furthermore, we will show that the naive approach does well, because it effectively removes the sequence-dependent component of the measured intensities with the help of the control hybridization experiment. CONCLUSIONS: Sequence-dependent normalization of microarray data hardly improves the detection of enriched DNA or RNA fragments. The "success" of the sequence-independent naive approach is only possible due to the control experiment and requires proper scaling of the measured intensities.

The formation of transcription-factor-binding sites is an important evolutionary process. Here, we show that methylation and deamination of CpG dinucleotides generate in vivo p53-binding sites in numerous Alu elements and in non-repetitive DNA in a species-specific manner. In light of this, we propose that the deamination of methylated CpGs constitutes a universal mechanism for de novo generation of various transcription-factor-binding sites in Alus.

Eukaryotic DNA is organized into a macromolecular structure called chromatin. The basic repeating unit of chromatin is the nucleosome, which consists of two copies of each of the four core histones and DNA. The nucleosomal organization and the positions of nucleosomes have profound effects on all DNA-dependent processes. Understanding the factors that influence nucleosome positioning is therefore of general interest. Among the many determinants of nucleosome positioning, the DNA sequence has been proposed to have a major role. Here, we analyzed more than 860,000 nucleosomal DNA sequences to identify sequence features that guide the formation of nucleosomes in vivo. We found that both a periodic enrichment of AT base pairs and an out-of-phase oscillating enrichment of GC base pairs as well as the overall preference for GC base pairs are determinants of nucleosome positioning. The preference for GC pairs can be related to a lower energetic cost required for deformation of the DNA to wrap around the histones. In line with this idea, we found that only incorporation of both signal components into a sequence model for nucleosome formation results in maximal predictive performance on a genome-wide scale. In this manner, one achieves greater predictive power than published approaches. Our results confirm the hypothesis that the DNA sequence has a major role in nucleosome positioning in vivo.

We have determined diversities exceeding 10(12) different sequences in an annealing and melting assay using synthetic randomized oligonucleotides as a standard. For such high diversities, the annealing kinetics differ from those observed for low diversities, favouring the remelting curve after annealing as the best indicator of complexity. Direct comparisons of nucleic acid pools obtained from an aptamer selection demonstrate that even highly complex populations can be evaluated by using DiStRO, without the need of complicated calculations.

Bicoid (Bcd) is the anterior determinant in Drosophila. Accordingly, loss of Bcd causes loss of head and thorax and their replacement with posterior structures. bcd mRNA is maternally deposited at the anterior pole and Bcd forms an anterior-to-posterior (AP) concentration gradient. The expression of a series of zygotic head genes is thought to be differentially regulated by distinct threshold concentrations of the Bcd gradient. Thereby Bcd functions as a morphogen, instructing fields of cells to take on specific fates. Here, we show that spatial limits of anterior genes are also set in the absence of a Bcd gradient and depend on factors of the maternal terminal system. The receptor tyrosine kinase Torso (Tor), a key component of this system, is active in the pole regions of the embryo. Its activity downregulates the maternally deposited repressor Capicua (Cic), leaving high Cic activity in the central regions and decreasingly lower Cic activities toward the poles. We show that the positions of posterior boundaries of Bcd target genes are dependent not only on Bcd, but also on Tor-mediated Cic activity. The results indicate that Cic can mediate repression through distinct binding sites within a Bcd responsive enhancer and that gene activation by Bcd is antagonized by Cic. The activating and repressive effects of Bcd and Cic, respectively, are integrated by the Bcd target gene enhancer. We conclude that the spatial domains of head gene expression are determined by Bcd in concert with Tor-dependent repressors.

Motif overrepresentation analysis of proximal promoters is a common approach to characterize the regulatory properties of co-expressed sets of genes. Here we show that these approaches perform well on mammalian CpG-depleted promoter sets that regulate expression in terminally differentiated tissues such as liver and heart. In contrast, CpG-rich promoters show very little overrepresentation signal, even when associated with genes that display highly constrained spatiotemporal expression. For instance, while approximately 50% of heart specific genes possess CpG-rich promoters we find that the frequently observed enrichment of MEF2-binding sites upstream of heart-specific genes is solely due to contributions from CpG-depleted promoters. Similar results are obtained for all sets of tissue-specific genes indicating that CpG-rich and CpG-depleted promoters differ fundamentally in their distribution of regulatory inputs around the transcription start site. In order not to dilute the respective transcription factor binding signals, the two promoter types should thus be treated as separate sets in any motif overrepresentation analysis.

Animal genomes possess highly conserved cis-regulatory sequences that are often found near genes that regulate transcription and development. Researchers have proposed that the strong conservation of these sequences may affect the evolution of the surrounding genome, both by repressing rearrangement, and possibly by promoting duplicate gene retention. Conflicting data, however, have made the validity of these propositions unclear. Here, we use a new computational method to identify phylogenetically conserved non-coding elements (PCNEs) in a manner that is not biased by rearrangement and duplication. This method is powerful enough to identify more than a thousand PCNEs that have been conserved between vertebrates and the basal chordate amphioxus. We test 42 of our PCNEs in transgenic zebrafish assays---including examples from vertebrates and amphioxus---and find that the majority are functional enhancers. We find that PCNEs are enriched around genes with ancient synteny conservation, and that this association is strongest for extragenic PCNEs, suggesting that cis-regulatory interdigitation plays a key role in repressing genome rearrangement. Next, we classify mouse and zebrafish genes according to association with PCNEs, synteny conservation, duplication history, and presence in bidirectional promoter pairs, and use this data to cluster gene functions into a series of distinct evolutionary patterns. These results demonstrate that subfunctionalization of conserved cis-regulation has not been the primary determinate of gene duplicate retention in vertebrates. Instead, the data support the Gene Balance Hypothesis, which proposes that duplicate retention has been driven by selection against dosage imbalances in genes with many protein connections.

ABSTRACT: BACKGROUND: The transcription factor OCT4 is highly expressed in pluripotent embryonic stem cells which are derived from the inner cell mass of mammalian blastocysts. Pluripotency and self renewal are controlled by a transcription regulatory network governed by the transcription factors OCT4, SOX2 and NANOG. Recent studies on reprogramming somatic cells to induced pluripotent stem cells highlight OCT4 as a key regulator of pluripotency. RESULTS: We have carried out an integrated analysis of high-throughput data (ChIP-on-chip and RNAi experiments along with promoter sequence analysis of putative target genes) and identified a core OCT4 regulatory network in human embryonic stem cells consisting of 33 target genes. Enrichment analysis with these target genes revealed that this integrative analysis increases the functional information content by factors of 1.3 - 4.7 compared to the individual studies. In order to identify potential regulatory co-factors of OCT4, we performed a de novo motif analysis. In addition to known validated OCT4 motifs we obtained binding sites similar to motifs recognized by further regulators of pluripotency and development; e.g. the heterodimer of the transcription factors C-MYC and MAX, a prerequisite for C-MYC transcriptional activity that leads to cell growth and proliferation. CONCLUSIONS: Our analysis shows how heterogeneous functional information can be integrated in order to reconstruct gene regulatory networks. As a test case we identified a core OCT4-regulated network that is important for the analysis of stem cell characteristics and cellular differentiation. Functional information is largely enriched using different experimental results. The de novo motif discovery identified well-known regulators closely connected to the OCT4 network as well as potential new regulators of pluripotency and differentiation. These results provide the basis for further targeted functional studies.

Signaling cascades are triggered by environmental stimulation and propagate the signal to regulate transcription. Systematic reconstruction of the underlying regulatory mechanisms requires pathway-targeted, informative experimental data. However, practical experimental design approaches are still in their infancy. Here, we propose a framework that iterates design of experiments and identification of regulatory relationships downstream of a given pathway. The experimental design component, called MEED, aims to minimize the amount of laboratory effort required in this process. To avoid ambiguity in the identification of regulatory relationships, the choice of experiments maximizes diversity between expression profiles of genes regulated through different mechanisms. The framework takes advantage of expert knowledge about the pathways under study, formalized in a predictive logical model. By considering model-predicted dependencies between experiments, MEED is able to suggest a whole set of experiments that can be carried out simultaneously. Our framework was applied to investigate interconnected signaling pathways in yeast. In comparison with other approaches, MEED suggested the most informative experiments for unambiguous identification of transcriptional regulation in this system.

Abstract We present a new formulation of phylogenetic reconstruction named maximum similarity. We describe basic algorithms based on the maximum similarity objective for computing distances between subtrees and for combining two subtrees. We present distance methods for constructing an initial tree and updating the initial tree by incorporating those basic algorithms into the Neighbor Joining (NJ) method and the Nearest-Neighbor Interchange (NNI) framework of the FastME program. The new distance methods have been implemented as a computer program named MS. The time requirement of the MS program is about five times the cost of computing observed sequence distances. The MS program was compared by simulation with four existing programs: NJ, FastME, STC, and Weighbor. Experimental results show that incorporating the maximum similarity objective into existing methods leads to improvements both in topology and in branch length.

SUMMARY: Chromatin immunoprecipitation combined with DNA microarrays (ChIP-chip) has evolved as a popular technique to study DNA-protein-binding or post-translational chromatin/histone modifications at the genomic level. However, the raw microarray intensities generate a massive amount of data, creating a need for efficient analysis algorithms and statistical methods to identified enriched regions. RESULTS: We present a fast, free and powerful, open source R package, rMAT, that allows the identification of regions enriched for transcription factor binding sites in ChIP-chip experiments on Affymetrix tiling arrays. AVAILABILITY: The R-package rMAT is available from the Bioconductor web site at http://bioconductor.org and runs on Linux, MAC OS and MS-Windows. rMAT is distributed under the terms of the Artistic Licence 2.0. CONTACT: arnaud.droit@ircm.qc.ca, raphael.gottardo@ircm.qc.ca.

Transcriptome profiling has become a routine tool in biology. For Arabidopsis, the Affymetrix ATH1 expression array is most commonly used, but it lacks about a third of all annotated genes present in the reference strain. An alternative are tiling arrays, but previous designs have not allowed the simultaneous analysis of both strands on a single array. We introduce AGRONOMICS1, a new Affymetrix Arabidopsis microarray that contains the complete paths of both genome strands with on average one 25mer probe per 35 bp genome sequence window. In addition, the new AGRONOMICS1 array contains all perfect match (PM) probes from the original ATH1 array, allowing for seamless integration of the very large existing ATH1 knowledge base. The AGRONOMICS1 array can be used for diverse functional genomics application such as reliable expression profiling of more than 30,000 genes, detection of alternative splicing and chromatin immunoprecipitation coupled to microarrays (ChIP-chip). Here, we describe the design of the array and compare its performance to that of the ATH1 array. We found results from both microarrays to be of similar quality, but that AGRONOMICS1 arrays yield robust expression information for many more genes, as expected. Analysis of the ATH1 probes on AGRONOMICS1 arrays produces results that closely mirror those of ATH1 arrays. Finally, the AGRONOMICS1 array is shown to be useful for ChIP-chip experiments. We show that heterochromatic H3K9me2 is strongly confined to the gene body of target genes in euchromatic chromosome regions, suggesting that spreading of heterochromatin is limited outside of pericentromeric regions.

ABSTRACT: BACKGROUND: We have incorporated Bayesian model regularization with biophysical modeling of protein-DNA interactions, and of genome-wide nucleosome positioning to study protein-DNA interactions, using a high-throughput dataset. The newly developed method (BayesPI) includes the estimation of a transcription factor (TF) binding energy matrices, the computation of binding affinity of a TF target site and the corresponding chemical potential. RESULTS: The method was successfully tested on synthetic ChIP-chip datasets, real yeast ChIP-chip experiments. Subsequently, it was used to estimate condition-specific and species-specific protein-DNA interaction for several yeast TFs. CONCLUSIONS: The results revealed that the modification of the protein binding parameters and the variation of the individual nucleotide affinity in either recognition or flanking sequences occurred under different stresses and in different species. The findings suggest that such modifications may be adaptive and play roles in the formation of the environment-specific binding patterns of yeast TFs and in the divergence of TF binding sites across the related yeast species.

ABSTRACT: BACKGROUND: Chromatin immunoprecipitation on tiling arrays (ChIP-chip) has been employed to examine features such as protein binding and histone modifications on a genome-wide scale in a variety of cell types. Array data from the latter studies typically have a high proportion of enriched probes whose signals vary considerably (due to heterogeneity in the cell population), and this makes their normalization and downstream analysis difficult. RESULTS: Here we present strategies for analyzing such experiments, focusing our discussion on the analysis of Bromodeoxyruridine (BrdU) immunoprecipitation on tiling array (BrdU-IP-chip) datasets. BrdU-IP-chip experiments map large, recently replicated genomic regions and have similar characteristics to histone modification/location data. To prepare such data for downstream analysis we employ a dynamic programming algorithm that identifies a set of putative unenriched probes, which we use for both within-array and between-array normalization. We also introduce a second dynamic programming algorithm that incorporates a priori knowledge to identify and quantify positive signals in these datasets. CONCLUSIONS: Highly enriched IP-chip datasets are often difficult to analyze with traditional array normalization and analysis strategies. Here we present and test a set of analytical tools for their normalization and quantification that allows for accurate identification and analysis of enriched regions.

ABSTRACT: BACKGROUND: High-throughput sequencing technology has become popular and widely used to study protein and DNA interactions. Chromatin immunoprecipitation, followed by sequencing of the resulting samples, produces large amounts of data that can be used to map transcription factor binding sites and histone modifications. METHOD: Our proposed statistical algorithm, BayesPeak, uses a fully Bayesian hidden Markov model to detect enriched locations in the genome. The structure accommodates the natural features of the Solexa/Illumina sequencing data and allows for overdispersion in the abundance of reads in different regions. Moreover, a control sample can be incorporated in the analysis to account for experimental and sequence biases. Markov chain Monte Carlo algorithms are applied to estimate the posterior distributions of the model parameters, and posterior probabilities are used to detect the sites of interest. CONCLUSIONS: We have presented a flexible approach for identifying peaks from ChIP-seq reads, suitable for use on both transcription factor binding and histone modification data. Our method estimates probabilities of enrichment that can be used in downstream analysis. The method is assessed using experimentally verified data and is shown to provide high-confidence calls with low false positive rates.

Here we describe an adapted ChIP-on-chip protocol for the analysis of DNA topoisomerase chromosomal binding in Saccharomyces cerevisiae cells. The ChIP-on-chip technique is based on the immunoprecipitation of crosslinked chromatin (ChIP, chromatin immunoprecipitation), followed by DNA amplification and hybridization to high-density oligonucleotide arrays (Chip). Comparison of the signal intensities of immunoprecipitated and control fractions provides a measurement of the protein-DNA association along entire genomes. ChIP-on-chip analysis of DNA topoisomerase binding to chromosomal DNA opens a window to the understanding of the in vivo contribution of these enzymes to the different DNA transactions taking place concomitantly within the context of the highly organized eukaryotic genome. Chromosomal binding profiles obtained from synchronized cells allow scoring the temporal and spatial restriction of these enzymes at different cell cycle stages. By using this approach, novel aspects of DNA topoisomerase function in chromosome metabolism might be unmasked.

In order to fully understand the functions of a DNA-binding protein it is necessary to identify all of its binding sites in chromosomes and assess the role of each site in the overall biological function of the factor. An approach ChIP-on-Chip which combines the chromatin immunoprecipitation technique with chromosomal DNA microarray analysis, has proven to be a powerful means for the chromosome-wide identification of protein binding sites. This approach can also be used to characterize chromosome-wide variations in patterns of post-translational protein modifications, for example histone modifications. This chapter presents methodologies for the ChIP-on-Chip analysis, using as an example the identification of chromosome-wide binding sites for the TATA-binding protein in mitotic cells.

MOTIVATION: ChIP-chip has been widely used for various genome-wide biological investigations. Given the small number of replicates (typically 2-3) per biological sample, methods of analysis that control the variance are desirable but in short supply. We propose a Double Error Shrinkage (DES) method by using moving average statistics based on local-pooled error estimates which effectively control both heterogeneous error variances and correlation structures of an extremely large number of individual probes on tiling arrays. RESULTS: Applying DES to ChIP-chip tiling array study for discovering genome-wide protein-binding sites, we identified 8,400 target regions that include highly likely TFIID binding sites. About 33% of these were well matched with the known transcription starting sites on the DBTSS library, while many other newly-identified sites have a high chance to be real binding sites based on a high positive predictive value of DES. We also showed the superior performance of DES compared to other commonly-used methods for detecting actual protein binding sites. CONTACT: tspark@snu.ac.kr; jaeklee@virginia.edu.

Transcriptional regulation is largely enacted by transcription factors (TFs) binding DNA. Large numbers of TF binding motifs have been revealed by ChIP-chip experiments followed by computational DNA motif discovery. However, the success of motif discovery algorithms has been limited when applied to sequences bound in vivo (such as those identified by ChIP-chip) because the observed TF-DNA interactions are not necessarily direct: some TFs predominantly associate with DNA indirectly through protein partners, while others exhibit both direct and indirect binding. Here, we present the first method for distinguishing between direct and indirect TF-DNA interactions, integrating in vivo TF binding data, in vivo nucleosome occupancy data, and motifs from in vitro protein binding microarray experiments. When applied to yeast ChIP-chip data, our method reveals that only 48% of the data sets can be readily explained by direct binding of the profiled TF, while 16% can be explained by indirect DNA binding. In the remaining 36%, none of the motifs used in our analysis was able to explain the ChIP-chip data, either because the data were too noisy or because the set of motifs was incomplete. As more in vitro TF DNA binding motifs become available, our method could be used to build a complete catalog of direct and indirect TF-DNA interactions. Our method is not restricted to yeast or to ChIP-chip data, but can be applied in any system for which both in vivo binding data and in vitro DNA binding motifs are available.

Chromosome conformation capture (3C) methodology was developed to study spatial organization of long genomic regions in living cells. Briefly, chromatin is fixed with formaldehyde in vivo to cross-link interacting sites, digested with a restriction enzyme and ligated at a low DNA concentration so that ligation between cross-linked fragments is favored over ligation between random fragments. Ligation products are then analyzed and quantified by PCR. So far, semi-quantitative PCR methods were widely used to estimate the ligation frequencies. However, it is often important to estimate the ligation frequencies more precisely which is only possible by using the real-time PCR. At the same time, it is equally necessary to monitor the specificity of PCR amplification. That is why the real-time PCR with TaqMan probes is becoming more and more popular in 3C studies. In this chapter, we describe the general protocol for 3C analysis with the subsequent estimation of ligation frequencies by using the real-time PCR technology with TaqMan probes. We discuss in details all steps of the experimental procedure paying special attention to weak points and possible ways to solve the problems. A special attention is also paid to the problems in interpretation of the results and necessary control experiments. Besides, in theory, we consider other approaches to analysis of the ligation products used in frames of the so-called 4C and 5C methods. The recently developed chromatin immunoprecipitation (ChIP)-loop assay representing a combination of 3C and ChIP is also discussed.