A statistical framework for modeling gene expression using chromatin features and application to modENCODE datasets

Cheng, Chao; Yan, Koon-Kiu; Yip, Kevin Y.; Rozowsky, Joel; Alexander, Roger P.; Shou, Chong; Gerstein, Mark

doi:10.1186/gb-2011-12-2-r15

Cited by 134 publications

(173 citation statements)

References 63 publications

Supporting

Mentioning

154

Contrasting

Order By: Relevance

“…Previous work has shown that prediction of factor-responsive targets from an individual ChIP-seq experiment can be improved by incorporating additional information, such as co-correlated expression across hundreds of different microarray studies (Lai et al 2010;Cheng et al 2011;Marbach et al 2012). Here, we show that by integrating 98 ChIP-seq data sets for 57 transcription factors, we can identify low-complexity binding sites that significantly improve the fraction of factor-responsive targets.…”

Section: Low-complexity Targets Correlate Better With Factor-responsimentioning

confidence: 78%

“…Compendia of ChIP-seq data sets can be used to construct regulatory networks and to find novel pairs of transcription factors with similar sets of bound targets that suggest new cases of transcription factor co-association (Niu et al 2011;Gerstein et al 2012). Additionally, in combination with data sets describing histone modifications and other measures of chromatin state, such compendia can be used to predict gene expression (Cheng et al 2011Marbach et al 2012). These examples illustrate how combining data for many transcription factors can provide emergent insights about gene regulation that cannot be found by studying one factor by itself.…”

Section: [Supplemental Materials Is Available For This Article]mentioning

confidence: 99%

See 1 more Smart Citation

Integrative analysis of C. elegans modENCODE ChIP-seq data sets to infer gene regulatory interactions

Nostrand

Kim

2013

Genome Res.

View full text Add to dashboard Cite

The C. elegans modENCODE Consortium has defined in vivo binding sites for a large array of transcription factors by ChIPseq. In this article, we present examples that illustrate how this compendium of ChIP-seq data can drive biological insights not possible with analysis of individual factors. First, we analyze the number of independent factors bound to the same locus, termed transcription factor complexity, and find that low-complexity sites are more likely to respond to altered expression of a single bound transcription factor. Next, we show that comparison of binding sites for the same factor across developmental stages can reveal insight into the regulatory network of that factor, as we find that the transcription factor UNC-62 has distinct binding profiles at different stages due to distinct cofactor co-association as well as tissue-specific alternative splicing. Finally, we describe an approach to infer potential regulators of gene expression changes found in profiling experiments (such as DNA microarrays) by screening these altered genes to identify significant enrichment for targets of a transcription factor identified in ChIP-seq data sets. After confirming that this approach can correctly identify the upstream regulator on expression data sets for which the regulator was previously known, we applied this approach to identify novel candidate regulators of transcriptional changes with age. The analysis revealed nine candidate aging regulators, of which three were previously known to have a role in longevity. We experimentally showed that two of the new candidate aging regulators can extend lifespan when overexpressed, indicating that this approach can identify novel functional regulators of complex processes.

show abstract

Section: Low-complexity Targets Correlate Better With Factor-responsimentioning

confidence: 78%

Section: [Supplemental Materials Is Available For This Article]mentioning

confidence: 99%

Integrative analysis of C. elegans modENCODE ChIP-seq data sets to infer gene regulatory interactions

Nostrand

Kim

2013

Genome Res.

View full text Add to dashboard Cite

show abstract

“…Recently, additional methods have been developed to predict expression from histone ChIP-seq [22][23][24][25] . These methods were used to predict RNA-seq-based expression levels, using a large number of histone ChIP-seq data sets as input.…”

Section: Npgmentioning

confidence: 99%

Uniform, optimal signal processing of mapped deep-sequencing data

et al. 2013

View full text Add to dashboard Cite

“…We have previously shown that histone modifications are strong indicators of expression levels [46,47]. Therefore, we next explored the relationship between DNA methylation and histone modifications in terms of indicating gene expression, and tested whether information on gene expression conveyed by DNA methylation is totally subsumed by that of histone modifications.…”

Section: Quantitative Relationship With Histone Modificationsmentioning

confidence: 99%

Whole-genome bisulfite sequencing of multiple individuals reveals complementary roles of promoter and gene body methylation in transcriptional regulation

et al. 2014

Self Cite

View full text Add to dashboard Cite

Background: DNA methylation is an important type of epigenetic modification involved in gene regulation. Although strong DNA methylation at promoters is widely recognized to be associated with transcriptional repression, many aspects of DNA methylation remain not fully understood, including the quantitative relationships between DNA methylation and expression levels, and the individual roles of promoter and gene body methylation. Results: Here we present an integrated analysis of whole-genome bisulfite sequencing and RNA sequencing data from human samples and cell lines. We find that while promoter methylation inversely correlates with gene expression as generally observed, the repressive effect is clear only on genes with a very high DNA methylation level. By means of statistical modeling, we find that DNA methylation is indicative of the expression class of a gene in general, but gene body methylation is a better indicator than promoter methylation. These findings are general in that a model constructed from a sample or cell line could accurately fit the unseen data from another. We further find that promoter and gene body methylation have minimal redundancy, and either one is sufficient to signify low expression. Finally, we obtain increased modeling power by integrating histone modification data with the DNA methylation data, showing that neither type of information fully subsumes the other. Conclusion: Our results suggest that DNA methylation outside promoters also plays critical roles in gene regulation. Future studies on gene regulatory mechanisms and disease-associated differential methylation should pay more attention to DNA methylation at gene bodies and other non-promoter regions.

show abstract

A statistical framework for modeling gene expression using chromatin features and application to modENCODE datasets

Cited by 134 publications

References 63 publications

Integrative analysis of C. elegans modENCODE ChIP-seq data sets to infer gene regulatory interactions

Integrative analysis of C. elegans modENCODE ChIP-seq data sets to infer gene regulatory interactions

Uniform, optimal signal processing of mapped deep-sequencing data

Whole-genome bisulfite sequencing of multiple individuals reveals complementary roles of promoter and gene body methylation in transcriptional regulation

Contact Info

Product

Resources

About