CoChat
Sign Up
Cited by 181Genomics and Chromatin DynamicsChromatin Remodeling and CancerCancer-related molecular mechanisms research

The DOI matches its Crossref record; 79 of the 80 references it cites were found in a registry or an index.

An integrative ENCODE resource for cancer genomics

Open access
Zhang, Jing, Lee, Donghoon, Dhiman, Vineet, Jiang, Peng, Xu, Jie, McGillivray, Patrick, Yang, Hongbo, Liu, Jason, Meyerson, William, Clarke, Declan, Gu, Mengting, Li, Shantao, Lou, Shaoke, Xu, Jinrui, Lochovsky, Lucas, Ung, Matthew, Ma, Lijia, Yu, Shan, Cao, Qin, Harmanci, Arif, Yan, Koon-Kiu, Sethi, Anurag, Gürsoy, Gamze, Schoenberg, Michael Rutenberg, Rozowsky, Joel, Warrell, Jonathan, Emani, Prashant, Yang, Yucheng T., Galeev, Timur, Kong, Xiangmeng, Liu, Shuang, Li, Xiaotong, Krishnan, Jayanth, Feng, Yanlin, Rivera-Mulia, Juan Carlos, Adrian, Jessica, Broach, James R, Bolt, Michael, Moran, Jennifer, Fitzgerald, Dominic, Dileep, Vishnu, Liu, Tingting, Mei, Shenglin, Sasaki, Takayo, Trevilla-Garcia, Claudia, Wang, Su, Wang, Yanli, Zang, Chongzhi, Wang, Daifeng, Klein, Robert J., Snyder, Michael, Gilbert, David M., Yip, Kevin, Cheng, Chao, Yue, Feng, Liu, X. Shirley, White, Kevin P., Gerstein, Mark·2020·Nature Communications
DOI 10.1038/s41467-020-14743-wPMC PMC7391744

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

ENCODE comprises thousands of functional genomics datasets, and the encyclopedia covers hundreds of cell types, providing a universal annotation for genome interpretation. However, for particular applications, it may be advantageous to use a customized annotation. Here, we develop such a custom annotation by leveraging advanced assays, such as eCLIP, Hi-C, and whole-genome STARR-seq on a number of data-rich ENCODE cell types. A key aspect of this annotation is comprehensive and experimentally derived networks of both transcription factors and RNA-binding proteins (TFs and RBPs). Cancer, a disease of system-wide dysregulation, is an ideal application for such a network-based annotation. Specifically, for cancer-associated cell types, we put regulators into hierarchies and measure their network change (rewiring) during oncogenesis. We also extensively survey TF-RBP crosstalk, highlighting how SUB1, a previously uncharacterized RBP, drives aberrant tumor expression and amplifies the effect of MYC, a well-known oncogenic TF. Furthermore, we show how our annotation allows us to place oncogenic transformations in the context of a broad cell space; here, many normal-to-tumor transitions move towards a stem-like state, while oncogene knockdowns show an opposing trend. Finally, we organize the resource into a coherent workflow to prioritize key elements and variants, in addition to regulators. We showcase the application of this prioritization to somatic burdening, cancer differential expression and GWAS. Targeted validations of the prioritized regulators, elements and variants using siRNA knockdowns, CRISPR-based editing, and luciferase assays demonstrate the value of the ENCODE resource.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41467-020-14743-w

Cites

79 of 79 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Overview of the ENCODEC resource. Table columns list cell types and rows list assays. Blue table boundary: Cell types with assays in the ENCODE Encyclopedia highlight the breadth of the resource. The large number of cell types allows for comparative analyses between cell-types, as well as cell-type specific analyses. Green table boundary: cell-type specific analyses based on deep annotations of cell lines. The integration of assays allows for high-resolution investigation of genomic biology. Inset: we use annotations from cell-type specific ENCODE assays to build extended gene definitions—coding and non-coding elements that are linked according to their interaction and associated function (top). We relate transcription factors (TFs) and RNA binding proteins (RBPs) in a joint network hierarchy that describes their regulatory potential (middle). By comparing regulatory networks in tumor and normal ENCODE samples, we develop rewiring networks that may relate to regulatory changes that occur in the context of normal-to-tumor transition (bottom).
Fig. 1 Overview of the ENCODEC resource. Table columns list cell types and rows list assays. Blue table boundary: Cell types with assays in the ENCODE Encyclopedia highlight the breadth of the resource. The large number of cell types allows for comparative analyses between cell-types, as well as cell-type specific analyses. Green table boundary: cell-type specific analyses based on deep annotations of cell lines. The integration of assays allows for high-resolution investigation of genomic biology. Inset: we use annotations from cell-type specific ENCODE assays to build extended gene definitions—coding and non-coding elements that are linked according to their interaction and associated function (top). We relate transcription factors (TFs) and RNA binding proteins (RBPs) in a joint network hierarchy that describes their regulatory potential (middle). By comparing regulatory networks in tumor and normal ENCODE samples, we develop rewiring networks that may relate to regulatory changes that occur in the context of normal-to-tumor transition (bottom).CC BY · PubMed Central
Fig. 2 Regulatory network hierarchies. a TFs and b RBPs are systematically organized into a hierarchy, forming a joint TF-RBP regulatory network. Higher layer elements tend to regulate lower layer elements. c The regulatory potentials of TFs/RBPs to drive tumor-to-normal expression changes are shown as a heatmap; red and blue indicate up- and down-regulation respectively. d Elevated MYC regulatory activity is associated with reduced disease-specific survival (DSS) in breast cancer (i); MYC knockdown in MCF-7 leads to significantly larger expression reduction in MYC target genes (ii). e MYC expression is more positively correlated with its target genes as compared to other TFs (top); MYC frequently forms FFLs with NRF1. These are mostly coherent FFLs and OR-gate logic predominates (bottom). f Elevated SUB1 regulation activity is associated with reduced overall survival (OS) in lung cancer (i); SUB1 knockdown in HepG2 leads to reduced target gene expression (ii); Targets of SUB1 show slower mRNA decay rate (iii); for cancer-associated target genes of MYC and SUB1, gene expression is decreased with both MYC and SUB1 knockdown (KD), compared with knockdown of either MYC or SUB1 individually, and compared to control (iv).
Fig. 2 Regulatory network hierarchies. a TFs and b RBPs are systematically organized into a hierarchy, forming a joint TF-RBP regulatory network. Higher layer elements tend to regulate lower layer elements. c The regulatory potentials of TFs/RBPs to drive tumor-to-normal expression changes are shown as a heatmap; red and blue indicate up- and down-regulation respectively. d Elevated MYC regulatory activity is associated with reduced disease-specific survival (DSS) in breast cancer (i); MYC knockdown in MCF-7 leads to significantly larger expression reduction in MYC target genes (ii). e MYC expression is more positively correlated with its target genes as compared to other TFs (top); MYC frequently forms FFLs with NRF1. These are mostly coherent FFLs and OR-gate logic predominates (bottom). f Elevated SUB1 regulation activity is associated with reduced overall survival (OS) in lung cancer (i); SUB1 knockdown in HepG2 leads to reduced target gene expression (ii); Targets of SUB1 show slower mRNA decay rate (iii); for cancer-associated target genes of MYC and SUB1, gene expression is decreased with both MYC and SUB1 knockdown (KD), compared with knockdown of either MYC or SUB1 individually, and compared to control (iv).CC BY · PubMed Central
Fig. 3 TF-Gene network rewiring. Green and red arrows designate edge gain and loss, respectively. a Cell-type specific network using K562 and GM12878: top layer TFs significantly drive tumor-normal differential expression; bottom layer TFs are more often associated with burdened binding sites. b JUND is a top edge-gainer in CML, and its targets demonstrate increased gene expression. However, few of its binding sites are affected by SVs or SNVs. c Rewiring index in CML by direct edge counts using both proximal and distal networks (top) and by gene community analysis (bottom). Comparisons to TF-gene rewiring networks in other cancers are also shown.
Fig. 3 TF-Gene network rewiring. Green and red arrows designate edge gain and loss, respectively. a Cell-type specific network using K562 and GM12878: top layer TFs significantly drive tumor-normal differential expression; bottom layer TFs are more often associated with burdened binding sites. b JUND is a top edge-gainer in CML, and its targets demonstrate increased gene expression. However, few of its binding sites are affected by SVs or SNVs. c Rewiring index in CML by direct edge counts using both proximal and distal networks (top) and by gene community analysis (bottom). Comparisons to TF-gene rewiring networks in other cancers are also shown.CC BY · PubMed Central
Fig. 4 Oncogenic transformation and cell state. We project the expression profiles (left, poly-A long RNA-seq), proximal network (second from right, CTCF ChIP-seq), and distal network (right, candidate cis-regulatory elements) of the ENCODE cell types to a lower dimension space. Stem-like cell types formed a cluster, suggesting stem-like cell types have a distinct profile from normal and cancerous cell types. Further, we find that cancerous cell types tend to locate closer to stem-like clusters. Oncogene knockdown in K562 led to more transcriptomic similarity to a normal cell-type, and tumor suppressor gene (TSG) knockdown led to greater similarity to a tumor cell-type (second from left, top, in comparison to GM12878). In general, we find that oncogene knockdown leads to a slight reversion towards normal state along the stem-like component (second from left, bottom).
Fig. 4 Oncogenic transformation and cell state. We project the expression profiles (left, poly-A long RNA-seq), proximal network (second from right, CTCF ChIP-seq), and distal network (right, candidate cis-regulatory elements) of the ENCODE cell types to a lower dimension space. Stem-like cell types formed a cluster, suggesting stem-like cell types have a distinct profile from normal and cancerous cell types. Further, we find that cancerous cell types tend to locate closer to stem-like clusters. Oncogene knockdown in K562 led to more transcriptomic similarity to a normal cell-type, and tumor suppressor gene (TSG) knockdown led to greater similarity to a tumor cell-type (second from left, top, in comparison to GM12878). In general, we find that oncogene knockdown leads to a slight reversion towards normal state along the stem-like component (second from left, bottom).CC BY · PubMed Central
Fig. 5 Extended genes and mutation burden analysis. a Mutation status in extended genes can explain expression differences for a larger number of genes than other annotations, such as annotations of coding sequences (CDS). b A 130-kbp deletion in the breast cancer cell line T47D potentially links a distal enhancer to the promoter of ERBB4, leading to its activation. This change does not affect coding sequences, highlighting the value of an extended gene annotation. c Cancer-associated GWAS SNVs display greater enrichment with the inclusion of proximal and distal annotations in extended gene definitions. d Somatic structural variant breakpoints in K562 tend to be associated with the activating histone mark H4K20me1, but not in GM12878.
Fig. 5 Extended genes and mutation burden analysis. a Mutation status in extended genes can explain expression differences for a larger number of genes than other annotations, such as annotations of coding sequences (CDS). b A 130-kbp deletion in the breast cancer cell line T47D potentially links a distal enhancer to the promoter of ERBB4, leading to its activation. This change does not affect coding sequences, highlighting the value of an extended gene annotation. c Cancer-associated GWAS SNVs display greater enrichment with the inclusion of proximal and distal annotations in extended gene definitions. d Somatic structural variant breakpoints in K562 tend to be associated with the activating histone mark H4K20me1, but not in GM12878.CC BY · PubMed Central
Fig. 6 Variant prioritization and validation. a A stepwise prioritization scheme for genomic regulators, elements, and variants, using the ENCODEC resources. At each step of prioritization, we indicate criteria for prioritization, as well as the applicable validation assay. b Small-scale validation of prioritized variants using a luciferase reporter assay. Candidate region 5 showed the most significant degree of differential expression and was selected for follow-up analysis. c Multiscale integrative analysis of candidate region 5 with assorted functional genomics data. The affected region is observed in the context of large-scale Hi–C linkages (top), as well as element-level signal tracks of histone modification marks and DNase hypersensitivity together with various TF binding events (middle), and nucleotide level disruption of the FOSL2 motif (bottom).
Fig. 6 Variant prioritization and validation. a A stepwise prioritization scheme for genomic regulators, elements, and variants, using the ENCODEC resources. At each step of prioritization, we indicate criteria for prioritization, as well as the applicable validation assay. b Small-scale validation of prioritized variants using a luciferase reporter assay. Candidate region 5 showed the most significant degree of differential expression and was selected for follow-up analysis. c Multiscale integrative analysis of candidate region 5 with assorted functional genomics data. The affected region is observed in the context of large-scale Hi–C linkages (top), as well as element-level signal tracks of histone modification marks and DNase hypersensitivity together with various TF binding events (middle), and nucleotide level disruption of the FOSL2 motif (bottom).CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 80 references

Cited by

Sorted by citations

Items 101–181 of 181

Related

See all

Metadata

Authors
Zhang, J., Lee, D., Dhiman, V., Jiang, P., Xu, J., McGillivray, P., Yang, H., Liu, J., Meyerson, W., Clarke, D., Gu, M., Li, S., Lou, S., Xu, J., Lochovsky, L., Ung, M., Ma, L., Yu, S., Cao, Q., Harmanci, A., Yan, K.-K., Sethi, A., Gürsoy, G., Schoenberg, M. R., Rozowsky, J., Warrell, J., Emani, P., Yang, Y. T., Galeev, T., Kong, X., Liu, S., Li, X., Krishnan, J., Feng, Y., Rivera-Mulia, J. C., Adrian, J., Broach, J. R., Bolt, M., Moran, J., Fitzgerald, D., Dileep, V., Liu, T., Mei, S., Sasaki, T., Trevilla-Garcia, C., Wang, S., Wang, Y., Zang, C., Wang, D., Klein, R. J., Snyder, M., Gilbert, D. M., Yip, K., Cheng, C., Yue, F., Liu, X. S., White, K. P., & Gerstein, M.
Affiliations
  • Yale University (US)
  • University of Chicago (US)
  • National Institutes of Health (US)
  • Northwestern University (US)
  • Dartmouth College (US)
  • Westlake University (CN)
  • Chinese University of Hong Kong (HK)
  • The University of Texas Health Science Center at Houston (US)
  • Pennsylvania State University (US)
  • Florida State University (US)
  • Harvard University (US)
  • University of Virginia (US)
  • University of Wisconsin–Madison (US)
  • Icahn School of Medicine at Mount Sinai (US)
  • Stanford University (US)
Corresponding author
  • Yue, Feng
  • Liu, X. Shirley
  • White, Kevin P.
  • Gerstein, Mark
ORCID
Venue
Nature Communications
Publisher
Springer Science and Business Media LLC
Published
2020-07-29
Volume
11
Issue
1
Pages
3696-3696
ISSN
2041-1723 (ISSN-L)
DOI
10.1038/s41467-020-14743-w
PMC
PMC7391744
OpenAlex
W3045759143
Cited by
172
References
80
Citations by year
2026: 23 · 2025: 37 · 2024: 26 · 2023: 24 · 2022: 28 · 2021: 27 · 2020: 5
FWCI
7.46
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • eScholarship (California Digital Library) (submittedVersion, cc-by)
Funders
U.S. Department of Health & Human Services | NIH | National Human Genome Research Institute (U24HG009446)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Chromatin Dynamics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Chromatin Remodeling and Cancer
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cancer Research › Cancer-related molecular mechanisms research
Keywords
ENCODE, cancer genomics, transcriptional regulatory networks, RNA-binding proteins, network rewiring, carcinogenesis, c-Myc, SUB1, variant prioritization, genome-wide association study, eCLIP, Hi-C, STARR-seq
MeSH terms
Cell Transformation, Neoplastic, Humans, Mutation, Neoplasms, Transcription Factors, Reproducibility of Results, Genomics, Databases, Genetic, Cell Line, Tumor, Gene Regulatory Networks

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review