1
|
Sokolova K, Chen KM, Hao Y, Zhou J, Troyanskaya OG. Deep Learning Sequence Models for Transcriptional Regulation. Annu Rev Genomics Hum Genet 2024; 25:105-122. [PMID: 38594933 DOI: 10.1146/annurev-genom-021623-024727] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 04/11/2024]
Abstract
Deciphering the regulatory code of gene expression and interpreting the transcriptional effects of genome variation are critical challenges in human genetics. Modern experimental technologies have resulted in an abundance of data, enabling the development of sequence-based deep learning models that link patterns embedded in DNA to the biochemical and regulatory properties contributing to transcriptional regulation, including modeling epigenetic marks, 3D genome organization, and gene expression, with tissue and cell-type specificity. Such methods can predict the functional consequences of any noncoding variant in the human genome, even rare or never-before-observed variants, and systematically characterize their consequences beyond what is tractable from experiments or quantitative genetics studies alone. Recently, the development and application of interpretability approaches have led to the identification of key sequence patterns contributing to the predicted tasks, providing insights into the underlying biological mechanisms learned and revealing opportunities for improvement in future models.
Collapse
Affiliation(s)
- Ksenia Sokolova
- Department of Computer Science and Lewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, New Jersey, USA; , ,
| | - Kathleen M Chen
- Department of Computer Science and Lewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, New Jersey, USA; , ,
| | - Yun Hao
- Flatiron Institute, Simons Foundation, New York, NY, USA;
| | - Jian Zhou
- Lyda Hill Department of Bioinformatics, University of Texas Southwestern Medical Center, Dallas, Texas, USA;
| | - Olga G Troyanskaya
- Princeton Precision Health, Princeton University, Princeton, New Jersey, USA
- Flatiron Institute, Simons Foundation, New York, NY, USA;
- Department of Computer Science and Lewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, New Jersey, USA; , ,
| |
Collapse
|
2
|
Thompson M, Martín M, Olmo TS, Rajesh C, Koo PK, Bolognesi B, Lehner B. Interpretably deep learning amyloid nucleation by massive experimental quantification of random sequences. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2024:2024.07.13.603366. [PMID: 39071305 PMCID: PMC11275847 DOI: 10.1101/2024.07.13.603366] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 07/30/2024]
Abstract
Insoluble amyloid aggregates are the hallmarks of more than fifty human diseases, including the most common neurodegenerative disorders. The process by which soluble proteins nucleate to form amyloid fibrils is, however, quite poorly characterized. Relatively few sequences are known that form amyloids with high propensity and this data shortage likely limits our capacity to understand, predict, engineer, and prevent the formation of amyloid fibrils. Here we quantify the nucleation of amyloids at an unprecedented scale and use the data to train a deep learning model of amyloid nucleation. In total, we quantify the nucleation rates of >100,000 20-amino-acid-long peptides. This large and diverse dataset allows us to train CANYA, a convolution-attention hybrid neural network. CANYA is fast and outperforms existing methods with stable performance across diverse prediction tasks. Interpretability analyses reveal CANYA's decision-making process and learned grammar, providing mechanistic insights into amyloid nucleation. Our results illustrate the power of massive experimental analysis of random sequence-spaces and provide an interpretable and robust neural network model to predict amyloid nucleation.
Collapse
|
3
|
Takou M, Bellis ES, Lasky JR. Predicting gene expression responses to environment in Arabidopsis thaliana using natural variation in DNA sequence. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2024:2024.04.25.591174. [PMID: 38712066 PMCID: PMC11071634 DOI: 10.1101/2024.04.25.591174] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/08/2024]
Abstract
The evolution of gene expression responses are a critical component of adaptation to variable environments. Predicting how DNA sequence influences expression is challenging because the genotype to phenotype map is not well resolved for cis regulatory elements, transcription factor binding, regulatory interactions, and epigenetic features, not to mention how these factors respond to environment. We tested if flexible machine learning models could learn some of the underlying cis- regulatory genotype to phenotype map. We tested this approach using cold-responsive transcriptome profiles in 5 diverse Arabidopsis thaliana accessions. We first tested for evidence that cis regulation plays a role in environmental response, finding 14 and 15 motifs that were significantly enriched within the up- and down-stream regions of cold-responsive differentially regulated genes (DEGs). We next applied convolutional neural networks (CNNs), which learn de novo cis- regulatory motifs in DNA sequences to predict expression response to environment. We found that CNNs predicted differential expression with moderate accuracy, with evidence that predictions were hindered by biological complexity of regulation and the large potential regulatory code. Overall, DEGs between specific environments can be predicted based on variation in cis- regulatory sequences, although more information needs to be incorporated and better models may be required.
Collapse
|
4
|
He AY, Danko CG. Dissection of core promoter syntax through single nucleotide resolution modeling of transcription initiation. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2024:2024.03.13.583868. [PMID: 38559255 PMCID: PMC10979970 DOI: 10.1101/2024.03.13.583868] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 04/04/2024]
Abstract
Our understanding of how the DNA sequences of cis-regulatory elements encode transcription initiation patterns remains limited. Here we introduce CLIPNET, a deep learning model trained on population-scale PRO-cap data that accurately predicts the position and quantity of transcription initiation with single nucleotide resolution from DNA sequence. Interpretation of CLIPNET revealed a complex regulatory syntax consisting of DNA-protein interactions in five major positions between -200 and +50 bp relative to the transcription start site, as well as more subtle positional preferences among different transcriptional activators. Transcriptional activator and core promoter motifs occupy different positions and play distinct roles in regulating initiation, with the former driving initiation quantity and the latter initiation position. We identified core promoter motifs that explain initiation patterns in the majority of promoters and enhancers, including DPR motifs and AT-rich TBP binding sequences in TATA-less promoters. Our results provide insights into the sequence architecture governing transcription initiation.
Collapse
Affiliation(s)
- Adam Y. He
- Baker Institute for Animal Health, College of Veterinary Medicine, Cornell University
- Graduate Field of Computational Biology, Cornell University
| | - Charles G. Danko
- Baker Institute for Animal Health, College of Veterinary Medicine, Cornell University
- Department of Biomedical Sciences, College of Veterinary Medicine, Cornell University
| |
Collapse
|
5
|
Toneyan S, Koo PK. Interpreting Cis-Regulatory Interactions from Large-Scale Deep Neural Networks for Genomics. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2024:2023.07.03.547592. [PMID: 37461616 PMCID: PMC10349992 DOI: 10.1101/2023.07.03.547592] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Indexed: 07/28/2023]
Abstract
The rise of large-scale, sequence-based deep neural networks (DNNs) for predicting gene expression has introduced challenges in their evaluation and interpretation. Current evaluations align DNN predictions with experimental perturbation assays, which provides insights into the generalization capabilities within the studied loci but offers a limited perspective of what drives their predictions. Moreover, existing model explainability tools focus mainly on motif analysis, which becomes complex when interpreting longer sequences. Here we introduce CREME, an in silico perturbation toolkit that interrogates large-scale DNNs to uncover rules of gene regulation that it learns. Using CREME, we investigate Enformer, a prominent DNN in gene expression prediction, revealing cis-regulatory elements (CREs) that directly enhance or silence target genes. We explore the intricate complexity of higher-order CRE interactions, the relationship between CRE distance from transcription start sites on gene expression, as well as the biochemical features of enhancers and silencers learned by Enformer. Moreover, we demonstrate the flexibility of CREME to efficiently uncover a higher-resolution view of functional sequence elements within CREs. This work demonstrates how CREME can be employed to translate the powerful predictions of large-scale DNNs to study open questions in gene regulation.
Collapse
|
6
|
Tang Z, Koo PK. Evaluating the representational power of pre-trained DNA language models for regulatory genomics. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2024:2024.02.29.582810. [PMID: 38464101 PMCID: PMC10925287 DOI: 10.1101/2024.02.29.582810] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 03/12/2024]
Abstract
The emergence of genomic language models (gLMs) offers an unsupervised approach to learn a wide diversity of cis-regulatory patterns in the non-coding genome without requiring labels of functional activity generated by wet-lab experiments. Previous evaluations have shown pre-trained gLMs can be leveraged to improve prediction performance across a broad range of regulatory genomics tasks, albeit using relatively simple benchmark datasets and baseline models. Since the gLMs in these studies were tested upon fine-tuning their weights for each downstream task, determining whether gLM representations embody a foundational understanding of cis-regulatory biology remains an open question. Here we evaluate the representational power of pre-trained gLMs to predict and interpret cell-type-specific functional genomics data that span DNA and RNA regulation. Our findings suggest that current gLMs do not offer substantial advantages over conventional machine learning approaches that use one-hot encoded sequences. This work highlights a major limitation with current gLMs, raising potential issues in conventional pre-training strategies for the non-coding genome.
Collapse
Affiliation(s)
- Ziqi Tang
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, NY, USA
| | - Peter K Koo
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, NY, USA
| |
Collapse
|
7
|
Taskiran II, Spanier KI, Dickmänken H, Kempynck N, Pančíková A, Ekşi EC, Hulselmans G, Ismail JN, Theunis K, Vandepoel R, Christiaens V, Mauduit D, Aerts S. Cell-type-directed design of synthetic enhancers. Nature 2024; 626:212-220. [PMID: 38086419 PMCID: PMC10830415 DOI: 10.1038/s41586-023-06936-2] [Citation(s) in RCA: 18] [Impact Index Per Article: 18.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/06/2022] [Accepted: 12/05/2023] [Indexed: 01/19/2024]
Abstract
Transcriptional enhancers act as docking stations for combinations of transcription factors and thereby regulate spatiotemporal activation of their target genes1. It has been a long-standing goal in the field to decode the regulatory logic of an enhancer and to understand the details of how spatiotemporal gene expression is encoded in an enhancer sequence. Here we show that deep learning models2-6, can be used to efficiently design synthetic, cell-type-specific enhancers, starting from random sequences, and that this optimization process allows detailed tracing of enhancer features at single-nucleotide resolution. We evaluate the function of fully synthetic enhancers to specifically target Kenyon cells or glial cells in the fruit fly brain using transgenic animals. We further exploit enhancer design to create 'dual-code' enhancers that target two cell types and minimal enhancers smaller than 50 base pairs that are fully functional. By examining the state space searches towards local optima, we characterize enhancer codes through the strength, combination and arrangement of transcription factor activator and transcription factor repressor motifs. Finally, we apply the same strategies to successfully design human enhancers, which adhere to enhancer rules similar to those of Drosophila enhancers. Enhancer design guided by deep learning leads to better understanding of how enhancers work and shows that their code can be exploited to manipulate cell states.
Collapse
Affiliation(s)
- Ibrahim I Taskiran
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
| | - Katina I Spanier
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
| | - Hannah Dickmänken
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
| | - Niklas Kempynck
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
| | - Alexandra Pančíková
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
- VIB-KULeuven Center for Cancer Biology, Leuven, Belgium
| | - Eren Can Ekşi
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
| | - Gert Hulselmans
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
| | - Joy N Ismail
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
- UK Dementia Research Institute at Imperial College London, London, UK
| | - Koen Theunis
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
| | - Roel Vandepoel
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
| | - Valerie Christiaens
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
| | - David Mauduit
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium
- Department of Human Genetics, KU Leuven, Leuven, Belgium
| | - Stein Aerts
- Laboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, Belgium.
- VIB-KULeuven Center for Brain & Disease Research, Leuven, Belgium.
- Department of Human Genetics, KU Leuven, Leuven, Belgium.
| |
Collapse
|
8
|
de Boer CG, Taipale J. Hold out the genome: a roadmap to solving the cis-regulatory code. Nature 2024; 625:41-50. [PMID: 38093018 DOI: 10.1038/s41586-023-06661-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/17/2023] [Accepted: 09/20/2023] [Indexed: 01/05/2024]
Abstract
Gene expression is regulated by transcription factors that work together to read cis-regulatory DNA sequences. The 'cis-regulatory code' - how cells interpret DNA sequences to determine when, where and how much genes should be expressed - has proven to be exceedingly complex. Recently, advances in the scale and resolution of functional genomics assays and machine learning have enabled substantial progress towards deciphering this code. However, the cis-regulatory code will probably never be solved if models are trained only on genomic sequences; regions of homology can easily lead to overestimation of predictive performance, and our genome is too short and has insufficient sequence diversity to learn all relevant parameters. Fortunately, randomly synthesized DNA sequences enable testing a far larger sequence space than exists in our genomes, and designed DNA sequences enable targeted queries to maximally improve the models. As the same biochemical principles are used to interpret DNA regardless of its source, models trained on these synthetic data can predict genomic activity, often better than genome-trained models. Here we provide an outlook on the field, and propose a roadmap towards solving the cis-regulatory code by a combination of machine learning and massively parallel assays using synthetic DNA.
Collapse
Affiliation(s)
- Carl G de Boer
- School of Biomedical Engineering, University of British Columbia, Vancouver, British Columbia, Canada.
| | - Jussi Taipale
- Applied Tumor Genomics Research Program, Faculty of Medicine, University of Helsinki, Helsinki, Finland.
- Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden.
- Department of Biochemistry, University of Cambridge, Cambridge, UK.
| |
Collapse
|
9
|
Kim GB, Kim JY, Lee JA, Norsigian CJ, Palsson BO, Lee SY. Functional annotation of enzyme-encoding genes using deep learning with transformer layers. Nat Commun 2023; 14:7370. [PMID: 37963869 PMCID: PMC10645960 DOI: 10.1038/s41467-023-43216-z] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/28/2023] [Accepted: 11/03/2023] [Indexed: 11/16/2023] Open
Abstract
Functional annotation of open reading frames in microbial genomes remains substantially incomplete. Enzymes constitute the most prevalent functional gene class in microbial genomes and can be described by their specific catalytic functions using the Enzyme Commission (EC) number. Consequently, the ability to predict EC numbers could substantially reduce the number of un-annotated genes. Here we present a deep learning model, DeepECtransformer, which utilizes transformer layers as a neural network architecture to predict EC numbers. Using the extensively studied Escherichia coli K-12 MG1655 genome, DeepECtransformer predicted EC numbers for 464 un-annotated genes. We experimentally validated the enzymatic activities predicted for three proteins (YgfF, YciO, and YjdM). Further examination of the neural network's reasoning process revealed that the trained neural network relies on functional motifs of enzymes to predict EC numbers. Thus, DeepECtransformer is a method that facilitates the functional annotation of uncharacterized genes.
Collapse
Affiliation(s)
- Gi Bae Kim
- Metabolic and Biomolecular Engineering National Research Laboratory, Department of Chemical and Biomolecular Engineering (BK21 four), Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea
- Systems Metabolic Engineering and Systems Healthcare Cross-Generation Collaborative Laboratory, Department of Chemical and Biomolecular Engineering (BK21 four), KAIST, Daejeon, 34141, Republic of Korea
- KAIST Institute for the BioCentury and KAIST Institute for Artificial Intelligence, KAIST, Daejeon, 34141, Republic of Korea
| | - Ji Yeon Kim
- Metabolic and Biomolecular Engineering National Research Laboratory, Department of Chemical and Biomolecular Engineering (BK21 four), Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea
- Systems Metabolic Engineering and Systems Healthcare Cross-Generation Collaborative Laboratory, Department of Chemical and Biomolecular Engineering (BK21 four), KAIST, Daejeon, 34141, Republic of Korea
- KAIST Institute for the BioCentury and KAIST Institute for Artificial Intelligence, KAIST, Daejeon, 34141, Republic of Korea
| | - Jong An Lee
- Metabolic and Biomolecular Engineering National Research Laboratory, Department of Chemical and Biomolecular Engineering (BK21 four), Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea
- Systems Metabolic Engineering and Systems Healthcare Cross-Generation Collaborative Laboratory, Department of Chemical and Biomolecular Engineering (BK21 four), KAIST, Daejeon, 34141, Republic of Korea
- KAIST Institute for the BioCentury and KAIST Institute for Artificial Intelligence, KAIST, Daejeon, 34141, Republic of Korea
| | - Charles J Norsigian
- Division of Biological Sciences, University of California San Diego, La Jolla, CA, 92093, USA
- Department of Bioengineering, University of California San Diego, La Jolla, CA, 92093, USA
| | - Bernhard O Palsson
- Department of Bioengineering, University of California San Diego, La Jolla, CA, 92093, USA
- Bioinformatics and Systems Biology Program, University of California San Diego, La Jolla, CA, 92093, USA
- Novo Nordisk Foundation Center for Biosustainability, 2800, Kongens Lyngby, Denmark
| | - Sang Yup Lee
- Metabolic and Biomolecular Engineering National Research Laboratory, Department of Chemical and Biomolecular Engineering (BK21 four), Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea.
- Systems Metabolic Engineering and Systems Healthcare Cross-Generation Collaborative Laboratory, Department of Chemical and Biomolecular Engineering (BK21 four), KAIST, Daejeon, 34141, Republic of Korea.
- KAIST Institute for the BioCentury and KAIST Institute for Artificial Intelligence, KAIST, Daejeon, 34141, Republic of Korea.
- BioProcess Engineering Research Center and BioInformatics Research Center, KAIST, Daejeon, 34141, Republic of Korea.
| |
Collapse
|
10
|
Schleussner N, Cauchy P, Franke V, Giefing M, Fornes O, Vankadari N, Assi SA, Costanza M, Weniger MA, Akalin A, Anagnostopoulos I, Bukur T, Casarotto MG, Damm F, Daumke O, Edginton-White B, Gebhardt JCM, Grau M, Grunwald S, Hansmann ML, Hartmann S, Huber L, Kärgel E, Lusatis S, Noerenberg D, Obier N, Pannicke U, Fischer A, Reisser A, Rosenwald A, Schwarz K, Sundararaj S, Weilemann A, Winkler W, Xu W, Lenz G, Rajewsky K, Wasserman WW, Cockerill PN, Scheidereit C, Siebert R, Küppers R, Grosschedl R, Janz M, Bonifer C, Mathas S. Transcriptional reprogramming by mutated IRF4 in lymphoma. Nat Commun 2023; 14:6947. [PMID: 37935654 PMCID: PMC10630337 DOI: 10.1038/s41467-023-41954-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/29/2022] [Accepted: 09/20/2023] [Indexed: 11/09/2023] Open
Abstract
Disease-causing mutations in genes encoding transcription factors (TFs) can affect TF interactions with their cognate DNA-binding motifs. Whether and how TF mutations impact upon the binding to TF composite elements (CE) and the interaction with other TFs is unclear. Here, we report a distinct mechanism of TF alteration in human lymphomas with perturbed B cell identity, in particular classic Hodgkin lymphoma. It is caused by a recurrent somatic missense mutation c.295 T > C (p.Cys99Arg; p.C99R) targeting the center of the DNA-binding domain of Interferon Regulatory Factor 4 (IRF4), a key TF in immune cells. IRF4-C99R fundamentally alters IRF4 DNA-binding, with loss-of-binding to canonical IRF motifs and neomorphic gain-of-binding to canonical and non-canonical IRF CEs. IRF4-C99R thoroughly modifies IRF4 function by blocking IRF4-dependent plasma cell induction, and up-regulates disease-specific genes in a non-canonical Activator Protein-1 (AP-1)-IRF-CE (AICE)-dependent manner. Our data explain how a single mutation causes a complex switch of TF specificity and gene regulation and open the perspective to specifically block the neomorphic DNA-binding activities of a mutant TF.
Collapse
Affiliation(s)
- Nikolai Schleussner
- Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Biology of Malignant Lymphomas, 13125, Berlin, Germany
- Hematology, Oncology, and Cancer Immunology, Charité - Universitätsmedizin Berlin, Corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, 10117, Berlin, Germany
- Experimental and Clinical Research Center (ECRC), a joint cooperation between Charité and MDC, Berlin, Germany
| | - Pierre Cauchy
- Max Planck Institute of Immunobiology and Epigenetics, 79108, Freiburg, Germany
- Institute of Cancer and Genomic Sciences, College of Medical and Dental Sciences, University of Birmingham, Birmingham, B15 2TT, UK
- University Medical Center Freiburg, 79106, Freiburg, Germany
- German Cancer Consortium (DKTK), German Cancer Research Center (DKFZ), 69120, Heidelberg, Germany
| | - Vedran Franke
- Bioinformatics and Omics Data Science Platform, Berlin Institute for Medical Systems Biology, Max-Delbrück-Center, Berlin, Germany
| | - Maciej Giefing
- Institute of Human Genetics, Polish Academy of Sciences, Poznan, 60-479, Poland
- Institute of Human Genetics, Christian-Albrechts-University Kiel, 24105, Kiel, Germany
| | - Oriol Fornes
- Centre for Molecular Medicine and Therapeutics, Department of Medical Genetics, BC Children's Hospital Research Institute, University of British Columbia, Vancouver, BC, V5Z 4H4, Canada
| | - Naveen Vankadari
- Department of Biochemistry and Pharmacology, Bio21 Molecular Science and Biotechnology Institute, The University of Melbourne, Melbourne, VIC, 3000, Australia
| | - Salam A Assi
- Institute of Cancer and Genomic Sciences, College of Medical and Dental Sciences, University of Birmingham, Birmingham, B15 2TT, UK
| | - Mariantonia Costanza
- Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Biology of Malignant Lymphomas, 13125, Berlin, Germany
- Hematology, Oncology, and Cancer Immunology, Charité - Universitätsmedizin Berlin, Corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, 10117, Berlin, Germany
- Experimental and Clinical Research Center (ECRC), a joint cooperation between Charité and MDC, Berlin, Germany
| | - Marc A Weniger
- Institute of Cell Biology (Cancer Research), University of Duisburg-Essen, 45122, Essen, Germany
| | - Altuna Akalin
- Bioinformatics and Omics Data Science Platform, Berlin Institute for Medical Systems Biology, Max-Delbrück-Center, Berlin, Germany
| | - Ioannis Anagnostopoulos
- Institute of Pathology, Universität Würzburg and Comprehensive Cancer Centre Mainfranken (CCCMF), Würzburg, Germany
| | - Thomas Bukur
- TRON gGmbH - Translationale Onkologie an der Universitätsmedizin der Johannes Gutenberg-Universität Mainz, Mainz, Germany
| | - Marco G Casarotto
- Research School of Biology, The Australian National University, Canberra, ACT, Australia
| | - Frederik Damm
- Hematology, Oncology, and Cancer Immunology, Charité - Universitätsmedizin Berlin, Corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, 10117, Berlin, Germany
| | - Oliver Daumke
- Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Structural Biology, 13125, Berlin, Germany
| | - Benjamin Edginton-White
- Institute of Cancer and Genomic Sciences, College of Medical and Dental Sciences, University of Birmingham, Birmingham, B15 2TT, UK
| | | | - Michael Grau
- Department of Physics, University of Marburg, 35052, Marburg, Germany
- Medical Department A for Hematology, Oncology and Pneumology, University Hospital Münster, Münster, Germany
| | - Stephan Grunwald
- Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Structural Biology, 13125, Berlin, Germany
| | - Martin-Leo Hansmann
- Frankfurt Institute of Advanced Studies, Frankfurt am Main, Germany
- Institute for Pharmacology and Toxicology, Goethe University, Frankfurt am Main, Germany
| | - Sylvia Hartmann
- Dr. Senckenberg Institute of Pathology, Goethe University Frankfurt, Frankfurt am Main, Germany
| | - Lionel Huber
- Max Planck Institute of Immunobiology and Epigenetics, 79108, Freiburg, Germany
| | - Eva Kärgel
- Signal Transduction in Tumor Cells, Max-Delbrück-Center for Molecular Medicine, Berlin, Germany
| | - Simone Lusatis
- Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Biology of Malignant Lymphomas, 13125, Berlin, Germany
- Hematology, Oncology, and Cancer Immunology, Charité - Universitätsmedizin Berlin, Corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, 10117, Berlin, Germany
- Experimental and Clinical Research Center (ECRC), a joint cooperation between Charité and MDC, Berlin, Germany
| | - Daniel Noerenberg
- Hematology, Oncology, and Cancer Immunology, Charité - Universitätsmedizin Berlin, Corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, 10117, Berlin, Germany
| | - Nadine Obier
- Max Planck Institute of Immunobiology and Epigenetics, 79108, Freiburg, Germany
- Institute of Cancer and Genomic Sciences, College of Medical and Dental Sciences, University of Birmingham, Birmingham, B15 2TT, UK
| | - Ulrich Pannicke
- Institute for Transfusion Medicine, University of Ulm, Ulm, Germany
| | - Anja Fischer
- Institute of Human Genetics, Ulm University and Ulm University Medical Center, 89081, Ulm, Germany
| | - Anja Reisser
- Department of Physics, Institute of Biophysics, Ulm University, Ulm, Germany
| | - Andreas Rosenwald
- Institute of Pathology, Universität Würzburg and Comprehensive Cancer Centre Mainfranken (CCCMF), Würzburg, Germany
| | - Klaus Schwarz
- Institute for Transfusion Medicine, University of Ulm, Ulm, Germany
- Institute for Clinical Transfusion Medicine and Immunogenetics Ulm, German Red Cross Blood Service Baden-Württemberg-Hessen, Ulm, Germany
| | - Srinivasan Sundararaj
- Research School of Biology, The Australian National University, Canberra, ACT, Australia
| | - Andre Weilemann
- Medical Department A for Hematology, Oncology and Pneumology, University Hospital Münster, Münster, Germany
| | - Wiebke Winkler
- Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Biology of Malignant Lymphomas, 13125, Berlin, Germany
- Hematology, Oncology, and Cancer Immunology, Charité - Universitätsmedizin Berlin, Corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, 10117, Berlin, Germany
- Experimental and Clinical Research Center (ECRC), a joint cooperation between Charité and MDC, Berlin, Germany
| | - Wendan Xu
- Medical Department A for Hematology, Oncology and Pneumology, University Hospital Münster, Münster, Germany
| | - Georg Lenz
- Medical Department A for Hematology, Oncology and Pneumology, University Hospital Münster, Münster, Germany
| | - Klaus Rajewsky
- Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Immune Regulation and Cancer, 13125, Berlin, Germany
| | - Wyeth W Wasserman
- Centre for Molecular Medicine and Therapeutics, Department of Medical Genetics, BC Children's Hospital Research Institute, University of British Columbia, Vancouver, BC, V5Z 4H4, Canada
| | - Peter N Cockerill
- Institute of Cancer and Genomic Sciences, College of Medical and Dental Sciences, University of Birmingham, Birmingham, B15 2TT, UK
| | - Claus Scheidereit
- Signal Transduction in Tumor Cells, Max-Delbrück-Center for Molecular Medicine, Berlin, Germany
| | - Reiner Siebert
- Institute of Human Genetics, Christian-Albrechts-University Kiel, 24105, Kiel, Germany
- Institute of Human Genetics, Ulm University and Ulm University Medical Center, 89081, Ulm, Germany
| | - Ralf Küppers
- German Cancer Consortium (DKTK), German Cancer Research Center (DKFZ), 69120, Heidelberg, Germany
- Institute of Cell Biology (Cancer Research), University of Duisburg-Essen, 45122, Essen, Germany
| | - Rudolf Grosschedl
- Max Planck Institute of Immunobiology and Epigenetics, 79108, Freiburg, Germany
| | - Martin Janz
- Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Biology of Malignant Lymphomas, 13125, Berlin, Germany
- Hematology, Oncology, and Cancer Immunology, Charité - Universitätsmedizin Berlin, Corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, 10117, Berlin, Germany
- Experimental and Clinical Research Center (ECRC), a joint cooperation between Charité and MDC, Berlin, Germany
| | - Constanze Bonifer
- Institute of Cancer and Genomic Sciences, College of Medical and Dental Sciences, University of Birmingham, Birmingham, B15 2TT, UK
| | - Stephan Mathas
- Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Biology of Malignant Lymphomas, 13125, Berlin, Germany.
- Hematology, Oncology, and Cancer Immunology, Charité - Universitätsmedizin Berlin, Corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, 10117, Berlin, Germany.
- Experimental and Clinical Research Center (ECRC), a joint cooperation between Charité and MDC, Berlin, Germany.
- German Cancer Consortium (DKTK), German Cancer Research Center (DKFZ), 69120, Heidelberg, Germany.
| |
Collapse
|
11
|
Klie A, Laub D, Talwar JV, Stites H, Jores T, Solvason JJ, Farley EK, Carter H. Predictive analyses of regulatory sequences with EUGENe. NATURE COMPUTATIONAL SCIENCE 2023; 3:946-956. [PMID: 38177592 PMCID: PMC10768637 DOI: 10.1038/s43588-023-00544-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 01/12/2023] [Accepted: 09/27/2023] [Indexed: 01/06/2024]
Abstract
Deep learning has become a popular tool to study cis-regulatory function. Yet efforts to design software for deep-learning analyses in regulatory genomics that are findable, accessible, interoperable and reusable (FAIR) have fallen short of fully meeting these criteria. Here we present elucidating the utility of genomic elements with neural nets (EUGENe), a FAIR toolkit for the analysis of genomic sequences with deep learning. EUGENe consists of a set of modules and subpackages for executing the key functionality of a genomics deep learning workflow: (1) extracting, transforming and loading sequence data from many common file formats; (2) instantiating, initializing and training diverse model architectures; and (3) evaluating and interpreting model behavior. We designed EUGENe as a simple, flexible and extensible interface for streamlining and customizing end-to-end deep-learning sequence analyses, and illustrate these principles through application of the toolkit to three predictive modeling tasks. We hope that EUGENe represents a springboard towards a collaborative ecosystem for deep-learning applications in genomics research.
Collapse
Affiliation(s)
- Adam Klie
- Department of Medicine, University of California San Diego, La Jolla, CA, USA
- Bioinformatics and Systems Biology Program, University of California San Diego, La Jolla, CA, USA
| | - David Laub
- Department of Medicine, University of California San Diego, La Jolla, CA, USA
- Bioinformatics and Systems Biology Program, University of California San Diego, La Jolla, CA, USA
| | - James V Talwar
- Department of Medicine, University of California San Diego, La Jolla, CA, USA
- Bioinformatics and Systems Biology Program, University of California San Diego, La Jolla, CA, USA
| | | | - Tobias Jores
- Department of Genome Sciences, University of Washington, Seattle, WA, USA
| | - Joe J Solvason
- Department of Medicine, University of California San Diego, La Jolla, CA, USA
- Bioinformatics and Systems Biology Program, University of California San Diego, La Jolla, CA, USA
- Department of Molecular Biology, University of California San Diego, La Jolla, CA, USA
| | - Emma K Farley
- Department of Medicine, University of California San Diego, La Jolla, CA, USA
- Bioinformatics and Systems Biology Program, University of California San Diego, La Jolla, CA, USA
- Department of Molecular Biology, University of California San Diego, La Jolla, CA, USA
| | - Hannah Carter
- Department of Medicine, University of California San Diego, La Jolla, CA, USA.
- Bioinformatics and Systems Biology Program, University of California San Diego, La Jolla, CA, USA.
| |
Collapse
|
12
|
Stanfill SB, Hecht SS, Joerger AC, González PJ, Maia LB, Rivas MG, Moura JJG, Gupta AK, Le Brun NE, Crack JC, Hainaut P, Sparacino-Watkins C, Tyx RE, Pillai SD, Zaatari GS, Henley SJ, Blount BC, Watson CH, Kaina B, Mehrotra R. From cultivation to cancer: formation of N-nitrosamines and other carcinogens in smokeless tobacco and their mutagenic implications. Crit Rev Toxicol 2023; 53:658-701. [PMID: 38050998 DOI: 10.1080/10408444.2023.2264327] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/21/2023] [Accepted: 09/20/2023] [Indexed: 12/07/2023]
Abstract
Tobacco use is a major cause of preventable morbidity and mortality globally. Tobacco products, including smokeless tobacco (ST), generally contain tobacco-specific N-nitrosamines (TSNAs), such as N'-nitrosonornicotine (NNN) and 4-(methylnitrosamino)-1-(3-pyridyl)-butanone (NNK), which are potent carcinogens that cause mutations in critical genes in human DNA. This review covers the series of biochemical and chemical transformations, related to TSNAs, leading from tobacco cultivation to cancer initiation. A key aim of this review is to provide a greater understanding of TSNAs: their precursors, the microbial and chemical mechanisms that contribute to their formation in ST, their mutagenicity leading to cancer due to ST use, and potential means of lowering TSNA levels in tobacco products. TSNAs are not present in harvested tobacco but can form due to nitrosating agents reacting with tobacco alkaloids present in tobacco during certain types of curing. TSNAs can also form during or following ST production when certain microorganisms perform nitrate metabolism, with dissimilatory nitrate reductases converting nitrate to nitrite that is then released into tobacco and reacts chemically with tobacco alkaloids. When ST usage occurs, TSNAs are absorbed and metabolized to reactive compounds that form DNA adducts leading to mutations in critical target genes, including the RAS oncogenes and the p53 tumor suppressor gene. DNA repair mechanisms remove most adducts induced by carcinogens, thus preventing many but not all mutations. Lastly, because TSNAs and other agents cause cancer, previously documented strategies for lowering their levels in ST products are discussed, including using tobacco with lower nornicotine levels, pasteurization and other means of eliminating microorganisms, omitting fermentation and fire-curing, refrigerating ST products, and including nitrite scavenging chemicals as ST ingredients.
Collapse
Affiliation(s)
- Stephen B Stanfill
- Tobacco and Volatiles Branch, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA, USA
| | - Stephen S Hecht
- Masonic Cancer Center, University of Minnesota, Minneapolis, MN, USA
| | - Andreas C Joerger
- Structural Genomics Consortium (SGC), Institute of Pharmaceutical Chemistry, Goethe University Frankfurt, Frankfurt am Main, Germany
| | - Pablo J González
- Department of Physics, Universidad Nacional Litoral, and CONICET, Santa Fe, Argentina
| | - Luisa B Maia
- Department of Chemistry, LAQV, REQUIMTE, NOVA School of Science and Technology (FCT NOVA), Caparica, Portugal
| | - Maria G Rivas
- Department of Physics, Universidad Nacional Litoral, and CONICET, Santa Fe, Argentina
| | - José J G Moura
- Department of Chemistry, LAQV, REQUIMTE, NOVA School of Science and Technology (FCT NOVA), Caparica, Portugal
| | | | - Nick E Le Brun
- School of Chemistry, Centre for Molecular and Structural Biochemistry, University of East Anglia, Norwich, UK
| | - Jason C Crack
- School of Chemistry, Centre for Molecular and Structural Biochemistry, University of East Anglia, Norwich, UK
| | - Pierre Hainaut
- Institute for Advanced Biosciences, Grenoble Alpes University, Grenoble, France
| | - Courtney Sparacino-Watkins
- University of Pittsburgh, School of Medicine, Division of Pulmonary Allergy and Critical Care Medicine, Vascular Medicine Institute, PA, USA
| | - Robert E Tyx
- Tobacco and Volatiles Branch, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA, USA
| | - Suresh D Pillai
- Department of Food Science & Technology, National Center for Electron Beam Research, Texas A&M University, College Station, TX, USA
| | - Ghazi S Zaatari
- Department of Pathology and Laboratory Medicine, American University of Beirut, Beirut, Lebanon
| | - S Jane Henley
- Division of Cancer Prevention and Control, Centers for Disease Control and Prevention, Atlanta, GA, USA
| | - Benjamin C Blount
- Tobacco and Volatiles Branch, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA, USA
| | - Clifford H Watson
- Tobacco and Volatiles Branch, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA, USA
| | - Bernd Kaina
- Institute of Toxicology, University Medical Center, Mainz, Germany
| | - Ravi Mehrotra
- Centre for Health, Innovation and Policy Foundation, Noida, India
| |
Collapse
|
13
|
Nambiar A, Dubinkina V, Liu S, Maslov S. FUN-PROSE: A deep learning approach to predict condition-specific gene expression in fungi. PLoS Comput Biol 2023; 19:e1011563. [PMID: 37971967 PMCID: PMC10653424 DOI: 10.1371/journal.pcbi.1011563] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/19/2023] [Accepted: 09/30/2023] [Indexed: 11/19/2023] Open
Abstract
mRNA levels of all genes in a genome is a critical piece of information defining the overall state of the cell in a given environmental condition. Being able to reconstruct such condition-specific expression in fungal genomes is particularly important to metabolically engineer these organisms to produce desired chemicals in industrially scalable conditions. Most previous deep learning approaches focused on predicting the average expression levels of a gene based on its promoter sequence, ignoring its variation across different conditions. Here we present FUN-PROSE-a deep learning model trained to predict differential expression of individual genes across various conditions using their promoter sequences and expression levels of all transcription factors. We train and test our model on three fungal species and get the correlation between predicted and observed condition-specific gene expression as high as 0.85. We then interpret our model to extract promoter sequence motifs responsible for variable expression of individual genes. We also carried out input feature importance analysis to connect individual transcription factors to their gene targets. A sizeable fraction of both sequence motifs and TF-gene interactions learned by our model agree with previously known biological information, while the rest corresponds to either novel biological facts or indirect correlations.
Collapse
Affiliation(s)
- Ananthan Nambiar
- Department of Bioengineering, University of Illinois Urbana-Champaign, Urbana, Illinois, United States of America
- Carl R. Woese Institute for Genomic Biology, Urbana, Illinois, United States of America
| | - Veronika Dubinkina
- Department of Bioengineering, University of Illinois Urbana-Champaign, Urbana, Illinois, United States of America
- Carl R. Woese Institute for Genomic Biology, Urbana, Illinois, United States of America
- The Gladstone Institute of Data Science and Biotechnology, San Francisco, California, United States of America
| | - Simon Liu
- Carl R. Woese Institute for Genomic Biology, Urbana, Illinois, United States of America
- Department of Computer Science, University of Illinois Urbana-Champaign, Urbana, Illinois, United States of America
| | - Sergei Maslov
- Department of Bioengineering, University of Illinois Urbana-Champaign, Urbana, Illinois, United States of America
- Carl R. Woese Institute for Genomic Biology, Urbana, Illinois, United States of America
- Department of Physics, University of Illinois Urbana-Champaign, Urbana, Illinois, United States of America
- Computing, Environment and Life Sciences, Argonne National Laboratory, Lemont, Illinois, United States of America
| |
Collapse
|
14
|
Friedman RZ, Ramu A, Lichtarge S, Myers CA, Granas DM, Gause M, Corbo JC, Cohen BA, White MA. Active learning of enhancer and silencer regulatory grammar in photoreceptors. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2023:2023.08.21.554146. [PMID: 37662358 PMCID: PMC10473580 DOI: 10.1101/2023.08.21.554146] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 09/05/2023]
Abstract
Cis-regulatory elements (CREs) direct gene expression in health and disease, and models that can accurately predict their activities from DNA sequences are crucial for biomedicine. Deep learning represents one emerging strategy to model the regulatory grammar that relates CRE sequence to function. However, these models require training data on a scale that exceeds the number of CREs in the genome. We address this problem using active machine learning to iteratively train models on multiple rounds of synthetic DNA sequences assayed in live mammalian retinas. During each round of training the model actively selects sequence perturbations to assay, thereby efficiently generating informative training data. We iteratively trained a model that predicts the activities of sequences containing binding motifs for the photoreceptor transcription factor Cone-rod homeobox (CRX) using an order of magnitude less training data than current approaches. The model's internal confidence estimates of its predictions are reliable guides for designing sequences with high activity. The model correctly identified critical sequence differences between active and inactive sequences with nearly identical transcription factor binding sites, and revealed order and spacing preferences for combinations of motifs. Our results establish active learning as an effective method to train accurate deep learning models of cis-regulatory function after exhausting naturally occurring training examples in the genome.
Collapse
Affiliation(s)
- Ryan Z. Friedman
- The Edison Family Center for Genome Sciences & Systems Biology, Washington University School of Medicine, Saint Louis, MO, 63110
- Department of Genetics, Washington University School of Medicine, Saint Louis, MO, 63110
| | - Avinash Ramu
- The Edison Family Center for Genome Sciences & Systems Biology, Washington University School of Medicine, Saint Louis, MO, 63110
- Department of Genetics, Washington University School of Medicine, Saint Louis, MO, 63110
| | - Sara Lichtarge
- The Edison Family Center for Genome Sciences & Systems Biology, Washington University School of Medicine, Saint Louis, MO, 63110
- Department of Genetics, Washington University School of Medicine, Saint Louis, MO, 63110
| | - Connie A. Myers
- Department of Pathology and Immunology, Washington University School of Medicine, Saint Louis, MO, 63110
| | - David M. Granas
- The Edison Family Center for Genome Sciences & Systems Biology, Washington University School of Medicine, Saint Louis, MO, 63110
- Department of Genetics, Washington University School of Medicine, Saint Louis, MO, 63110
| | - Maria Gause
- Department of Pathology and Immunology, Washington University School of Medicine, Saint Louis, MO, 63110
| | - Joseph C. Corbo
- Department of Pathology and Immunology, Washington University School of Medicine, Saint Louis, MO, 63110
| | - Barak A. Cohen
- The Edison Family Center for Genome Sciences & Systems Biology, Washington University School of Medicine, Saint Louis, MO, 63110
- Department of Genetics, Washington University School of Medicine, Saint Louis, MO, 63110
| | - Michael A. White
- The Edison Family Center for Genome Sciences & Systems Biology, Washington University School of Medicine, Saint Louis, MO, 63110
- Department of Genetics, Washington University School of Medicine, Saint Louis, MO, 63110
| |
Collapse
|
15
|
Novakovsky G, Fornes O, Saraswat M, Mostafavi S, Wasserman WW. ExplaiNN: interpretable and transparent neural networks for genomics. Genome Biol 2023; 24:154. [PMID: 37370113 DOI: 10.1186/s13059-023-02985-y] [Citation(s) in RCA: 6] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/20/2022] [Accepted: 06/12/2023] [Indexed: 06/29/2023] Open
Abstract
Deep learning models such as convolutional neural networks (CNNs) excel in genomic tasks but lack interpretability. We introduce ExplaiNN, which combines the expressiveness of CNNs with the interpretability of linear models. ExplaiNN can predict TF binding, chromatin accessibility, and de novo motifs, achieving performance comparable to state-of-the-art methods. Its predictions are transparent, providing global (cell state level) as well as local (individual sequence level) biological insights into the data. ExplaiNN can serve as a plug-and-play platform for pretrained models and annotated position weight matrices. ExplaiNN aims to accelerate the adoption of deep learning in genomic sequence analysis by domain experts.
Collapse
Affiliation(s)
- Gherman Novakovsky
- Department of Medical Genetics, Centre for Molecular Medicine and Therapeutics, BC Children's Hospital Research Institute, University of British Columbia, Vancouver, BC, Canada
| | - Oriol Fornes
- Department of Medical Genetics, Centre for Molecular Medicine and Therapeutics, BC Children's Hospital Research Institute, University of British Columbia, Vancouver, BC, Canada
| | - Manu Saraswat
- Department of Medical Genetics, Centre for Molecular Medicine and Therapeutics, BC Children's Hospital Research Institute, University of British Columbia, Vancouver, BC, Canada
- Division of Computational Genomics and Systems Genetics, German Cancer Research Center (DKFZ), Heidelberg, Germany
- European Molecular Biology Laboratory (EMBL), Genome Biology Unit, Heidelberg, Germany
| | - Sara Mostafavi
- Paul G. Allen School of Computer Science and Engineering, University of Washington (UW), Seattle, USA
| | - Wyeth W Wasserman
- Department of Medical Genetics, Centre for Molecular Medicine and Therapeutics, BC Children's Hospital Research Institute, University of British Columbia, Vancouver, BC, Canada.
| |
Collapse
|
16
|
Janizek JD, Dincer AB, Celik S, Chen H, Chen W, Naxerova K, Lee SI. Uncovering expression signatures of synergistic drug responses via ensembles of explainable machine-learning models. Nat Biomed Eng 2023; 7:811-829. [PMID: 37127711 PMCID: PMC11149694 DOI: 10.1038/s41551-023-01034-0] [Citation(s) in RCA: 4] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/12/2021] [Accepted: 04/01/2023] [Indexed: 05/03/2023]
Abstract
Machine learning may aid the choice of optimal combinations of anticancer drugs by explaining the molecular basis of their synergy. By combining accurate models with interpretable insights, explainable machine learning promises to accelerate data-driven cancer pharmacology. However, owing to the highly correlated and high-dimensional nature of transcriptomic data, naively applying current explainable machine-learning strategies to large transcriptomic datasets leads to suboptimal outcomes. Here by using feature attribution methods, we show that the quality of the explanations can be increased by leveraging ensembles of explainable machine-learning models. We applied the approach to a dataset of 133 combinations of 46 anticancer drugs tested in ex vivo tumour samples from 285 patients with acute myeloid leukaemia and uncovered a haematopoietic-differentiation signature underlying drug combinations with therapeutic synergy. Ensembles of machine-learning models trained to predict drug combination synergies on the basis of gene-expression data may improve the feature attribution quality of complex machine-learning models.
Collapse
Affiliation(s)
- Joseph D Janizek
- Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, WA, USA
- Medical Scientist Training Program, University of Washington, Seattle, WA, USA
| | - Ayse B Dincer
- Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, WA, USA
| | - Safiye Celik
- Recursion Pharmaceuticals, Salt Lake City, UT, USA
| | - Hugh Chen
- Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, WA, USA
| | - William Chen
- Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, WA, USA
| | - Kamila Naxerova
- Center for Systems Biology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA.
- Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA.
| | - Su-In Lee
- Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, WA, USA.
| |
Collapse
|
17
|
Majdandzic A, Rajesh C, Koo PK. Correcting gradient-based interpretations of deep neural networks for genomics. Genome Biol 2023; 24:109. [PMID: 37161475 PMCID: PMC10169356 DOI: 10.1186/s13059-023-02956-3] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/04/2022] [Accepted: 04/28/2023] [Indexed: 05/11/2023] Open
Abstract
Post hoc attribution methods can provide insights into the learned patterns from deep neural networks (DNNs) trained on high-throughput functional genomics data. However, in practice, their resultant attribution maps can be challenging to interpret due to spurious importance scores for seemingly arbitrary nucleotides. Here, we identify a previously overlooked attribution noise source that arises from how DNNs handle one-hot encoded DNA. We demonstrate this noise is pervasive across various genomic DNNs and introduce a statistical correction that effectively reduces it, leading to more reliable attribution maps. Our approach represents a promising step towards gaining meaningful insights from DNNs in regulatory genomics.
Collapse
Affiliation(s)
- Antonio Majdandzic
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, 1 Bungtown Road, Cold Spring Harbor, NY, USA
| | - Chandana Rajesh
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, 1 Bungtown Road, Cold Spring Harbor, NY, USA
| | - Peter K Koo
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, 1 Bungtown Road, Cold Spring Harbor, NY, USA.
| |
Collapse
|
18
|
Lee NK, Tang Z, Toneyan S, Koo PK. EvoAug: improving generalization and interpretability of genomic deep neural networks with evolution-inspired data augmentations. Genome Biol 2023; 24:105. [PMID: 37143118 PMCID: PMC10161416 DOI: 10.1186/s13059-023-02941-w] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/12/2022] [Accepted: 04/17/2023] [Indexed: 05/06/2023] Open
Abstract
Deep neural networks (DNNs) hold promise for functional genomics prediction, but their generalization capability may be limited by the amount of available data. To address this, we propose EvoAug, a suite of evolution-inspired augmentations that enhance the training of genomic DNNs by increasing genetic variation. Random transformation of DNA sequences can potentially alter their function in unknown ways, so we employ a fine-tuning procedure using the original non-transformed data to preserve functional integrity. Our results demonstrate that EvoAug substantially improves the generalization and interpretability of established DNNs across prominent regulatory genomics prediction tasks, offering a robust solution for genomic DNNs.
Collapse
Affiliation(s)
- Nicholas Keone Lee
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, 1 Bungtown Road, Cold Spring Harbor, NY, USA
| | - Ziqi Tang
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, 1 Bungtown Road, Cold Spring Harbor, NY, USA
| | - Shushan Toneyan
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, 1 Bungtown Road, Cold Spring Harbor, NY, USA
| | - Peter K Koo
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, 1 Bungtown Road, Cold Spring Harbor, NY, USA.
| |
Collapse
|
19
|
Boersma M, Wolsink J, Sourabh S, Hoogduin LA, Kandhai D. Measure cross-sectoral structural similarities from financial networks. Sci Rep 2023; 13:7124. [PMID: 37130862 PMCID: PMC10153030 DOI: 10.1038/s41598-023-34034-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/25/2022] [Accepted: 04/22/2023] [Indexed: 05/04/2023] Open
Abstract
Auditing is a multi-billion dollar market, with auditors assessing the trustworthiness of financial data, contributing to financial stability in a more interconnected and faster-changing world. We measure cross-sectoral structural similarities between firms using microscopic real-world transaction data. We derive network representations of companies from their transaction datasets, and we compute an embedding vector for each network. Our approach is based on the analysis of 300+ real transaction datasets that provide auditors with relevant insights. We detect significant changes in bookkeeping structure and the similarity between clients. For various tasks, we obtain good classification accuracy. Moreover, closely related companies are near in the embedding space while different industries are further apart suggesting that the measure captures relevant aspects. Besides the direct applications in computational audit, we expect this approach to be of use at multiple scales, from firms to countries, potentially elucidating structural risks at a broader scale.
Collapse
Affiliation(s)
- M Boersma
- Computational Science Lab, University of Amsterdam, Amsterdam, The Netherlands.
- KPMG, Amstelveen, The Netherlands.
| | - J Wolsink
- Computational Science Lab, University of Amsterdam, Amsterdam, The Netherlands
- KPMG, Amstelveen, The Netherlands
| | - S Sourabh
- Computational Science Lab, University of Amsterdam, Amsterdam, The Netherlands
| | | | - D Kandhai
- Computational Science Lab, University of Amsterdam, Amsterdam, The Netherlands
| |
Collapse
|
20
|
Novakovsky G, Dexter N, Libbrecht MW, Wasserman WW, Mostafavi S. Obtaining genetics insights from deep learning via explainable artificial intelligence. Nat Rev Genet 2023; 24:125-137. [PMID: 36192604 DOI: 10.1038/s41576-022-00532-2] [Citation(s) in RCA: 63] [Impact Index Per Article: 63.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 08/31/2022] [Indexed: 01/24/2023]
Abstract
Artificial intelligence (AI) models based on deep learning now represent the state of the art for making functional predictions in genomics research. However, the underlying basis on which predictive models make such predictions is often unknown. For genomics researchers, this missing explanatory information would frequently be of greater value than the predictions themselves, as it can enable new insights into genetic processes. We review progress in the emerging area of explainable AI (xAI), a field with the potential to empower life science researchers to gain mechanistic insights into complex deep learning models. We discuss and categorize approaches for model interpretation, including an intuitive understanding of how each approach works and their underlying assumptions and limitations in the context of typical high-throughput biological datasets.
Collapse
Affiliation(s)
- Gherman Novakovsky
- Centre for Molecular Medicine and Therapeutics, Department of Medical Genetics, BC Children's Hospital Research Institute, University of British Columbia, Vancouver, British Columbia, Canada.,Bioinformatics Graduate Program, University of British Columbia, Vancouver, British Columbia, Canada
| | - Nick Dexter
- Department of Mathematics, Simon Fraser University, Burnaby, British Columbia, Canada.,School of Computing Science, Simon Fraser University, Burnaby, British Columbia, Canada
| | - Maxwell W Libbrecht
- School of Computing Science, Simon Fraser University, Burnaby, British Columbia, Canada.
| | - Wyeth W Wasserman
- Centre for Molecular Medicine and Therapeutics, Department of Medical Genetics, BC Children's Hospital Research Institute, University of British Columbia, Vancouver, British Columbia, Canada.
| | - Sara Mostafavi
- Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, WA, USA. .,Canadian Institute for Advanced Research, Toronto, Ontario, Canada.
| |
Collapse
|
21
|
Deep learning in regulatory genomics: from identification to design. Curr Opin Biotechnol 2023; 79:102887. [PMID: 36640453 DOI: 10.1016/j.copbio.2022.102887] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/28/2022] [Revised: 11/12/2022] [Accepted: 12/14/2022] [Indexed: 01/14/2023]
Abstract
Genomics and deep learning are a natural match since both are data-driven fields. Regulatory genomics refers to functional noncoding DNA regulating gene expression. In recent years, deep learning applications on regulatory genomics have achieved remarkable advances so-much-so that it has revolutionized the rules of the game of the computational methods in this field. Here, we review two emerging trends: (i) the modeling of very long input sequence (up to 200 kb), which requires self-matched modularization of model architecture; (ii) on the balance of model predictability and model interpretability because the latter is more able to meet biological demands. Finally, we discuss how to employ these two routes to design synthetic regulatory DNA, as a promising strategy for optimizing crop agronomic properties.
Collapse
|
22
|
Koo PK, Ploenzke M, Anand P, Paul S, Majdandzic A. ResidualBind: Uncovering Sequence-Structure Preferences of RNA-Binding Proteins with Deep Neural Networks. Methods Mol Biol 2023; 2586:197-215. [PMID: 36705906 DOI: 10.1007/978-1-0716-2768-6_12] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/28/2023]
Abstract
Deep neural networks have demonstrated improved performance at predicting sequence specificities of DNA- and RNA-binding proteins. However, it remains unclear why they perform better than previous methods that rely on k-mers and position weight matrices. Here, we highlight a recent deep learning-based software package, called ResidualBind, that analyzes RNA-protein interactions using only RNA sequence as an input feature and performs global importance analysis for model interpretability. We discuss practical considerations for model interpretability to uncover learned sequence motifs and their secondary structure preferences.
Collapse
Affiliation(s)
- Peter K Koo
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, USA.
| | - Matt Ploenzke
- Department of Biostatistics, Harvard University, Cambridge, MA, USA
| | | | - Steffan Paul
- Bioinformatics Program, Harvard Medical School, Boston, MA, USA
| | - Antonio Majdandzic
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, USA
| |
Collapse
|
23
|
A multimorphic mutation in IRF4 causes human autosomal dominant combined immunodeficiency. Sci Immunol 2023; 8:eade7953. [PMID: 36662884 PMCID: PMC10825898 DOI: 10.1126/sciimmunol.ade7953] [Citation(s) in RCA: 9] [Impact Index Per Article: 9.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/08/2022] [Accepted: 12/22/2022] [Indexed: 01/22/2023]
Abstract
Interferon regulatory factor 4 (IRF4) is a transcription factor (TF) and key regulator of immune cell development and function. We report a recurrent heterozygous mutation in IRF4, p.T95R, causing an autosomal dominant combined immunodeficiency (CID) in seven patients from six unrelated families. The patients exhibited profound susceptibility to opportunistic infections, notably Pneumocystis jirovecii, and presented with agammaglobulinemia. Patients' B cells showed impaired maturation, decreased immunoglobulin isotype switching, and defective plasma cell differentiation, whereas their T cells contained reduced TH17 and TFH populations and exhibited decreased cytokine production. A knock-in mouse model of heterozygous T95R showed a severe defect in antibody production both at the steady state and after immunization with different types of antigens, consistent with the CID observed in these patients. The IRF4T95R variant maps to the TF's DNA binding domain, alters its canonical DNA binding specificities, and results in a simultaneous multimorphic combination of loss, gain, and new functions for IRF4. IRF4T95R behaved as a gain-of-function hypermorph by binding to DNA with higher affinity than IRF4WT. Despite this increased affinity for DNA, the transcriptional activity on IRF4 canonical genes was reduced, showcasing a hypomorphic activity of IRF4T95R. Simultaneously, IRF4T95R functions as a neomorph by binding to noncanonical DNA sites to alter the gene expression profile, including the transcription of genes exclusively induced by IRF4T95R but not by IRF4WT. This previously undescribed multimorphic IRF4 pathophysiology disrupts normal lymphocyte biology, causing human disease.
Collapse
|
24
|
Kawaguchi RK, Tang Z, Fischer S, Rajesh C, Tripathy R, Koo PK, Gillis J. Learning single-cell chromatin accessibility profiles using meta-analytic marker genes. Brief Bioinform 2023; 24:bbac541. [PMID: 36549922 PMCID: PMC9851328 DOI: 10.1093/bib/bbac541] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/27/2022] [Revised: 09/29/2022] [Accepted: 11/08/2022] [Indexed: 12/24/2022] Open
Abstract
MOTIVATION Single-cell assay for transposase accessible chromatin using sequencing (scATAC-seq) is a valuable resource to learn cis-regulatory elements such as cell-type specific enhancers and transcription factor binding sites. However, cell-type identification of scATAC-seq data is known to be challenging due to the heterogeneity derived from different protocols and the high dropout rate. RESULTS In this study, we perform a systematic comparison of seven scATAC-seq datasets of mouse brain to benchmark the efficacy of neuronal cell-type annotation from gene sets. We find that redundant marker genes give a dramatic improvement for a sparse scATAC-seq annotation across the data collected from different studies. Interestingly, simple aggregation of such marker genes achieves performance comparable or higher than that of machine-learning classifiers, suggesting its potential for downstream applications. Based on our results, we reannotated all scATAC-seq data for detailed cell types using robust marker genes. Their meta scATAC-seq profiles are publicly available at https://gillisweb.cshl.edu/Meta_scATAC. Furthermore, we trained a deep neural network to predict chromatin accessibility from only DNA sequence and identified key motifs enriched for each neuronal subtype. Those predicted profiles are visualized together in our database as a valuable resource to explore cell-type specific epigenetic regulation in a sequence-dependent and -independent manner.
Collapse
Affiliation(s)
| | - Ziqi Tang
- Cold Spring Harbor Laboratory, Cold Spring Harbor 11724, USA
| | - Stephan Fischer
- Cold Spring Harbor Laboratory, Cold Spring Harbor 11724, USA
| | - Chandana Rajesh
- Cold Spring Harbor Laboratory, Cold Spring Harbor 11724, USA
| | - Rohit Tripathy
- Cold Spring Harbor Laboratory, Cold Spring Harbor 11724, USA
| | - Peter K Koo
- Cold Spring Harbor Laboratory, Cold Spring Harbor 11724, USA
| | - Jesse Gillis
- Cold Spring Harbor Laboratory, Cold Spring Harbor 11724, USA
- Department of Physiology and Donnelly Centre for Cellular & Biomolecular Research Department, University of Toronto, Ontario M5S 3E1, Canada
| |
Collapse
|
25
|
Artz O, Ackermann A, Taylor L, Koo PK, Pedmale UV. Light and temperature regulate m 6A-RNA modification to regulate growth in plants. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2023:2023.01.17.524395. [PMID: 36711495 PMCID: PMC9882139 DOI: 10.1101/2023.01.17.524395] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Indexed: 01/19/2023]
Abstract
N6-methyladenosine is a highly dynamic, abundant mRNA modification which is an excellent potential mechanism for fine tuning gene expression. Plants adapt to their surrounding light and temperature environment using complex gene regulatory networks. The role of m6A in controlling gene expression in response to variable environmental conditions has so far been unexplored. Here, we map the transcriptome-wide m6A landscape under various light and temperature environments. Identified m6A-modifications show a highly specific spatial distribution along transcripts with enrichment occurring in 5'UTR regions and around transcriptional end sites. We show that the position of m6A modifications on transcripts might influence cellular transcript localization and the presence of m6A-modifications is associated with alternative polyadenylation, a process which results in multiple RNA isoforms with varying 3'UTR lengths. RNA with m6A-modifications exhibit a higher preference for shorter 3'UTRs. These shorter 3'UTR regions might directly influence transcript abundance and localization by including or excluding cis-regulatory elements. We propose that environmental stimuli might change the m6A landscape of plants as one possible way of fine tuning gene regulation through alternative polyadenylation and transcript localization.
Collapse
Affiliation(s)
- Oliver Artz
- Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 11724. USA
| | - Amanda Ackermann
- Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 11724. USA
| | - Laura Taylor
- Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 11724. USA
| | - Peter K. Koo
- Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 11724. USA
| | - Ullas V. Pedmale
- Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 11724. USA
| |
Collapse
|
26
|
Toneyan S, Tang Z, Koo PK. Evaluating deep learning for predicting epigenomic profiles. NAT MACH INTELL 2022. [DOI: 10.1038/s42256-022-00570-9] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/12/2022]
|
27
|
Park M, Singh S, Khan SR, Abrar MA, Grisanti F, Rahman MS, Samee MAH. Multinomial Convolutions for Joint Modeling of Regulatory Motifs and Sequence Activity Readouts. Genes (Basel) 2022; 13:genes13091614. [PMID: 36140783 PMCID: PMC9498894 DOI: 10.3390/genes13091614] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/30/2022] [Revised: 08/31/2022] [Accepted: 09/02/2022] [Indexed: 11/29/2022] Open
Abstract
A common goal in the convolutional neural network (CNN) modeling of genomic data is to discover specific sequence motifs. Post hoc analysis methods aid in this task but are dependent on parameters whose optimal values are unclear and applying the discovered motifs to new genomic data is not straightforward. As an alternative, we propose to learn convolutions as multinomial distributions, thus streamlining interpretable motif discovery with CNN model fitting. We developed MuSeAM (Multinomial CNNs for Sequence Activity Modeling) by implementing multinomial convolutions in a CNN model. Through benchmarking, we demonstrate the efficacy of MuSeAM in accurately modeling genomic data while fitting multinomial convolutions that recapitulate known transcription factor motifs.
Collapse
Affiliation(s)
- Minjun Park
- Department of Integrative Physiology, Baylor College of Medicine, Houston, TX 77030, USA
| | - Salvi Singh
- Department of Integrative Physiology, Baylor College of Medicine, Houston, TX 77030, USA
| | - Samin Rahman Khan
- Computer Science and Engineering, Bangladesh University of Engineering and Technology, Dhaka 1000, Bangladesh
| | | | - Francisco Grisanti
- Department of Integrative Physiology, Baylor College of Medicine, Houston, TX 77030, USA
| | - M. Sohel Rahman
- Computer Science and Engineering, Bangladesh University of Engineering and Technology, Dhaka 1000, Bangladesh
| | - Md. Abul Hassan Samee
- Department of Integrative Physiology, Baylor College of Medicine, Houston, TX 77030, USA
- Correspondence:
| |
Collapse
|
28
|
Kassani PH, Lu F, Guen YL, Belloy ME, He Z. Deep neural networks with controlled variable selection for the identification of putative causal genetic variants. NAT MACH INTELL 2022; 4:761-771. [PMID: 37859729 PMCID: PMC10586424 DOI: 10.1038/s42256-022-00525-0] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/24/2021] [Accepted: 07/26/2022] [Indexed: 11/09/2022]
Abstract
Deep neural networks (DNNs) have been successfully utilized in many scientific problems for their high prediction accuracy, but their application to genetic studies remains challenging due to their poor interpretability. Here we consider the problem of scalable, robust variable selection in DNNs for the identification of putative causal genetic variants in genome sequencing studies. We identified a pronounced randomness in feature selection in DNNs due to its stochastic nature, which may hinder interpretability and give rise to misleading results. We propose an interpretable neural network model, stabilized using ensembling, with controlled variable selection for genetic studies. The merit of the proposed method includes: flexible modelling of the nonlinear effect of genetic variants to improve statistical power; multiple knockoffs in the input layer to rigorously control the false discovery rate; hierarchical layers to substantially reduce the number of weight parameters and activations, and improve computational efficiency; and stabilized feature selection to reduce the randomness in identified signals. We evaluate the proposed method in extensive simulation studies and apply it to the analysis of Alzheimer's disease genetics. We show that the proposed method, when compared with conventional linear and nonlinear methods, can lead to substantially more discoveries.
Collapse
Affiliation(s)
- Peyman H. Kassani
- Department of Neurology and Neurological Sciences, Stanford University, Stanford, CA, USA
| | - Fred Lu
- Department of Statistics, Stanford University, Stanford, CA, USA
| | - Yann Le Guen
- Department of Neurology and Neurological Sciences, Stanford University, Stanford, CA, USA
| | - Michael E. Belloy
- Department of Neurology and Neurological Sciences, Stanford University, Stanford, CA, USA
| | - Zihuai He
- Department of Neurology and Neurological Sciences, Stanford University, Stanford, CA, USA
- Quantitative Sciences Unit, Department of Medicine (Biomedical Informatics Research), Stanford University, Stanford, CA, USA
| |
Collapse
|
29
|
Yoon CH, Torrance R, Scheinerman N. Machine learning in medicine: should the pursuit of enhanced interpretability be abandoned? JOURNAL OF MEDICAL ETHICS 2022; 48:581-585. [PMID: 34006600 PMCID: PMC9411871 DOI: 10.1136/medethics-2020-107102] [Citation(s) in RCA: 29] [Impact Index Per Article: 14.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/25/2020] [Revised: 03/21/2021] [Accepted: 04/08/2021] [Indexed: 06/01/2023]
Abstract
We argue why interpretability should have primacy alongside empiricism for several reasons: first, if machine learning (ML) models are beginning to render some of the high-risk healthcare decisions instead of clinicians, these models pose a novel medicolegal and ethical frontier that is incompletely addressed by current methods of appraising medical interventions like pharmacological therapies; second, a number of judicial precedents underpinning medical liability and negligence are compromised when 'autonomous' ML recommendations are considered to be en par with human instruction in specific contexts; third, explainable algorithms may be more amenable to the ascertainment and minimisation of biases, with repercussions for racial equity as well as scientific reproducibility and generalisability. We conclude with some reasons for the ineludible importance of interpretability, such as the establishment of trust, in overcoming perhaps the most difficult challenge ML will face in a high-stakes environment like healthcare: professional and public acceptance.
Collapse
Affiliation(s)
- Chang Ho Yoon
- Big Data Institute, Oxford University, Oxford, UK
- Medical Sciences Doctoral Training Centre, Oxford University, Oxford, UK
- Nuffield Department of Population Health, University of Oxford Richard Doll Building, Oxford, UK
| | - Robert Torrance
- Faculty of Public Health and Policy, London School of Hygiene and Tropical Medicine, London, UK
| | - Naomi Scheinerman
- Department of Medical Ethics and Health Policy, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA
| |
Collapse
|
30
|
Nair S, Shrikumar A, Schreiber J, Kundaje A. fastISM: performant in silico saturation mutagenesis for convolutional neural networks. Bioinformatics 2022; 38:2397-2403. [PMID: 35238376 PMCID: PMC9048647 DOI: 10.1093/bioinformatics/btac135] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/26/2021] [Revised: 02/09/2022] [Accepted: 03/01/2022] [Indexed: 11/14/2022] Open
Abstract
MOTIVATION Deep-learning models, such as convolutional neural networks, are able to accurately map biological sequences to associated functional readouts and properties by learning predictive de novo representations. In silico saturation mutagenesis (ISM) is a popular feature attribution technique for inferring contributions of all characters in an input sequence to the model's predicted output. The main drawback of ISM is its runtime, as it involves multiple forward propagations of all possible mutations of each character in the input sequence through the trained model to predict the effects on the output. RESULTS We present fastISM, an algorithm that speeds up ISM by a factor of over 10× for commonly used convolutional neural network architectures. fastISM is based on the observations that the majority of computation in ISM is spent in convolutional layers, and a single mutation only disrupts a limited region of intermediate layers, rendering most computation redundant. fastISM reduces the gap between backpropagation-based feature attribution methods and ISM. It far surpasses the runtime of backpropagation-based methods on multi-output architectures, making it feasible to run ISM on a large number of sequences. AVAILABILITY AND IMPLEMENTATION An easy-to-use Keras/TensorFlow 2 implementation of fastISM is available at https://github.com/kundajelab/fastISM. fastISM can be installed using pip install fastism. A hands-on tutorial can be found at https://colab.research.google.com/github/kundajelab/fastISM/blob/master/notebooks/colab/DeepSEA.ipynb. SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.
Collapse
Affiliation(s)
- Surag Nair
- Department of Computer Science, Stanford University, Stanford, CA 94305, USA
| | | | | | | |
Collapse
|
31
|
Yaish O, Orenstein Y. Computational modeling of mRNA degradation dynamics using deep neural networks. Bioinformatics 2022; 38:1087-1101. [PMID: 34849591 DOI: 10.1093/bioinformatics/btab800] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/29/2021] [Revised: 11/12/2021] [Accepted: 11/22/2021] [Indexed: 02/04/2023] Open
Abstract
MOTIVATION messenger RNA (mRNA) degradation plays critical roles in post-transcriptional gene regulation. A major component of mRNA degradation is determined by 3'-UTR elements. Hence, researchers are interested in studying mRNA dynamics as a function of 3'-UTR elements. A recent study measured the mRNA degradation dynamics of tens of thousands of 3'-UTR sequences using a massively parallel reporter assay. However, the computational approach used to model mRNA degradation was based on a simplifying assumption of a linear degradation rate. Consequently, the underlying mechanism of 3'-UTR elements is still not fully understood. RESULTS Here, we developed deep neural networks to predict mRNA degradation dynamics and interpreted the networks to identify regulatory elements in the 3'-UTR and their positional effect. Given an input of a 110 nt-long 3'-UTR sequence and an initial mRNA level, the model predicts mRNA levels of eight consecutive time points. Our deep neural networks significantly improved prediction performance of mRNA degradation dynamics compared with extant methods for the task. Moreover, we demonstrated that models predicting the dynamics of two identical 3'-UTR sequences, differing by their poly(A) tail, performed better than single-task models. On the interpretability front, by using Integrated Gradients, our convolutional neural networks (CNNs) models identified known and novel cis-regulatory sequence elements of mRNA degradation. By applying a novel systematic evaluation of model interpretability, we demonstrated that the recurrent neural network models are inferior to the CNN models in terms of interpretability and that random initialization ensemble improves both prediction and interoperability performance. Moreover, using a mutagenesis analysis, we newly discovered the positional effect of various 3'-UTR elements. AVAILABILITY AND IMPLEMENTATION All the code developed through this study is available at github.com/OrensteinLab/DeepUTR/. SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.
Collapse
Affiliation(s)
- Ofir Yaish
- School of Electrical and Computer Engineering, Ben-Gurion University of the Negev, Beer-Sheva 8410501, Israel
| | - Yaron Orenstein
- School of Electrical and Computer Engineering, Ben-Gurion University of the Negev, Beer-Sheva 8410501, Israel
| |
Collapse
|
32
|
Huminiecki Ł. Virtual Gene Concept and a Corresponding Pragmatic Research Program in Genetical Data Science. ENTROPY (BASEL, SWITZERLAND) 2021; 24:17. [PMID: 35052043 PMCID: PMC8774939 DOI: 10.3390/e24010017] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 09/30/2021] [Revised: 12/02/2021] [Accepted: 12/14/2021] [Indexed: 06/14/2023]
Abstract
Mendel proposed an experimentally verifiable paradigm of particle-based heredity that has been influential for over 150 years. The historical arguments have been reflected in the near past as Mendel's concept has been diversified by new types of omics data. As an effect of the accumulation of omics data, a virtual gene concept forms, giving rise to genetical data science. The concept integrates genetical, functional, and molecular features of the Mendelian paradigm. I argue that the virtual gene concept should be deployed pragmatically. Indeed, the concept has already inspired a practical research program related to systems genetics. The program includes questions about functionality of structural and categorical gene variants, about regulation of gene expression, and about roles of epigenetic modifications. The methodology of the program includes bioinformatics, machine learning, and deep learning. Education, funding, careers, standards, benchmarks, and tools to monitor research progress should be provided to support the research program.
Collapse
Affiliation(s)
- Łukasz Huminiecki
- Evolutionary, Computational, and Statistical Genetics, Department of Molecula Biology, Institute of Genetics and Animal Biotechnology, Polish Academy of Sciences, Postępu 36A, Jastrzębiec, 05-552 Warsaw, Poland
| |
Collapse
|
33
|
Zrimec J, Buric F, Kokina M, Garcia V, Zelezniak A. Learning the Regulatory Code of Gene Expression. Front Mol Biosci 2021; 8:673363. [PMID: 34179082 PMCID: PMC8223075 DOI: 10.3389/fmolb.2021.673363] [Citation(s) in RCA: 6] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/27/2021] [Accepted: 05/24/2021] [Indexed: 11/13/2022] Open
Abstract
Data-driven machine learning is the method of choice for predicting molecular phenotypes from nucleotide sequence, modeling gene expression events including protein-DNA binding, chromatin states as well as mRNA and protein levels. Deep neural networks automatically learn informative sequence representations and interpreting them enables us to improve our understanding of the regulatory code governing gene expression. Here, we review the latest developments that apply shallow or deep learning to quantify molecular phenotypes and decode the cis-regulatory grammar from prokaryotic and eukaryotic sequencing data. Our approach is to build from the ground up, first focusing on the initiating protein-DNA interactions, then specific coding and non-coding regions, and finally on advances that combine multiple parts of the gene and mRNA regulatory structures, achieving unprecedented performance. We thus provide a quantitative view of gene expression regulation from nucleotide sequence, concluding with an information-centric overview of the central dogma of molecular biology.
Collapse
Affiliation(s)
- Jan Zrimec
- Department of Biology and Biological Engineering, Chalmers University of Technology, Gothenburg, Sweden
| | - Filip Buric
- Department of Biology and Biological Engineering, Chalmers University of Technology, Gothenburg, Sweden
| | - Mariia Kokina
- Department of Biology and Biological Engineering, Chalmers University of Technology, Gothenburg, Sweden
- Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark, Kongens Lyngby, Denmark
| | - Victor Garcia
- School of Life Sciences and Facility Management, Zurich University of Applied Sciences, Wädenswil, Switzerland
| | - Aleksej Zelezniak
- Department of Biology and Biological Engineering, Chalmers University of Technology, Gothenburg, Sweden
- Science for Life Laboratory, Stockholm, Sweden
| |
Collapse
|
34
|
Schreiber J, Singh R. Machine learning for profile prediction in genomics. Curr Opin Chem Biol 2021; 65:35-41. [PMID: 34107341 DOI: 10.1016/j.cbpa.2021.04.008] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/31/2021] [Revised: 04/21/2021] [Accepted: 04/24/2021] [Indexed: 02/08/2023]
Abstract
A recent deluge of publicly available multi-omics data has fueled the development of machine learning methods aimed at investigating important questions in genomics. Although the motivations for these methods vary, a task that is commonly adopted is that of profile prediction, where predictions are made for one or more forms of biochemical activity along the genome, for example, histone modification, chromatin accessibility, or protein binding. In this review, we give an overview of the research works performing profile prediction, define two broad categories of profile prediction tasks, and discuss the types of scientific questions that can be answered in each.
Collapse
Affiliation(s)
| | - Ritambhara Singh
- Department of Computer Science, Center for Computational Molecular Biology, Brown University, United States.
| |
Collapse
|
35
|
Koo PK, Majdandzic A, Ploenzke M, Anand P, Paul SB. Global importance analysis: An interpretability method to quantify importance of genomic features in deep neural networks. PLoS Comput Biol 2021; 17:e1008925. [PMID: 33983921 PMCID: PMC8118286 DOI: 10.1371/journal.pcbi.1008925] [Citation(s) in RCA: 33] [Impact Index Per Article: 11.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/18/2020] [Accepted: 03/30/2021] [Indexed: 12/15/2022] Open
Abstract
Deep neural networks have demonstrated improved performance at predicting the sequence specificities of DNA- and RNA-binding proteins compared to previous methods that rely on k-mers and position weight matrices. To gain insights into why a DNN makes a given prediction, model interpretability methods, such as attribution methods, can be employed to identify motif-like representations along a given sequence. Because explanations are given on an individual sequence basis and can vary substantially across sequences, deducing generalizable trends across the dataset and quantifying their effect size remains a challenge. Here we introduce global importance analysis (GIA), a model interpretability method that quantifies the population-level effect size that putative patterns have on model predictions. GIA provides an avenue to quantitatively test hypotheses of putative patterns and their interactions with other patterns, as well as map out specific functions the network has learned. As a case study, we demonstrate the utility of GIA on the computational task of predicting RNA-protein interactions from sequence. We first introduce a convolutional network, we call ResidualBind, and benchmark its performance against previous methods on RNAcompete data. Using GIA, we then demonstrate that in addition to sequence motifs, ResidualBind learns a model that considers the number of motifs, their spacing, and sequence context, such as RNA secondary structure and GC-bias.
Collapse
Affiliation(s)
- Peter K. Koo
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, United States of America
| | - Antonio Majdandzic
- Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, United States of America
| | - Matthew Ploenzke
- Department of Biostatistics, Harvard University, Cambridge, Massachusetts, United States of America
| | - Praveen Anand
- Dana-Farber Cancer Institute, Boston, Massachusetts, United States of America
| | - Steffan B. Paul
- Bioinformatics and Integrative Genomics Program, Harvard Medical School, Boston, Massachusetts, United States of America
| |
Collapse
|