Chromatin interaction–aware gene regulatory modeling with graph attention networks
Open accessThe DOI matches its Crossref record; 16 of the 16 references it cites were found in a registry or an index.
Abstract
Linking distal enhancers to genes and modeling their impact on target gene expression are longstanding unresolved problems in regulatory genomics and critical for interpreting noncoding genetic variation. Here, we present a new deep learning approach called GraphReg that exploits 3D interactions from chromosome conformation capture assays to predict gene expression from 1D epigenomic data or genomic DNA sequence. By using graph attention networks to exploit the connectivity of distal elements up to 2 Mb away in the genome, GraphReg more faithfully models gene regulation and more accurately predicts gene expression levels than the state-of-the-art deep learning methods for this task. Feature attribution used with GraphReg accurately identifies functional enhancers of genes, as validated by CRISPRi-FlowFISH and TAP-seq assays, outperforming both convolutional neural networks (CNNs) and the recently proposed activity-by-contact model. Sequence-based GraphReg also accurately predicts direct transcription factor (TF) targets as validated by CRISPRi TF knockout experiments via in silico ablation of TF binding motifs. GraphReg therefore represents an important advance in modeling the regulatory impact of epigenomic and sequence elements.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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
deepTools2: a next generation web server for deep-sequencing data analysis
Ramírez, F., Ryan, D. P., Grüning, B., Bhardwaj, V., Kilpert, F., Richter, A. S., Heyne, S., Dündar, F., & Manke, T. (2016) · Nucleic Acids Research · doi:10.1093/nar/gkw257 · Cited by 8,096 · cited 2×
“We also tried another normalization called reads per genomic coverage (RPGC) or 1× normalization from deepTools (Ramírez et al. 2016), and we evaluated the generalization performance across three cell lines: K562, GM12878, and hESC.”
MethodHow Powerful are Graph Neural Networks?
Xu, K., Hu, W., Leskovec, J., & Jegelka, S. (2018) · arXiv · doi:10.48550/arxiv.1810.00826 · Cited by 6,078
FIMO: scanning for occurrences of a given motif
Grant, C. E., Bailey, T. L., & Noble, W. S. (2011) · Bioinformatics · doi:10.1093/bioinformatics/btr064 · Cited by 4,247 · cited 1×
“We used FIMO software (Grant et al. 2011) to scan these 20bp sequences and findmotifmatches to theCIS-BPmotif database.”
Axiomatic attribution for deep networks
Sundararajan, M., Taly, A., & Yan, Q. (2017) · arXiv · doi:10.48550/arxiv.1703.01365 · Cited by 4,087
ATAC-seq: a method for assaying chromatin accessibility genome-wide
Buenrostro, J. D., Wu, B., Chang, H. Y., & Greenleaf, W. J. (2015) · Current Protocols in Molecular Biology · doi:10.1002/0471142727.mb2129s109 · Cited by 3,289
Learning Important Features Through Propagating Activation Differences
Shrikumar, A., Greenside, P., & Kundaje, A. (2017) · arXiv · doi:10.48550/arxiv.1704.02685 · Cited by 2,761
Effective gene expression prediction from sequence by integrating long-range interactions
Avsec, Ž., Agarwal, V., Visentin, D., Ledsam, J. R., Grabska-Barwinska, A., Taylor, K. R., Assael, Y., Jumper, J., Kohli, P., & Kelley, D. R. (2021) · Nature Methods · doi:10.1038/s41592-021-01252-x · Cited by 1,326 · cited 2×
“A recent work, Enformer (Avsec et al., 2021a), introduced a transformer model for the prediction of CAGE as well as epigenomic tracks from genomic sequence.”
BackgroundActivity-by-contact model of enhancer–promoter regulation from thousands of CRISPR perturbations
Fulco, C. P., Nasser, J., Jones, T. R., Munson, G., Bergman, D. T., Subramanian, V., Grossman, S. R., Anyoha, R., Doughty, B. R., Patwardhan, T. A., et al. (2019) · Nature Genetics · doi:10.1038/s41588-019-0538-0 · Cited by 1,111
Single-cell RNA-seq denoising using a deep count autoencoder
Eraslan, G., Simon, L. M., Mircea, M., Mueller, N. S., & Theis, F. J. (2019) · Nature Communications · doi:10.1038/s41467-018-07931-2 · Cited by 1,084
Ultrastructural Details of Mammalian Chromosome Architecture
Krietenstein, N., Abraham, S., Venev, S. V., Abdennur, N., Gibcus, J., Hsieh, T.-H. S., Parsi, K. M., Yang, L., Maehr, R., Mirny, L. A., et al. (2020) · Molecular Cell · doi:10.1016/j.molcel.2020.03.003 · Cited by 622
Base-resolution models of transcription-factor binding reveal soft motif syntax
Avsec, Ž., Weilert, M., Shrikumar, A., Krueger, S., Alexandari, A., Dalal, K., Fropf, R., McAnany, C., Gagneur, J., Kundaje, A., et al. (2021) · Nature Genetics · doi:10.1038/s41588-021-00782-6 · Cited by 610 · cited 1×
“Dilated CNNs have been used previously to increase the receptive field of CNN layers in deep learning models that predict epigenomic data from DNA sequence, such as Basenji [7], BPNet [14], and Xpresso [15].”
MethodDeep learning sequence-based ab initio prediction of variant effects on expression and disease risk
Zhou, J., Theesfeld, C. L., Yao, K., Chen, K. M., Wong, A. K., & Troyanskaya, O. G. (2018) · Nature Genetics · doi:10.1038/s41588-018-0160-6 · Cited by 559 · cited 3×
“, 2021b), ExPecto (Zhou et al., 2018), and Xpresso (Agarwal and Shendure, 2020).”
BackgroundSequential regulatory activity prediction across chromosomes with convolutional neural networks
Kelley, D. R., Reshef, Y. A., Bileschi, M., Belanger, D., McLean, C. Y., & Snoek, J. (2018) · Genome Research · doi:10.1101/gr.227819.117 · Cited by 537 · cited 6×
“However, for enhancer scoring and validation with FlowFISH analysis, as suggested in Basenji [7], we derived the saliency maps as the dot product of the 100bp bin representations (after the CNN layers and before the dilated CNN layers) and the gradient of the model prediction for the gene with respect to those bin rep…”
BackgroundMethodInfluentialTargeted Perturb-seq enables genome-scale genetic screens in single cells
Schraivogel, D., Gschwind, A. R., Milbank, J. H., Leonce, D. R., Jakob, P., Mathur, L., Korbel, J. O., Merten, C. A., Velten, L., & Steinmetz, L. M. (2020) · Nature Methods · doi:10.1038/s41592-020-0837-5 · Cited by 245
Predicting mRNA abundance directly from genomic sequence using deep convolutional neural networks
Agarwal, V., & Shendure, J. (2018) · doi:10.1101/416685 · Cited by 243
Predicting mRNA Abundance Directly from Genomic Sequence Using Deep Convolutional Neural Networks
Agarwal, V., & Shendure, J. (2020) · Cell Reports · doi:10.1016/j.celrep.2020.107663 · Cited by 243
Cited by
DeepSTARR predicts enhancer activity from DNA sequence and enables the de novo design of enhancers
de Almeida, B. P., Reiter, F., Pagani, M., & Stark, A. (2021) · doi:10.1101/2021.10.05.463203 · Cited by 284
de Almeida, B. P., Reiter, F., Pagani, M., & Stark, A. (2022) · Nature Genetics · doi:10.1038/s41588-022-01048-5 · Cited by 283
Graph Attention Networks: A Comprehensive Review of Methods and Applications
Vrahatis, A. G., Lazaros, K., & Kotsiantis, S. (2024) · Future Internet · doi:10.3390/fi16090318 · Cited by 274
Artificial intelligence in cancer research, diagnosis and therapy
Elemento, O., Leslie, C., Lundin, J., & Tourassi, G. (2021) · Nature Reviews Cancer · doi:10.1038/s41568-021-00399-1 · Cited by 264
Self-citationAn encyclopedia of enhancer-gene regulatory interactions in the human genome
Gschwind, A. R., Mualim, K. S., Karbalayghareh, A., Sheth, M. U., Dey, K. K., Jagoda, E., Nurtdinov, R. N., Xi, W., Tan, A. S., Jones, H., et al. (2023) · doi:10.1101/2023.11.09.563812 · Cited by 119
Self-citation
Metadata
- Authors
- Karbalayghareh, A., Sahin, M., & Leslie, C. S.
- Venue
- Genome Research
- Publisher
- Cold Spring Harbor Laboratory
- Published
- 2022-04-08
- Volume
- 32
- Issue
- 5
- Pages
- 930-944
- ISSN
- 1088-9051 (ISSN-L)
- 1549-5469
- OpenAlex
- W4225256479
- Cited by
- 71
- References
- 16
- Type
- posted content
- Language
- en
- Open access
- bronze
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.