The DOI matches its Crossref record; 161 of the 163 references it cites were found in a registry or an index, and 1 are not checked yet.
Base-resolution models of transcription-factor binding reveal soft motif syntax
Open accessPublished version of a preprint
Abstract
The arrangement (syntax) of transcription factor (TF) binding motifs is an important part of the cis-regulatory code, yet remains elusive. We introduce a deep learning model, BPNet, that uses DNA sequence to predict base-resolution chromatin immunoprecipitation (ChIP)-nexus binding profiles of pluripotency TFs. We develop interpretation tools to learn predictive motif representations and identify soft syntax rules for cooperative TF binding interactions. Strikingly, Nanog preferentially binds with helical periodicity, and TFs often cooperate in a directional manner, which we validate using clustered regularly interspaced short palindromic repeat (CRISPR)-induced point mutations. Our model represents a powerful general approach to uncover the motifs and syntax of cis-regulatory sequences in genomics data.
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.
Integrity signals
Registry
Cites
Statistics
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.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The BPNet architecture has been used widely in genomics and represents a very strong default architecture for modeling small-sized datasets from scratch through supervised learning.”
From Nucleotide Transformer: building and evaluating robust foundation models for human genomics (2025) · cites this paper as [9]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Machine-learning models trained to predict function from DNA sequences have been successful at characterizing regulatory syntax and interpreting genetic variant effects.”
From Predicting RNA-seq coverage from DNA sequence as a unifying model of gene regulation (2025) · cites this paper as [7]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Thus far, such models have focused on assays in which measured activity is proportional to local sequencing read counts.”
From Predicting RNA-seq coverage from DNA sequence as a unifying model of gene regulation (2025) · cites this paper as [7]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Finer resolution has potential benefits when the data support it15, but would extend the sequence length entering the quadratic complexity MHA and make the model engineering intractable on currently available hardware.”
From Effective gene expression prediction from sequence by integrating long-range interactions (2021) · cites this paper as [15]
The cited source says what the claim attributes to it.
In the citing paper
“Additionally, a deep learning algorithm applied to binding of pluripotency TFs at base resolution in mouse embryonic stem cells (ESCs) suggested a “soft motif syntax” whereby TFs help each other to bind directly to the enhancer with a lenient distance-dependency between binding sites [98].”
From Transcriptional enhancers and their communication with gene promoters (2021) · cites this paper as [98]
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
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
VMD: visual molecular dynamics
Humphrey, W., Dalke, A., & Schulten, K. (1996) · Journal of Molecular Graphics · doi:10.1016/0263-7855(96)00018-5 · Cited by 66,952
The Sequence Alignment/Map format and SAMtools
Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G., Abecasis, G., Durbin, R., & 1000 Genome Project Data Processing Subgroup (2009) · Bioinformatics · doi:10.1093/bioinformatics/btp352 · Cited by 62,041
Fast and accurate short read alignment with Burrows-Wheeler transform
Li, H., & Durbin, R. (2009) · Bioinformatics · doi:10.1093/bioinformatics/btp324 · Cited by 58,607
Cutadapt removes adapter sequences from high-throughput sequencing reads
Martin, M. (2011) · EMBnet.journal · doi:10.14806/ej.17.1.200 · Cited by 33,889
Kimura, M. (1980) · Journal of Molecular Evolution · doi:10.1007/bf01731581 · Cited by 29,778
BEDTools: a flexible suite of utilities for comparing genomic features
Quinlan, A. R., & Hall, I. M. (2010) · Bioinformatics · doi:10.1093/bioinformatics/btq033 · Cited by 27,437
Ultrafast and memory-efficient alignment of short DNA sequences to the human genome.
Langmead, B., Trapnell, C., Pop, M., & Salzberg, S. L. (2009) · Genome Biology · doi:10.1186/gb-2009-10-3-r25 · Cited by 21,849
Hierarchical Grouping to Optimize an Objective Function
Ward, J. H. (1963) · Journal of the American Statistical Association · doi:10.1080/01621459.1963.10500845 · Cited by 21,341
Research works printed but not linked: 1
- Shrikumar, A. et al. TF-MoDISco v0.4.2.2-alpha: technical note. Preprint at arXiv https://arxiv.org/abs/1811.00416 (2018). · arXiv
Cited by
Sorted by citations
Items 301–400 of 654
Anaemia, macrocytosis, vitamin B12 and folate levels in elderly Zimbabweans.
Allain, T. J., Gomo, Z. A. R., Wilson, A. O., Ndemera, B., Adamchak, D. J., & Matenga, J. A. (1997) · PubMed · Cited by 8
Uusi-Mäkelä, J., Afyounian, E., Tabaro, F., Häkkinen, T., Lussana, A., Shcherban, A., Annala, M., Nurminen, R., Kivinummi, K. K., Tammela, T. L. J., et al. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.09.08.287268 · Cited by 11
CPA-Perturb-seq: Multiplexed single-cell characterization of alternative polyadenylation regulatorsPreprint
Kowalski, M. H., Wessels, H., Linder, J., Choudhary, S., Hartman, A., Hao, Y., Mascio, I., Dalgarno, C., Kundaje, A. B., & Satija, R. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.02.09.527751 · Cited by 8
“We next used a motif discovery tool, TF-MoDISco 60,64 , to cluster the attribution scores of each perturbation into a set of salient motifs (Figure 5D, Supplementary Figure 5C-D).”
Self-citationGUANinE v1.0: Benchmark Datasets for Genomic AI Sequence-to-Function ModelsPreprint
Robson, E. S., & Ioannidis, N. M. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.10.12.562113 · Cited by 9
“…machinery to analyze genomes has grown in complexity, from sophisticated tree models to reconstruct demographic and evolutionary histories [59, 5] to neural network-based sequence-to-function maps (i.e. f : X → y ) for [epi]genomic annotation, imputation, and understanding [88, 7, 89, 39, 6].”
BackgroundCurrent genomic deep learning architectures generalize across grass species but not allelesPreprint
Wrightsman, T., Ferebee, T. H., Romay, M. C., Seetharam, A. S., AuBuchon‐Elder, T., Phillips, A. R., Syring, M., Hufford, M. B., Kellogg, E. A., & Buckler, E. S. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.04.11.589024 · Cited by 6
“…in predicting non-coding genomic features such as chromatin accessibility (Kelley, 2020; Wrightsman et al., 2022), transcription factor binding (Žiga Avsec, Weilert, et al., 2021; Mejía-Guerra & Buckler, 2019), and RNA abundance (Žiga Avsec, Agarwal, et al., 2021; Linder et al., 2023) directly…”
Massively parallel approaches for characterizing noncoding functional variation in human evolution
Rong, S., Root, E., & Reilly, S. K. (2024) · Current Opinion in Genetics & Development · doi:10.1016/j.gde.2024.102256 · Cited by 9
Interpreting regulatory mechanisms of Hippo signaling through a deep learning sequence model
Dalal, K., McAnany, C. E., Weilert, M., McKinney, M. C., Krueger, S., & Zeitlinger, J. (2025) · Cell Genomics · doi:10.1016/j.xgen.2025.100821 · Cited by 7
“…highest resolution and leveraged our previously developed deep learning model BPNet to predict the data at base resolution from genomic sequences of 1 kb. 15,20,22,38,39 This approach has high predictive accuracy and optimally re-solves sequence rules between closely spaced motifs within enhancers.”
InfluentialSelf-citationDeep learning for regulatory genomics: a survey of models, challenges, and applications
Ismail, F. N., Sengupta, A., & Amarasoma, S. (2025) · Bioinformatics Advances · doi:10.1093/bioadv/vbaf271 · Cited by 8
“For instance, BPNet (Avsec et al. 2021b) demonstrated cooperative binding syntax amongst factors in addition to predicting base-resolution transcription factor binding.”
Enhancer cooperativity in the folded genome
Friman, E. T., & Bickmore, W. A. (2025) · Current Opinion in Genetics & Development · doi:10.1016/j.gde.2025.102416 · Cited by 7
Modernización y cambio cultural: la persistencia de los valores tradicionales
Inglehart, R. (2005) · Quaderns de la Mediterrània = Cuadernos del Mediterráneo · Cited by 7
Wagner, A. (2024) · Bioinformatics · doi:10.1093/bioinformatics/btae317 · Cited by 7
“…(Jumper et al. 2021, Cheng et al. 2023), protein stability (Pancotti et al. 2021, Blaabjerg et al. 2023), protein-nucleic acid binding (Avsec et al. 2021, Alipanahi et al. 2015), DNA methylation (Angermueller et al. 2017), mutational effects on proteins and RNA (Riesselman et al.…”
BackgroundTissue-specific and tissue-agnostic effects of genome sequence variation modulating blood pressure
Lee, D., Han, S. K., Yaacov, O., Berk-Rauch, H., Mathiyalagan, P., Ganesh, S. K., & Chakravarti, A. (2023) · Cell Reports · doi:10.1016/j.celrep.2023.113351 · Cited by 7
“11 12–15 Over the past decade, several groups, includingus, have pioneered these approaches and demonstrated that functional CRE variants can be predicted from DNA sequences, 16–18 using a two-step approach.”
Sequence-based modelling of bacterial genomes enables accurate antibiotic resistance predictionPreprint
Wiatrak, M., Weimann, A., Dinan, A. M., Brbić, M., & Floto, R. A. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.01.03.574022 · Cited by 8
“CNN models have been shown to be able to extract patterns from the DNA sequence such as transcription factor binding motifs [21, 22].”
BackgroundMethodMulti-locus CRISPRi targeting with a single truncated guide RNA
Moore, M. M., Wekhande, S., Issner, R., Collins, A., Cruz, A. J., Liu, Y. V., Javed, N. M., Casaní-Galdón, S., Buenrostro, J. D., Epstein, C. B., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-56144-x · Cited by 6
Kayani, S., Hu, X., Shen, Q., Zou, B., Zhu, F., Yu, Z., Shah, M. A. R., Rehman, O. U., & Huo, S. (2025) · Critical Reviews in Biotechnology · doi:10.1080/07388551.2025.2503788 · Cited by 6
Zeng, W., Guo, H., Liu, Q., & Wong, W. H. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2419202122 · Cited by 8
“By accurately predicting the impact of genetic variants on transcription factor binding at a single base - pair resolution, BPNet enhances the precision of transcription factor binding prediction ( 41 ).”
The regulatory code of injury-responsive enhancers enables precision cell-state targeting in the CNS
Zamboni, M., Martínez-Martín, A., Rydholm, G., Häneke, T., Almeida, L. P., Seçilmiş, D., Ziegenhain, C., & Llorens-Bobadilla, E. (2025) · Nature Neuroscience · doi:10.1038/s41593-025-02131-w · Cited by 10
“The architecture of the models used in this paper is derived from ChromBPNet, which is, in turn, based on BPNet 18 .”
InfluentialGaur, P., Satheesh, T., Singh, R., Beig, H., Shahu, S., Ramakrishnan, S., Srivastava, M., Neogi, S., Dash, A., Singh, A., et al. (2025) · Nucleic Acids Research · doi:10.1093/nar/gkaf1428 · Cited by 6
“Our findings are particularly relevant in the context of recent experimental findings suggesting that weak-affinity flanking sequences around the core DNA-binding motifs of transcription factors play a crucial role in achieving their sequence specificity [ 47 , 48 ].”
TargetGAN: A generative AI framework for the design of plant core promoters with targeted activity
Xianglei, X., Yao, Q., Deng, K., Ge, Y., Xiong, Q., Lu, Y., & Hu, X. (2026) · Plant Communications · doi:10.1016/j.xplc.2026.101851 · Cited by 6
“To quantify the contribution of individual nucleotides to predicted promoter activity, we used DeepLIFT (Shrikumar et al., 2017), a gradient-based interpretation method (Avsec et al., 2021; de Almeida et al., 2022).”
McAnany, C. E., Weilert, M., Mehta, G., Kamulegeya, F., Gardner, J., Schreiber, J. M., Kundaje, A. B., & Zeitlinger, J. (2026) · Nature Communications · doi:10.1038/s41467-026-74807-1 · Cited by 6
“thebindingofasingleTFwillnotonlylearnthatTF ’ sbindingmotif,but also other sequence patterns, such as motifs for cooperative binding partners 3,13 .”
InfluentialSelf-citationA suboptimal OCT4-SOX2 binding site facilitates the naïve-state specific function of a Klf4 enhancer
Waite, J. B., Boytz, R., Traeger, A. R., Lind, T. M., Lumbao-Conradson, K., & Torigoe, S. E. (2024) · PLoS ONE · doi:10.1371/journal.pone.0311120 · Cited by 5
“While there are several previously published studies on enhancer grammar in pluripotent stem cells [21–24], it has remained unknown whether there are grammatical constraints to separate enhancer function across the different states of pluripotency.”
BackgroundSnyder, L. F., O'Brien, E. M., Zhao, J., Liang, J., Bruce, B. J., Zhang, Y., Zhu, W., Cassier, T., Schnicker, N., Zhou, X., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-59244-w · Cited by 10
“Another important function of combinatorial control and TF interdependence is to allow cells to integrate multiple upstream signals.”
Defining transcription factor nucleosome binding with Pioneer-seq
Tsompana, M., Wilson, P. D., Murugaiyan, V., Handelmann, C. R., & Buck, M. J. (2025) · PLoS Genetics · doi:10.1371/journal.pgen.1011813 · Cited by 6
“Klf4-2 is a longer motif (GCCCCGCCCCGCCCC) derived from [29], and Klf4-2RC is its reverse complement.”
Deciphering cis-regulatory grammar with deep learning
Miraldi, E. R., Chen, X., & Weirauch, M. T. (2021) · Nature Genetics · doi:10.1038/s41588-021-00814-1 · Cited by 6
Trerotola, M., Antolini, L., Beni, L., Guerra, E., Spadaccini, M., Verzulli, D., Moschella, A., & Alberti, S. (2022) · NAR Genomics and Bioinformatics · doi:10.1093/nargab/lqac008 · Cited by 6
“Consistent, genome-wide analyses have rarely identified statistically over-represented motif syntax rules, questioning whether they indeed exist and whether they impose evolutionary constraints on regulatory elements function (13,26).”
BackgroundTranslation Rate Prediction and Regulatory Motif Discovery with Multi-task Learning
Zheng, W., Fong, J. H. C., Wan, Y. K., Chu, A. H. Y., Huang, Y., Wong, A. S. L., & Ho, J. W. K. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-29119-7_9 · Cited by 6
Toward Identification of Functional Sequences and Variants in Noncoding DNA
Monti, R., & Ohler, U. (2023) · Annual Review of Biomedical Data Science · doi:10.1146/annurev-biodatasci-122120-110102 · Cited by 6
“…sequence can learn a variety of tasks at different levels of gene regulation, including binding of individual TFs (85, 86) or RNA-binding proteins (85, 91), accessibility of DNA (87), histone modifications (86, 94), transcription (95, 96), splicing (97), and 3D genome folding (98, 99) ( Figure 4 ).”
BackgroundInfluentialZhang, Y., Wang, Z., Liu, Y., Lu, L., Tan, X., & Zou, Q. (2021) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm52615.2021.9669387 · Cited by 6
“Recently, deep learning-based methods have been successfully applied in bioinformatics [6]–[11], such as TFs binding sites imputation.”
MethodBuilding integrative functional maps of gene regulation
Xu, J., Pratt, H., Moore, J. E., Gerstein, M., & Weng, Z. (2022) · Human Molecular Genetics · doi:10.1093/hmg/ddac195 · Cited by 6
Quantitative analysis of ZFY and CTCF reveals dependent recognition of tandem zinc finger proteinsPreprint
Zuo, Z., Billings, T., Walker, M., Petkov, P. M., Stormo, G. D., & Fordyce, P. M. (2019) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/637298 · Cited by 6
Zhou, H., Shrikumar, A., & Kundaje, A. B. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.11.04.368803 · Cited by 6
“For the base-pair-level signal profile prediction, we used genome-wide ChIP-nexus profiles of four TFs - Oct4, Sox2, Nanog and Klf4 in mouse embryonic stem cells [9].”
BackgroundMethodInfluentialSelf-citationDynamic regulatory module networks for inference of cell type–specific transcriptional networksPreprint
Siahpirani, A. F., Knaack, S., Chasman, D. A., Seirup, M., Sridharan, R., Stewart, R. M., Thomson, J. L., & Roy, S. (2022) · Genome Research · doi:10.1101/gr.276542.121 · Cited by 6
“Another direction would be to use more generic sequence features, such as k -mers (Setty and Leslie 2015) or deep learning models (Avsec et al. 2021), to enable the discoveryof novel regulatory elements and offer great flexibility in capturing sequence specificity and its role in predictive models…”
Altered AP-1, RUNX and EGR chromatin dynamics drive fibrotic lung diseasePreprint
Valenzi, E., Jia, M., Gerges, P., Fan, J., Tabib, T., Behara, R., Zhou, Y., Sembrat, J., Das, J., Benos, P. V., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.10.23.619858 · Cited by 6
“After removing bias, corrected signal profiles for each cell state were used to generate single-base pair resolution contribution scores, with de novo motif discovery by TF-MoDISco 23,24 and annotation by JASPAR TF motifs 25 .”
Deaminase-mediated chromatin accessibility profiling with single-allele resolutionPreprint
Yu, T., Li, Z., Gibbs, E., Iwase, R., Francoeur, M., Phan, Q. V., Zhao, J., Rosin, J., Cole, P. A., Pinello, L., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.12.17.628768 · Cited by 6
A generative framework for enhanced cell-type specificity in rationally designed mRNAsPreprint
Khoroshkin, M., Zinkevich, A. O., Aristova, E., Yousefi, H., Lee, S. B., Mittmann, T., Manegold, K., Penzar, D., Raleigh, D. R., Kulakovskiy, I. V., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.12.31.630783 · Cited by 8
Marderstein, A. R., Kundu, S., Padhi, E. M., Deshpande, S., Wang, A., Robb, E., Sun, Y., Yun, C. M., Pomales-Matos, D. A., Xie, Y., et al. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.02.18.638922 · Cited by 6
Self-citationValencia, J. D., & Hendrix, D. A. (2023) · PLOS Computational Biology · doi:10.1371/journal.pcbi.1011526 · Cited by 6
Ding, P., Wang, J., He, S., Gao, X., Yu, X., & Yu, B. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.111279 · Cited by 5
Altered AP-1, RUNX, and EGR chromatin dynamics drive human fibrotic lung disease
Valenzi, E., Jia, M., Gerges, P., Fan, J., Tabib, T., Behera, R., Mao, H., Xue, D., Ju, E.-M., Zhou, Y., et al. (2025) · Annals of the Rheumatic Diseases · doi:10.1016/j.ard.2025.12.003 · Cited by 11
Baderna, V., Barzaghi, G., Kleinendorst, R., Chatsirisupachai, K., Moniot-Perron, L., Doyle, C., Schopp, M., Hochepied, T., Libert, C., Odom, D. T., et al. (2026) · Nature Genetics · doi:10.1038/s41588-026-02703-x · Cited by 12
“63.”
Deconstructing the GWAS library: next-generation GWAS
Zhang, W., Koslowski, S., Benzaki, M., Lee, C. J. M., Zhu, Y., Mak, M. C. E., Zhu, Y., Loong, S. S. E., Lettre, G., Anene-Nzelu, C. G., et al. (2025) · Physiological Reviews · doi:10.1152/physrev.00025.2024 · Cited by 5
Computational Approaches for Understanding Sequence Variation Effects on the 3D Genome Architecture
Avdeyev, P., & Zhou, J. (2022) · Annual Review of Biomedical Data Science · doi:10.1146/annurev-biodatasci-102521-012018 · Cited by 5
Maytum, A., Edginton‐White, B., & Bonifer, C. (2023) · Stem Cell Investigation · doi:10.21037/sci-2023-011 · Cited by 3
“Our data is a valuable source for the study of how TF motifs and their arrangements regulate enhancer function and coupled with recently published methods such as (126) will provide insight into these combinations.”
Cho, H. J., Wang, Z., Cong, Y., Bekiranov, S., Zhang, A., & Zang, C. (2024) · Genes · doi:10.3390/genes15020144 · Cited by 5
A variegated model of transcription factor function in the immune system
Chowdhary, K., & Benoist, C. O. (2023) · Trends in Immunology · doi:10.1016/j.it.2023.05.001 · Cited by 5
The dynseq browser track shows context-specific features at nucleotide resolution
Nair, S., Barrett, A., Li, D., Raney, B. J., Lee, B. T., Kerpedjiev, P., Ramalingam, V., Pampari, A., Lekschas, F., Wang, T., et al. (2022) · Nature Genetics · doi:10.1038/s41588-022-01194-w · Cited by 6
Self-citationKoo, P. K., Ploenzke, M., Anand, P., Paul, S. B., & Majdandzic, A. (2020) · bioRxiv · doi:10.1101/2020.09.08.288068 · Cited by 5
Ranking Reprogramming Factors for Directed DifferentiationPreprint
Hammelman, J., Patel, T., Closser, M., Wichterle, H., & Gifford, D. K. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.05.14.444080 · Cited by 5
Concert: Genome-wide prediction of sequence elements that modulate DNA replication timingPreprint
Yang, Y., Wang, Y., Zhang, Y., & Ma, J. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.04.21.488684 · Cited by 5
“More recently, a few models have been proposed to use dilated convo-lutions to learn long-range spatial dependencies in DNA sequences, which are critical for the prediction of long-range gene regulation and chromatin interaction [18–21].”
BackgroundMultimodal foundation model predicts zero-shot functional perturbations and cell fate dynamicsPreprint
Yang, Z., Fan, X., Lan, M., Tang, X., Zheng, Z., Liu, B., You, Y., Tian, L., Church, G. M., Liu, X., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.12.19.629561 · Cited by 5
Genome-wide mapping of mesoscale neuronal RNA organization and condensationPreprint
Becker, L. A., Quinodoz, S. A., Comi, T. J., Kimchi, O., Knowles, D. A., & Brangwynne, C. P. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.04.19.649570 · Cited by 5
“DL models including convolutional neural networks 143–145 (CNN), recurrent neural networks 146,147 (RNN), transformers, and hybrids of these models 148 , have had substantial success as “sequence-to-function” models predicting diverse molecular phenotypes including transcription factor binding,…”
Tan, W. Z., Wang, Z., & Liu, L. (2024) · Agriculture · doi:10.3390/agriculture15010029 · Cited by 5
Liu, X., Rastegari, S., Huang, Y., Cheong, S. C., Liu, W., Zhao, W., Tian, Q., Wang, H., Guo, Y., Zhou, S., et al. (2025) · Methods · doi:10.1016/j.ymeth.2025.07.004 · Cited by 5
Implementing XAI in life sciences: Key challenges and pathways to solutions
KREJCAR, O., Abdullah, J., & Namazi, H. (2026) · Artificial Intelligence in the Life Sciences · doi:10.1016/j.ailsci.2026.100153 · Cited by 4
Wang, S. K., Li, J., Nair, S., Kosaraju, R., Chen, Y., Zhang, Y., Kundaje, A. B., Liu, Y., Wang, N., & Chang, H. Y. (2026) · Cell Reports · doi:10.1016/j.celrep.2025.116814 · Cited by 7
Self-citationLin, J., Li, Z., Zhao, Y., Luo, R., & Pinello, L. (2026) · Nature Communications · doi:10.1038/s41467-026-70535-8 · Cited by 24
“The rich and large dataset generated by these consortia has been crucial for training powerful deep-learning methods 11 , furthering our ability to dissect and understand gene regulatory mechanisms 12 – 15 .”
Genome evolution through polyploidy: Enhancing plant stress resilience in agriculture
Edger, P. P., Body, M. J. A., Donno, S. D., Platts, A. E., Wang, J., & Jiang, J. (2026) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2522064123 · Cited by 5
Recio, P. S., Mitra, N. J., Shively, C. A., Song, D., Jaramillo, G., Shady, K., Chen, X., & Mitra, R. D. (2023) · Nucleic Acids Research · doi:10.1093/nar/gkad320 · Cited by 4
Extensive enhancer crosstalk controls PPARG2 activation during adipogenesis
Cetnarowska, A., Hyldahl, M., Nygård, M., Dashti, H., Hansen, B. M., Holm, L. K., Madsen, K., Madsen, M. S., Shukla, V., Mitchell, E. D., et al. (2026) · Nature Communications · doi:10.1038/s41467-026-70401-7 · Cited by 4
Wang, Z., Xiong, S., Yu, Y., Zhou, J., & Zhang, Y. (2023) · Bioinformatics · doi:10.1093/bioinformatics/btad299 · Cited by 4
“BPNet (Avsec et al. 2021) and U-TransNet (Zhang et al.”
MethodCharacterization of an eye field-like state during optic vesicle organoid development
Owen, L. J., Rainger, J. K., Bengani, H., Kilanowski, F. M., FitzPatrick, D. R., & Papanastasiou, A. S. (2023) · Development · doi:10.1242/dev.201432 · Cited by 5
Learning chemical sensitivity reveals mechanisms of cellular response
Connell, W., Garcia, K., Goodarzi, H., & Keiser, M. J. (2024) · Communications Biology · doi:10.1038/s42003-024-06865-4 · Cited by 4
“Deeplearningmodelattribution methodsareprimarilyempirical,sothepotentialcompoundmechanismsof actiontheyrevealultimatelynecessitateprospectivebiologicaltesting 52,53 .”
Menichelli, C., Guitard, V., Martins, R. M., Lèbre, S., Lopez-Rubio, J.-J., Lecellier, C., & Bréhélin, L. (2021) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1008909 · Cited by 5
Resolving diverse protein–DNA footprints from exonuclease-based ChIP experiments
Biswas, A., & Narlikar, L. (2021) · Bioinformatics · doi:10.1093/bioinformatics/btab274 · Cited by 4
Deciphering the regulatory syntax of genomic DNA with deep learning
Lal, A. (2022) · Journal of Biosciences · doi:10.1007/s12038-022-00291-6 · Cited by 6
“For example, Avsec et al. (2021b) trained BPnet to predict TF–DNA binding, and then applied the trained model to synthetic sequences in which two TF-binding motifs were placed at different distances.”
MethodFinding motifs using DNA images derived from sparse representations
Chu, S., & Stormo, G. D. (2023) · Bioinformatics · doi:10.1093/bioinformatics/btad378 · Cited by 4
Zhuang, J., Feng, K., Teng, X., & Jia, C. (2023) · Mathematical Biosciences & Engineering · doi:10.3934/mbe.2023704 · Cited by 4
Romero, R., Menichelli, C., Vroland, C., Marin, J., Lèbre, S., Lecellier, C., & Bréhélin, L. (2024) · Genome biology · doi:10.1186/s13059-024-03321-8 · Cited by 6
“Hence, in contrast to CNN based methods [5, 62], our approach completely controls the predictive features input into the model.”
MethodLearning immune cell differentiationPreprint
Maslova, A., Ramírez, R. N., Ma, K., Schmutz, H., Wang, C., Fox, C., Ng, B., Benoist, C. O., Mostafavi, S., & the Immunological Genome Project (2019) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2019.12.21.885814 · Cited by 4
AMBIENT: Accelerated Convolutional Neural Network Architecture Search for Regulatory GenomicsPreprint
Zhang, Z., Cofer, E. M., & Troyanskaya, O. G. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.02.25.432960 · Cited by 5
Janizek, J. D., Dincer, A. B., Çelik, S., Chen, H., Chen, W. C., Naxerova, K., & Lee, S. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.10.06.463409 · Cited by 4
Chaudhri, V. K., & Singh, H. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.11.26.470154 · Cited by 4
“Another study has investigated the possibility of soft-syntax for the regulatory code using helical distance to expand the spectrum of binding preferences(36).”
BackgroundMethodAlam, A., Duncan, A. G., Mitchell, J. A., & Moses, A. M. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.04.09.536185 · Cited by 5
“Compared to deep learning approaches based on deep convolutional neural networks [27,94,95] or language models [28], the average motif score representation is trivially interpretable, and needs no downstream steps to identify important motif combinations (e.g., Abf1 and Reb1 working with Rpn4,…”
BackgroundContext transcription factors establish cooperative environments and mediate enhancer communicationPreprint
Kribelbauer-Swietek, J. F., Pushkarev, O., Gardeux, V., Russeil, J., van Mierlo, G., & Deplancke, B. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.05.05.539543 · Cited by 4
Transcription factors operate on a limited vocabulary of binding motifs in Arabidopsis thalianaPreprint
Zenker, S., Wulf, D., Meierhenrich, A., Becker, S., Eisenhut, M., Stracke, R., Weißhaar, B., & Bräutigam, A. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.08.28.555073 · Cited by 4
“Deep learning has been used to predict TFBMs (Li et al., 2018) and to predict gene expression from DNA sequence features (Avsec et al., 2021b).”
Overlapping binding sites underlie TF genomic occupancyPreprint
Khetan, S., & Bulyk, M. L. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.03.05.583629 · Cited by 4
“As machine learning methods are increasingly being developed to decipher cis -regulatory codes ( 77 – 79 ) and design synthetic cis -regulatory elements ( 80 ), consideration of consecutive overlapping binding sites may improve model performance.”
MethodTraining deep learning models on personalized genomic sequences improves variant effect predictionPreprint
He, A. Y., Palamuttam, N. P., & Danko, C. G. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.10.15.618510 · Cited by 6
“Individual CNNs use the CLIPNET architecture [17], which was inspired by the BPNet [8] architecture.”
Predictable Engineering of Signal-Dependent Cis-Regulatory ElementsPreprint
Cornwall-Scoones, J., Benzinger, D., Yu, T., Pezzotta, A., Sagner, A., Gerontogianni, L., Bernadet, S., Finnie, E., Boezio, G. L. M., Stuart, H. T., et al. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.03.07.642002 · Cited by 5
Shift augmentation improves DNA convolutional neural network indel effect predictionsPreprint
Korsakova, A., Srivastava, D., & Kelley, D. R. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.04.07.647656 · Cited by 7
“These include the sum of the difference or log fold change at each position in the predicted vectors (SUM), L2 norm, or Kullback-Leibler divergence (after normalization) [11].”
Predicting gene expression using millions of yeast promoters reveals cis -regulatory logic
Dash, T., & Bornelöv, S. (2024) · Bioinformatics Advances · doi:10.1093/bioadv/vbaf130 · Cited by 4
Masri, C. A., & Yu, J. H. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c01143 · Cited by 3
“Subsequent approaches 41 – 44 have further improved the performance of DNA sequence-based models.”
How negative sampling shapes the performance of transcription factor binding site prediction models
Tourne, N., Waele, G. D., Vermeirssen, V., & Waegeman, W. (2026) · Bioinformatics · doi:10.1093/bioinformatics/btag048 · Cited by 5
“Another approach uses base-resolution predictions, where both negative and positive positions are present in the input window (Avsec et al. 2021b, Zhang et al. 2021).”
Cross-species prediction reveals chromatin regions with increased accessibility in humans
Wang, L., Li, Y., Han, D., & Wang, Z. (2026) · Science Advances · doi:10.1126/sciadv.ady9169 · Cited by 4
Decoding common and rare noncoding variant effects across cellular and developmental contexts
Marderstein, A. R., Kundu, S., Padhi, E. M., Deshpande, S., Wang, A., Robb, E., Sun, Y., Yun, C. M., Pomales-Matos, D. A., Xie, Y., et al. (2026) · Nature Genetics · doi:10.1038/s41588-026-02619-6 · Cited by 18
“12 and 13, Supplementary Table 11 and Supplementary Note 20).”
Self-citationSmart software untangles gene regulation in cells
Perkel, J. M. (2022) · Nature · doi:10.1038/d41586-022-02826-1 · Cited by 3
MEPP: more transparent motif enrichment by profiling positional correlations
Santos, N. P. D., Duttke, S. H. C., Heinz, S., & Benner, C. W. (2022) · NAR Genomics and Bioinformatics · doi:10.1093/nargab/lqac075 · Cited by 3
EvoAug-TF: extending evolution-inspired data augmentations for genomic deep learning to TensorFlow
Yu, Y., Muthukumar, S., & Koo, P. K. (2024) · Bioinformatics · doi:10.1093/bioinformatics/btae092 · Cited by 3
Predicting cell population-specific gene expression from genomic sequence
Michielsen, L., Reinders, M. J. T., & Mahfouz, A. E. (2024) · Frontiers in Bioinformatics · doi:10.3389/fbinf.2024.1347276 · Cited by 6
“Several computational methods have been developed for this task (Kelley et al., 2016; Kelley et al., 2018; Zhou et al., 2018; Agarwal and Shendure, 2020; Zhang et al., 2020; Avsec et al., 2021a; Avsec et al., 2021b).”
BackgroundMethodInfluentialMoeckel, C. M., Mouratidis, I., Chantzi, N., Uzun, Y., & Georgakopoulos‐Soares, I. (2024) · BioEssays · doi:10.1002/bies.202300210 · Cited by 4
Yang, J., Adhikari, S. D., Wang, H., Huang, B., Qi, W., Cui, Y., & Wang, J. (2024) · Frontiers in Systems Biology · doi:10.3389/fsysb.2024.1402664 · Cited by 5
Identification of transcription factor co-binding patterns with non-negative matrix factorization
Rauluševičiūtė, I., Launay, T., Barzaghi, G., Nikumbh, S., Lenhard, B., Krebs, A. R., Castro-Mondragon, J. A., & Mathelier, A. (2024) · Nucleic Acids Research · doi:10.1093/nar/gkae743 · Cited by 5
“Finally, deep learning approaches can infer regulatory patterns from experimental data, with the capacity to pinpoint TF co-binding events (12, 27).”
BackgroundSong, W., & Ovcharenko, I. (2024) · iScience · doi:10.1016/j.isci.2024.111658 · Cited by 2
“15 scBasset predicts activating and repressive TFs according to expression-activity correlation at single-cell resolution.”
Predicting TF–Target Gene Association Using a Heterogeneous Network and Enhanced Negative Sampling
Le, T. T., & Dang, X. T. (2025) · Bioinformatics and Biology Insights · doi:10.1177/11779322251316130 · Cited by 3
“Among the CNN-based methods, Avsec et al 6 introduced the BPNet model.”
Liu, X., Wang, C., Sun, X., Qiao, Z., Yang, X., & Liu, Y. (2025) · Poultry Science · doi:10.1016/j.psj.2025.105096 · Cited by 3
“DNA motif analysis revealed a set of potential transcription TFs (Avsec et al., 2021; Wang et al., 2021), which might bind to DNA regions associated with genes involved in cell cycle and adipogenesis, thereby influencing AFT deposition.”
Multinomial Convolutions for Joint Modeling of Regulatory Motifs and Sequence Activity Readouts
Park, M., Singh, S., Khan, S. R., Abrar, M. A., Grisanti, F., Rahman, M. S., & Samee, M. A. H. (2022) · Genes · doi:10.3390/genes13091614 · Cited by 3
Vahab, N., Bonu, T., Kuhlmann, L., Ramialison, M., & Tyagi, S. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.108068 · Cited by 3
How DNA encodes the start of transcription
Wang, J., & Agarwal, V. (2024) · Science · doi:10.1126/science.adp0869 · Cited by 3
SCIG: Machine learning uncovers cell identity genes in single cells by genetic sequence codes
Kulandaisamy, A., Xia, B., Yu, Y., Chen, H., Pu, W. T., Zhang, L., & Chen, K. (2025) · Nucleic Acids Research · doi:10.1093/nar/gkaf431 · Cited by 4
“Recent studies demonstrated that ge-netic sequence signatures hold promise to be cis-regulatory codes for some aspects of biological function, e.g. RNA structure [ 22–24 ], protein-binding [ 25 ], RNA stability [ 26 ], and promoters and enhancers activities [ 27 , 28 ].”
Dunkenberger, L., Li, D. Y., Karakikes, I., & Quertermous, T. (2025) · Trends in Cell Biology · doi:10.1016/j.tcb.2025.05.001 · Cited by 3
A High-Throughput Genome-Integrated Assay Reveals Spatial Dependencies Governing Tcf7l2 Binding
Szczesnik, T., Chu, L., Ho, J. W. K., & Sherwood, R. I. (2020) · Cell Systems · doi:10.1016/j.cels.2020.08.004 · Cited by 3
Metadata
- Authors
- Avsec, Ž., Weilert, M., Shrikumar, A., Krueger, S., Alexandari, A., Dalal, K., Fropf, R., McAnany, C., Gagneur, J., Kundaje, A., & Zeitlinger, J.
- Affiliations
- Google DeepMind (United Kingdom) (GB)
- Stowers Institute for Medical Research (US)
- Stanford University (US)
- Technical University of Munich (DE)
- Corresponding author
- Kundaje, Anshul
- Zeitlinger, Julia
- ORCID
- Venue
- Nature Genetics
- Published
- 2021-02-18
- Volume
- 53
- Issue
- 3
- Pages
- 354-366
- ISSN
- 1061-4036 (ISSN-L)
- 1546-1718
- PMC
- PMC8812996
- OpenAlex
- W3130819465
- Cited by
- 610
- References
- 163
- Citations by year
- 2026: 118 · 2025: 108 · 2024: 124 · 2023: 102 · 2022: 90 · 2021: 49
- FWCI
- 29.88
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Licence
- Locations
- Nature Genetics (publishedVersion)
- bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- Funders
- U.S. Department of Health & Human Services | NIH | National Human Genome Research Institute (1R01HG010211, 1U01HG009431, 1R01HG009674)
- Howard Hughes Medical Institute (International Student Research Fellowship)
- U.S. Department of Health & Human Services | NIH | National Institute of General Medical Sciences (1DP2GM123485)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Chromatin Dynamics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene Regulatory Network Analysis
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Epigenetics and DNA Methylation
- Keywords
- transcription factor binding, pluripotency transcription factors, Nanog, helical periodicity, explainable deep learning
- MeSH terms
- Mouse Embryonic Stem Cells, Nanog Homeobox Protein, Deep Learning, Animals, Binding Sites, Transcription Factors, Reproducibility of Results, Neural Networks, Computer, Computational Biology, Chromatin Immunoprecipitation, Octamer Transcription Factor-3, Mice, SOXB1 Transcription Factors, Nucleotide Motifs, Clustered Regularly Interspaced Short Palindromic Repeats
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
