Visible Machine Learning for Biomedicine
Open accessAbstract
A major ambition of artificial intelligence lies in translating patient data to successful therapies. Machine learning models face particular challenges in biomedicine, however, including handling of extreme data heterogeneity and lack of mechanistic insight into predictions. Here, we argue for “visible” approaches that guide model structure with experimental biology.
AI snapshot
Written from the paper’s full text.
TL;DR
The paper argues that current black-box machine learning is intrinsically limited in biomedicine by extreme data heterogeneity and lack of mechanistic interpretation, and proposes “visible” machine learning guided by biological structure. It outlines examples and near-term milestones for interpretable, mechanism-aware biomedical models.
Key findings
- Current machine learning faces two acute biomedical challenges: input heterogeneity, where the same outcome can arise from vastly different inputs (e.g., cancer from many rare combinations of genetic alterations), and black-box predictions that do not link outputs to underlying mechanisms.
- Brute-force data scaling is limited because patient numbers are finite; even common conditions like cancer and heart disease may saturate at “a few million patient examples,” which the authors argue is modest compared with the data needs of complex biological systems.
- Visible learning uses prior biological structure—such as molecular networks and hierarchical cell/tissue organization—to guide model architecture, potentially addressing heterogeneity through biologically meaningful compression and enabling mechanistic interpretation of internal model states.
- Examples cited include Lin et al. (2017), who structured a deep neural network based on hierarchical organization of transcriptional regulatory factors; DCell (Ma et al., 2018), which maps neurons to biological modules to translate mutated genes to cell proliferation; and Alvarez et al. (2016), who used a transcriptional regulatory network to infer regulatory protein activities from mRNA expression.
Methods
- This is a commentary/perspective, not an empirical study; no new dataset, sample, or statistical analysis is reported.
- The argument synthesizes prior literature and examples from machine learning and computational biology to propose visible machine learning and near-term milestones.
Limitations
- The paper states limitations of current black-box machine learning: data heterogeneity, lack of mechanistic insight, finite patient datasets, and a trade-off between increased model resolution and model scope.
- As a commentary, it presents no primary validation, quantitative benchmarks, or systematic review; its proposals and milestones are not empirically tested here (evident from the text).
- The paper also notes that literature curation is difficult and misses unknown human biology, and that basic experimental data types are not usually well connected to patient-data analysis.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
0 of 19 cited sources checked · 0 supported · 0 flagged
- Interpretationreasoning
Current black-box machine learning will be intrinsically limited in applications to patient data because of the very high complexity of biological systems.
“Here, we argue that the answer is no–that the very high complexity of biological systems will intrinsically limit applications of current “black box” machine learning in patient data.”
Quote in the paperSubstantial claim- Rests on an assumption: Biological systems are almost certainly more complex than those addressed by machine learning in other areas.
- Interpretationcitation
Direct incorporation of the biological structure of cells and tissues leads to a ready-made working model of how biological inputs, such as genotype, are compressed to determine outcomes.
“In contrast, direct incorporation of the biological structure of cells and tissues, such as explored by Lin et al. (2017) and Ma et al. (2018), leads to a ready-made working model of how biological inputs, such as genotype, are compressed to determine outcomes.”
Quote in the paperSubstantial claim0 of 2 sources checked- cites Lin et al., 2017 source
Not checked in the free run.
- cites Ma et al., 2018
Not checked in the free run.
- Premisecitation
Network connections can nominate a set of genes whose input data should be aggregated to create a composite feature, such as pooling rarely altered genes in cancer into a single pathway that is recurrently altered across a patient population.
“Network connections can also nominate a set of genes whose input data should be aggregated to create a composite feature, such as pooling of rarely altered genes in cancer into a single pathway, which, viewed as a single aggregate, is recurrently altered across a patient population (Alvarez et al., 2016; Chuang et al., 2007).”
Quote in the paperNot a substantial claim0 of 2 sources checked - Premisecitation
In DCell, neurons are organized into banks, each of which maps to a distinct biological module within a large hierarchy of known cellular components and processes.
“Neurons are organized into banks, each of which maps to a distinct biological module within a large hierarchy of known cellular components and processes.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Ma et al., 2018
Not checked in the free run.
- Premisecitation
Biological systems are complex machines whose outputs emerge from a hierarchy of internal components.
“Like man-made engines, biological systems are also complex machines whose outputs emerge from a hierarchy of internal components (Simon, 1962).”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Simon, 1962
Not checked in the free run.
- Interpretationreasoning
Even for common conditions like cancer and heart disease, the number of available datasets will saturate at a few million patient examples.
“Even for common conditions like cancer and heart disease, the number of available datasets will saturate at a few million patient examples.”
Quote in the paperNot a substantial claim- Rests on an assumption: The total number of patients is finite.
- Premisecitation
Increased model resolution may come at the expense of model scope.
“For instance, an important lesson is that increased model resolution (e.g., introducing many detailed biochemical parameters) may come at the expense of model scope (e.g., the ability to address all biological elements and generalize to new patient cohorts).”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Fisher and Henzinger, 2007
Not checked in the free run.
- Interpretationreasoning
Even after obtaining the genome sequence of every patient, the genetic patterns driving disease may still remain undiscovered by current statistical methods.
“As a consequence, even after obtaining the genome sequence of every patient, the genetic patterns driving disease may still remain undiscovered by current statistical methods.”
Quote in the paperSubstantial claim- Rests on an assumption: Biological systems are almost certainly more complex than those addressed by machine learning in other areas.
- Rests on an assumption: The total number of patients is finite.
- Interpretationreasoning
Prior knowledge of biological structure might provide distinct advantages to models capable of incorporating this information.
“It is nonetheless easy to see how prior knowledge of biological structure might provide distinct advantages to models capable of incorporating this information–what we here call visible learning–and recent research has begun to prove the concept.”
Quote in the paperNot a substantial claim - Interpretationreasoning
Prior knowledge of biological structure can address the problem of data heterogeneity, since different input patterns, even when entirely distinct, converge on common higher-order biological processing units corresponding to discrete modular components of cells and tissues.
“First, prior knowledge of biological structure can address the problem of data heterogeneity, since different input patterns, even when entirely distinct from one another (i.e., without common patterns), converge on common higher-order biological processing units corresponding to discrete modular components of cells and tissues.”
Quote in the paperNot a substantial claim - Premisecitation
DCell is a model of a basic eukaryotic cell that uses a deep neural network to translate a list of mutated genes (genotype input) to the resulting cell proliferation rate (phenotype output).
“We extended such a strategy to assemble DCell, a model of a basic eukaryotic cell (Ma et al., 2018). DCell uses a deep neural network to translate a list of mutated genes (genotype input) to the resulting cell proliferation rate (phenotype output).”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Ma et al., 2018
Not checked in the free run.
- Premisecitation
Machine-learning systems face input heterogeneity as a challenge that is particularly acute in biomedical applications.
“Machine-learning systems face two recognized challenges that become particularly acute in biomedical applications. The first is input heterogeneity.”
Quote in the paperSubstantial claim0 of 3 sources checked - Interpretationreasoning
A predicted change in cell phenotype can be interpreted by examining the functional states of underlying cellular components internal to DCell whose neuron states are also highly affected.
“A predicted change in cell phenotype can then be interpreted by examining the functional states of underlying cellular components, internal to DCell, whose neuron states are also highly affected.”
Quote in the paperSubstantial claim - Premisecitation
Lin et al. (2017) predicted cell type and state using a deep neural network whose structure was determined based on the hierarchical organization of transcriptional regulatory factors in the nucleus.
“This idea was recently explored by Lin et al. (2017), who predicted cell type and state using a deep neural network for which the structure was determined based on the hierarchical organization of transcriptional regulatory factors in the nucleus.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Lin et al., 2017 source
Not checked in the free run.
- Premisecitation
Alvarez et al. (2016) used a transcriptional regulatory network to translate patient mRNA expression profiles to activities of regulatory proteins, most of which are difficult to interrogate experimentally.
“For example, Alvarez et al. (2016) used a transcriptional regulatory network to translate patient mRNA expression profiles to activities of regulatory proteins, most of which are difficult to interrogate experimentally.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Alvarez et al., 2016 source
Not checked in the free run.
- May combine two claims. 10
- Classified as probably not a substantial claim. 1
May combine two claims.
InterpretationreasoningA few million patient examples is modest compared to the amounts of data often needed to train a statistical model, compounded by the as-yet-unknown but undoubtedly high complexity of biological systems.
“While a million may seem large, this number is modest compared to the amounts of data often needed to train a statistical model, compounded by the as-yet-unknown, but undoubtedly high, complexity of biological systems.”
Quote in the paperNot a substantial claim- Rests on an assumption: Biological systems are almost certainly more complex than those addressed by machine learning in other areas.
May combine two claims.
PremisecitationSimilar heterogeneity arises in patient data from nearly all common diseases, including cardiovascular, metabolic, and neurodevelopmental disorders, in which recurrent patterns are elusive, making reliable predictions difficult.
“Similar heterogeneity arises in patient data from nearly all common diseases, including cardiovascular, metabolic, and neurodevelopmental disorders, in which recurrent patterns are elusive, making it difficult to make reliable predictions (Boyle et al., 2017).”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Boyle et al., 2017 source
Not checked in the free run.
May combine two claims.
PremisecitationCancer can arise from many different combinations of genetic alterations involving many potential genes, any one of which may be mutated only rarely, so each new patient presents a distinct constellation of molecular changes never before seen in nature.
“For example, cancer can arise as the result of many different combinations of genetic alterations involving many potential genes, any one of which may be mutated only rarely; as a consequence, each new patient presents a distinct constellation of molecular changes never before seen in nature (Alvarez et al., 2016; Kourou et al., 2014).”
Quote in the paperSubstantial claim0 of 2 sources checkedMay combine two claims.
PremisecitationIn biomedicine, many machine-learning methods are being developed to predict patient outcomes but typically do not link predictions to underlying mechanisms.
“Similarly, in biomedicine, many machine-learning methods are being developed to predict patient outcomes (Kourou et al., 2014), but these approaches typically do not link predictions to underlying mechanisms (Figure 1B).”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Kourou et al., 2014 source
Not checked in the free run.
May combine two claims.
PremisecitationModern machine learning models, including deep neural networks, are black boxes that focus on predicting outputs from inputs without regard for the mechanism or rationale by which an outcome is brought about, a challenge particularly acute in biomedicine.
“A second well-known challenge is that modern machine learning models, including deep neural networks, are black boxes, devices which focus on predicting outputs from inputs without regard for the mechanism or rationale by which a particular outcome is brought about.”
Quote in the paperSubstantial claim0 of 2 sources checkedMay combine two claims.
InterpretationreasoningExecution of a visible model reveals the states of internal biological systems, and the most striking internal states provide hypotheses about the underlying mechanisms governing patient phenotype.
“Given input patient data, execution of the model not only produces a final output state; it also reveals the states of internal biological systems. The most striking of these internal states provide hypotheses as to the underlying mechanisms governing patient phenotype, which is important because many internal biological states are difficult to measure through direct experimental observations.”
Quote in the paperSubstantial claimMay combine two claims.
InterpretationreasoningMachine learning will not replace the need for experimental cell and tissue biology; it will be substantially enabled by such knowledge, given the right visible intelligence infrastructure.
“That is, machine learning will not replace the need for experimental cell and tissue biology; it will be substantially enabled by such knowledge, given the right visible intelligence infrastructure.”
Quote in the paperNot a substantial claim- Rests on an assumption: Causal mechanistic insights are key to identifying drug targets and advancing basic biological knowledge.
May combine two claims.
InterpretationreasoningInternal states of the model may indicate biological components that can be targeted by therapeutic interventions or form the basis for in silico testing of treatment combinations.
“Internal states of the model may also indicate biological components that can be targeted by therapeutic interventions or form the basis for in silico testing of treatment combinations.”
Quote in the paperSubstantial claim- Rests on an assumption: Causal mechanistic insights are key to identifying drug targets and advancing basic biological knowledge.
May combine two claims.
InterpretationreasoningModels guided by biological structure can be interpreted mechanistically, informing understanding of the system and suggesting potential therapeutic strategies.
“Second, models guided by biological structure can be interpreted mechanistically, informing our understanding of the system and suggesting potential therapeutic strategies.”
Quote in the paperSubstantial claim- Rests on an assumption: Causal mechanistic insights are key to identifying drug targets and advancing basic biological knowledge.
Probably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.
InterpretationreasoningParallel development of these directions will enable a new generation of biomedical machine learning, replacing black-box models that focus on isolated problem domains with visible models that survey general biological systems.
“Parallel development of these directions will enable a new generation of biomedical machine learning, replacing black-box models that focus on isolated problem domains with visible models that survey general biological systems.”
Quote in the paperNot a substantial claim
The paper relies on this without support.
Biological systems are almost certainly more complex than those addressed by machine learning in other areas.
“Although such input heterogeneity is a property of many and perhaps all complex systems, biological systems are almost certainly more complex than those addressed by machine learning in other areas.”
The paper relies on this without support.
The total number of patients is finite.
“The total number of patients is finite, however.”
The paper relies on this without support.
Causal mechanistic insights are key to identifying drug targets and advancing basic biological knowledge.
“This is a missed opportunity, as causal mechanistic insights are key to identifying drug targets and advancing basic biological knowledge.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
AI/ML network architectures can be crafted following the logic of a hierarchy (ontology) that reflects the known machinery of cells.
“Ideker and co-workers showed that AI/ML network architectures can be crafted following the logic of a hierarchy (ontology) that reflects the known machinery of cells.182,183”
From Advancing Targeted Protein Degradation via Multiomics Profiling and Artificial Intelligence (2023) · cites this paper as [182]
Cites
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126 · see citing works
Mastering the game of Go with deep neural networks and tree search
Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., van den Driessche, G., Schrittwieser, J., Antonoglou, I., Panneershelvam, V., Lanctot, M., et al. (2016) · Nature · doi:10.1038/nature16961 · Cited by 15,612 · see citing works
“Why Should I Trust You?”: Explaining the Predictions of Any Classifier
Ribeiro, M., Singh, S., & Guestrin, C. (2016) · Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations · doi:10.18653/v1/n16-3020 · Cited by 6,682 · see citing works
The Architecture of Complexity
Simon, H. (1991) · Springer eBooks · doi:10.1007/978-1-4899-0718-9_31 · Cited by 6,459 · see citing works
Machine Learning Applications in Cancer Prognosis and Prediction
Kourou, K., Exarchos, T. P., Exarchos, K. P., Karamouzis, M. V., & Fotiadis, D. I. (2015) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2014.11.005 · Cited by 3,338 · see citing works
An Expanded View of Complex Traits: From Polygenic to Omnigenic
Boyle, E. A., Li, Y. I., & Pritchard, J. K. (2017) · Cell · doi:10.1016/j.cell.2017.05.038 · Cited by 2,851 · see citing works
Network‐based classification of breast cancer metastasis
Chuang, H., Lee, E., Liu, Y., Lee, D., & Ideker, T. (2007) · Molecular Systems Biology · doi:10.1038/msb4100180 · Cited by 1,431 · see citing works
Self-citationA Whole-Cell Computational Model Predicts Phenotype from Genotype
Karr, J. R., Sanghvi, J. C., Macklin, D. N., Gutschow, M. V., Jacobs, J. M., Bolival, B., Assad-Garcia, N., Glass, J. I., & Covert, M. W. (2012) · Cell · doi:10.1016/j.cell.2012.05.044 · Cited by 1,300 · cited 2× · see citing works
“emerging (Karr et al., 2012), an idea that fits naturally with deep (multi-layered) neural networks.”
BackgroundAlvarez, M. J., Shen, Y., Giorgi, F. M., Lachmann, A., Ding, B. B., Ye, B. H., & Califano, A. (2016) · Nature Genetics · doi:10.1038/ng.3593 · Cited by 966 · see citing works
The Architecture of Complexity
Barabási, A.-Ĺ. (2007) · IEEE Control Systems · doi:10.1109/mcs.2007.384127 · Cited by 819 · see citing works
Rationalizing Neural Predictions
Lei, T., Barzilay, R., & Jaakkola, T. (2016) · Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d16-1011 · Cited by 777 · see citing works
"Why Should I Trust You?": Explaining the Predictions of Any ClassifierPreprint
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1602.04938 · Cited by 587 · see citing works
Fisher, J., & Henzinger, T. A. (2007) · Nature Biotechnology · doi:10.1038/nbt1356 · Cited by 543 · see citing works
Self-citationUsing deep learning to model the hierarchical structure and function of a cell
Ma, J., Yu, M. K., Fong, S., Ono, K., Sage, E., Demchak, B., Sharan, R., & Ideker, T. (2018) · Nature Methods · doi:10.1038/nmeth.4627 · Cited by 435 · see citing works
Self-citationTorkamani, A., Andersen, K. G., Steinhubl, S. R., & Topol, E. J. (2017) · Cell · doi:10.1016/j.cell.2017.08.007 · Cited by 245 · see citing works
Using neural networks for reducing the dimensions of single-cell RNA-Seq data
Lin, C., Jain, S., Kim, H., & Bar-Joseph, Z. (2017) · Nucleic Acids Research · doi:10.1093/nar/gkx681 · Cited by 197 · see citing works
Leiserson, M. D. M., Eldridge, J. V., Ramachandran, S., & Raphael, B. J. (2013) · Current Opinion in Genetics & Development · doi:10.1016/j.gde.2013.09.003 · Cited by 93 · see citing works
Self-citation
Flagged first, then by citations
1 of the 18 references couldn’t be matched to a known paper.
Cited by
Sorted by citations
Artificial intelligence in COVID-19 drug repurposing
Zhou, Y., Wang, F., Tang, J., Nussinov, R., & Cheng, F. (2020) · The Lancet Digital Health · doi:10.1016/s2589-7500(20)30192-8 · Cited by 636 · see citing works
Predicting drug response and synergy using a deep learning model of human cancer cells
Kuenzi, B. M., Park, J., Fong, S. H., Sanchez, K. S., Lee, J., Kreisberg, J. F., Ma, J., & Ideker, T. (2020) · Cancer Cell · doi:10.1016/j.ccell.2020.09.014 · Cited by 544 · see citing works
Self-citationFrom Big Data to Precision Medicine
Hulsen, T., Jamuar, S. S., Moody, A. R., Karnes, J. H., Varga, O., Hedensted, S., Spreafico, R., Hafler, D. A., & McKinney, E. F. (2019) · Frontiers in Medicine · doi:10.3389/fmed.2019.00034 · Cited by 466 · see citing works
Systems Biology of Cancer Metastasis
Suhail, Y., Cain, M. P., Vanaja, K. G., Kurywchak, P. A., Levchenko, A., Kalluri, R., & Gupta, K. (2019) · Cell Systems · doi:10.1016/j.cels.2019.07.003 · Cited by 455 · see citing works
Machine Learning in Enzyme Engineering
Mazurenko, S., Prokop, Z., & Damborský, J. (2019) · ACS Catalysis · doi:10.1021/acscatal.9b04321 · Cited by 410 · see citing works
In vitro neurons learn and exhibit sentience when embodied in a simulated game-world
Kagan, B. J., Kitchen, A. C., Tran, N. T., Habibollahi, F., Khajehnejad, M., Parker, B. J., Bhat, A., Rollo, B., Razi, A., & Friston, K. J. (2022) · Neuron · doi:10.1016/j.neuron.2022.09.001 · Cited by 400 · see citing works
How to build the virtual cell with artificial intelligence: Priorities and opportunities
Bunne, C., Roohani, Y., Rosen, Y., Gupta, A., Zhang, X., Roed, M., Alexandrov, T., AlQuraishi, M., Brennan, P., Burkhardt, D. B., et al. (2024) · Cell · doi:10.1016/j.cell.2024.11.015 · Cited by 393 · see citing works
The Application of Deep Learning in Cancer Prognosis Prediction
Zhu, W., Xie, L., Han, J., & Guo, X. (2020) · Cancers · doi:10.3390/cancers12030603 · Cited by 384 · see citing works
Recent progress in multienzymes co-immobilization and multienzyme system applications
Ren, S., Li, C., Jiao, X., Jia, S., Jiang, Y., Bilal, M., & Cui, J. (2019) · Chemical Engineering Journal · doi:10.1016/j.cej.2019.05.141 · Cited by 345 · see citing works
A White-Box Machine Learning Approach for Revealing Antibiotic Mechanisms of Action
Yang, J. H., Wright, S. N., Hamblin, M., McCloskey, D., Alcantar, M. A., Schrübbers, L. C., Lopatkin, A. J., Satish, S., Nili, A., Palsson, B. Ø., et al. (2019) · Cell · doi:10.1016/j.cell.2019.04.016 · Cited by 317 · see citing works
Machine Learning in Nanoscience: Big Data at Small Scales
Brown, K. A., Brittman, S., Maccaferri, N., Jariwala, D., & Celano, U. (2020) · Nano Letters · doi:10.1021/acs.nanolett.9b04090 · Cited by 229 · see citing works
Accelerating antibiotic discovery through artificial intelligence
Melo, M. C. R., Maasch, J. R. M. A., & de la Fuente-Nunez, C. (2021) · Communications Biology · doi:10.1038/s42003-021-02586-0 · Cited by 234 · see citing works
Computational network biology: Data, models, and applications
Liu, C., Ma, Y., Zhao, J., Nussinov, R., Zhang, Y., Cheng, F., & Zhang, Z. (2019) · Physics Reports · doi:10.1016/j.physrep.2019.12.004 · Cited by 226 · see citing works
Kong, J., Lee, H., Kim, D., Han, S. K., Ha, D., Shin, K., & Kim, S. (2020) · Nature Communications · doi:10.1038/s41467-020-19313-8 · Cited by 219 · see citing works
Molecular networks in Network Medicine: Development and applications
Silverman, E. K., Schmidt, H. H. H. W., Anastasiadou, E., Altucci, L., Angelini, M., Badimón, L., Balligand, J., Benincasa, G., Capasso, G., Conte, F., et al. (2020) · WIREs Systems Biology and Medicine · doi:10.1002/wsbm.1489 · Cited by 211 · see citing works
Artificial intelligence for precision oncology: beyond patient stratification
Azuaje, F. J. (2019) · npj Precision Oncology · doi:10.1038/s41698-019-0078-1 · Cited by 203 · see citing works
Alabi, R. O., Youssef, O., Pirinen, M., Elmusrati, M. S., Mäkitie, A. A., Leivo, I. V. V., & Almangush, A. (2021) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2021.102060 · Cited by 178 · see citing works
Artificial Intelligence for the Otolaryngologist: A State of the Art Review
Bur, A. M., Shew, M. A., & New, J. (2019) · Otolaryngology · doi:10.1177/0194599819827507 · Cited by 172 · see citing works
Multimodal learning with graphs
Ektefaie, Y., Dasoulas, G., Noori, A., Farhat, M., & Zitnik, M. (2023) · Nature Machine Intelligence · doi:10.1038/s42256-023-00624-6 · Cited by 166 · see citing works
Machine learning to predict occult nodal metastasis in early oral squamous cell carcinoma
Bur, A. M., Holcomb, A. J., Goodwin, S., Woodroof, J. M., Karadaghy, O. A., Shnayder, Y., Kakarala, K., Brant, J. A., & Shew, M. A. (2019) · Oral Oncology · doi:10.1016/j.oraloncology.2019.03.011 · Cited by 147 · see citing works
Zhang, Y., Chen, J., Lin, Y., Chan, S., Zhou, J., Chow, D. S., Chang, P., Kwong, T. C., Yeh, D., Wang, X., et al. (2020) · European Radiology · doi:10.1007/s00330-020-07274-x · Cited by 144 · see citing works
Lind, A. P., & Anderson, P. C. (2019) · PLoS ONE · doi:10.1371/journal.pone.0219774 · Cited by 141 · see citing works
Detection and Classification of Pulmonary Nodules Using Convolutional Neural Networks: A Survey
Monkam, P., Qi, S., Ma, H., Gao, W., Yao, Y., & Qian, W. (2019) · IEEE Access · doi:10.1109/access.2019.2920980 · Cited by 131 · see citing works
Transcriptional trajectories of human kidney injury progression
Cippà, P. E., Sun, B., Liu, J., Chen, L., Naesens, M., & McMahon, A. P. (2018) · JCI Insight · doi:10.1172/jci.insight.123151 · Cited by 122 · see citing works
Dual Machine-Learning System to Aid Glaucoma Diagnosis Using Disc and Cup Feature Extraction
Civit-Masot, J., Domínguez-Morales, M., Vicente-Díaz, S., & Civit, A. (2020) · IEEE Access · doi:10.1109/access.2020.3008539 · Cited by 120 · see citing works
Karadaghy, O. A., Shew, M. A., New, J., & Bur, A. M. (2019) · JAMA Otolaryngology–Head & Neck Surgery · doi:10.1001/jamaoto.2019.0981 · Cited by 114 · see citing works
GenNet framework: interpretable deep learning for predicting phenotypes from genetic data
van Hilten, A., Kushner, S. A., Kayser, M., Ikram, M. A., Adams, H. H. H., Klaver, C. C. W., Niessen, W. J., & Roshchupkin, G. V. (2021) · Communications Biology · doi:10.1038/s42003-021-02622-z · Cited by 112 · see citing works
Privacy-Preserving Artificial Intelligence Techniques in Biomedicine
Torkzadehmahani, R., Nasirigerdeh, R., Blumenthal, D. B., Kacprowski, T., List, M., Matschinske, J., Späth, J., Wenke, N. K., & Baumbach, J. (2022) · Methods of Information in Medicine · doi:10.1055/s-0041-1740630 · Cited by 117 · see citing works
The social nature of mitochondria: Implications for human health
Picard, M. A., & Sandi, C. (2020) · Neuroscience & Biobehavioral Reviews · doi:10.1016/j.neubiorev.2020.04.017 · Cited by 102 · see citing works
Artificial Intelligence and Mechanistic Modeling for Clinical Decision Making in Oncology
Benzekry, S. (2020) · Clinical Pharmacology & Therapeutics · doi:10.1002/cpt.1951 · Cited by 89 · see citing works
Yoo, T. K., Ryu, I. H., Choi, H., Kim, J. K., Lee, I. S., Kim, J. S., Lee, G., & Rim, T. H. (2020) · Translational Vision Science & Technology · doi:10.1167/tvst.9.2.8 · Cited by 82 · see citing works
A biochemically-interpretable machine learning classifier for microbial GWAS
Kavvas, E. S., Yang, L., Monk, J. M., Heckmann, D., & Palsson, B. Ø. (2020) · Nature Communications · doi:10.1038/s41467-020-16310-9 · Cited by 82 · see citing works
Executable cancer models: successes and challenges
Clarke, M. A., & Fisher, J. (2020) · Nature reviews. Cancer · doi:10.1038/s41568-020-0258-x · Cited by 76 · see citing works
Self-citationPredicting Anticancer Drug Response With Deep Learning Constrained by Signaling Pathways
Zhang, H., Chen, Y., & Li, F. (2021) · Frontiers in Bioinformatics · doi:10.3389/fbinf.2021.639349 · Cited by 58 · see citing works
Park, S., Silva, E., Singhal, A., Kelly, M. R., Licon, K. S., Panagiotou, I., Fogg, C., Fong, S. H., Lee, J. J. Y., Zhao, X., et al. (2024) · Nature Cancer · doi:10.1038/s43018-024-00740-1 · Cited by 60 · see citing works
Self-citationSignal Peptide Efficiency: From High-Throughput Data to Prediction and Explanation
Grasso, S., Dabene, V., Hendriks, M. M. W. B., Zwartjens, P., Pellaux, R., Held, M., Panke, S., van Dijl, J. M., Meyer, A., & van Rij, T. (2023) · ACS Synthetic Biology · doi:10.1021/acssynbio.2c00328 · Cited by 58 · see citing works
Machine Learning‐Assisted High‐Throughput Screening of Nanozymes for Ulcerative Colitis
Zhao, X., Yu, Y., Xu, X., Zhang, Z., Chen, Z., Gao, Y., Zhong, L., Chen, J., Huang, J., Qin, J., et al. (2025) · Advanced Materials · doi:10.1002/adma.202417536 · Cited by 57 · see citing works
Integration of artificial intelligence in clinical laboratory medicine: Advancements and challenges
Xie, H., Jia, Y. H., & Liu, S. (2024) · Interdisciplinary medicine · doi:10.1002/inmd.20230056 · Cited by 52 · see citing works
Zhang, T., Hasib, M. M., Chiu, Y., Han, Z.-F., Jin, Y.-F., Flores, M. A., Chen, Y., & Huang, Y. (2022) · Cancers · doi:10.3390/cancers14194763 · Cited by 55 · see citing works
Advancing Targeted Protein Degradation via Multiomics Profiling and Artificial Intelligence
Duran-Frigola, M., Cigler, M., & Winter, G. E. (2023) · Journal of the American Chemical Society · doi:10.1021/jacs.2c11098 · Cited by 50 · see citing works
Masoudi‐Sobhanzadeh, Y., Omidi, Y., Amanlou, M., & Masoudi‐Nejad, A. (2019) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2019.05.006 · Cited by 47 · see citing works
Digital Insights Into Nucleotide Metabolism and Antibiotic Treatment Failure
Lopatkin, A. J., & Yang, J. H. (2021) · Frontiers in Digital Health · doi:10.3389/fdgth.2021.583468 · Cited by 47 · see citing works
Machine Learning to Predict Delays in Adjuvant Radiation following Surgery for Head and Neck Cancer
Shew, M. A., New, J., & Bur, A. M. (2019) · Otolaryngology · doi:10.1177/0194599818823200 · Cited by 43 · see citing works
Kim, S.-H., Jeon, E., Yu, S., Oh, K., Kim, C. K., Song, T., Kim, Y., Heo, S. H., Park, K., Kim, J., et al. (2021) · Scientific Reports · doi:10.1038/s41598-021-99920-7 · Cited by 43 · see citing works
Incorporating biological structure into machine learning models in biomedicine
Crawford, J., & Greene, C. S. (2020) · Current Opinion in Biotechnology · doi:10.1016/j.copbio.2019.12.021 · Cited by 41 · see citing works
Machine learning in the loop for tuberculosis diagnosis support
Orjuela-Cañón, Á. D., Jutinico, A. L., Awad, C. E., Vergara, E., & Palencia, M. A. (2022) · Frontiers in Public Health · doi:10.3389/fpubh.2022.876949 · Cited by 40 · see citing works
From genotype to phenotype: augmenting deep learning with networks and systems biology
Gazestani, V., & Lewis, N. E. (2019) · Current Opinion in Systems Biology · doi:10.1016/j.coisb.2019.04.001 · Cited by 39 · see citing works
Gal, J., Bailleux, C., Chardin, D., Pourcher, T., Gilhodes, J., Jing, L., Guigonis, J., Ferrero, J.-., Milano, G. A., Mograbi, B., et al. (2020) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2020.05.021 · Cited by 39 · see citing works
Designing interpretable deep learning applications for functional genomics: a quantitative analysis
van Hilten, A., Katz, S., Saccenti, E., Niessen, W. J., & Roshchupkin, G. V. (2024) · Briefings in Bioinformatics · doi:10.1093/bib/bbae449 · Cited by 39 · see citing works
Computational Methods for the Discovery of Metabolic Markers of Complex Traits
Lee, M. Y., & Hu, T. (2019) · Metabolites · doi:10.3390/metabo9040066 · Cited by 38 · see citing works
Big Data and Actuarial Science
Hassani, H., Unger, S., & Beneki, C. C. (2020) · Big Data and Cognitive Computing · doi:10.3390/bdcc4040040 · Cited by 37 · see citing works
Harnessing the Power of Artificial Intelligence in Otolaryngology and the Communication Sciences
Wilson, B. S., Tucci, D. L., Moses, D. A., Chang, E. F., Young, N. M., Zeng, F., Lesica, N. A., Bur, A. M., Kavookjian, H. L., Mussatto, C. C., et al. (2022) · Journal of the Association for Research in Otolaryngology · doi:10.1007/s10162-022-00846-2 · Cited by 36 · see citing works
Interpretable machine learning for prediction of clinical outcomes in acute ischemic stroke
Lee, J., Park, K. M., & Park, S.-H. (2023) · Frontiers in Neurology · doi:10.3389/fneur.2023.1234046 · Cited by 33 · see citing works
Comte, B., Baumbach, J., Benis, A., Basílio, J., Debeljak, N., Flobak, Å., Franken, C., Harel, N., HeFeng, F. Q., Kuiper, M. T. R., et al. (2020) · Network and Systems Medicine · doi:10.1089/nsm.2020.0004 · Cited by 33 · see citing works
Trends in Deep Learning for Property-driven Drug Design
Born, J., & Manica, M. (2021) · Current Medicinal Chemistry · doi:10.2174/0929867328666210729115728 · Cited by 35 · see citing works
AI-readiness Criteria for Biomedical Data
Clark, T., Caufield, H., Parker, J. A., Al Manir, S., Amorim, E., Eddy, J., Gim, N., Gow, B., Goar, W., Hansen, J. N., et al. (2024) · doi:10.1101/2024.10.23.619844 · Cited by 32 · see citing works
Shew, M. A., New, J., Wichova, H., Koestler, D. C., & Staecker, H. (2019) · Scientific Reports · doi:10.1038/s41598-019-40192-7 · Cited by 33 · see citing works
Casimiro‐Soriguer, C. S., Loucera, C., Peña-Chilet, M., & Dopazo, J. (2022) · Scientific Reports · doi:10.1038/s41598-021-04182-y · Cited by 36 · see citing works
Deep neural network prediction of genome-wide transcriptome signatures – beyond the Black-box
Magnusson, R., Tegner, J. N., & Gustafsson, M. (2022) · npj Systems Biology and Applications · doi:10.1038/s41540-022-00218-9 · Cited by 31 · see citing works
Jabeer, A., Temiz, M., Bakır-Güngör, B., & Yousef, M. (2023) · Frontiers in Genetics · doi:10.3389/fgene.2022.1076554 · Cited by 31 · see citing works
Advances in Artificial Intelligence to Diagnose Otitis Media: State of the Art Review
Ngombu, S., Binol, H., Gürcan, M. N., & Moberly, A. C. (2022) · Otolaryngology · doi:10.1177/01945998221083502 · Cited by 29 · see citing works
Pisu, F., Williamson, B. J., Nardi, V., Paraskevas, K. I., Puig, J., Vagal, A. S., Rubeis, G. D., Porcu, M., Cau, R., Benson, J. C., et al. (2024) · Circulation Cardiovascular Imaging · doi:10.1161/circimaging.123.016274 · Cited by 27 · see citing works
van Hilten, A., van Rooij, J., Heijmans, B. T., Hoen, P. A. C. ’., van Meurs, J. B. J., Jansen, R., Franke, L., Boomsma, D. I., Pool, R., van Dongen, J., et al. (2024) · npj Systems Biology and Applications · doi:10.1038/s41540-024-00405-w · Cited by 27 · see citing works
Erdem, C., Mutsuddy, A., Bensman, E. M., Dodd, W. B., Saint-Antoine, M., Bouhaddou, M., Blake, R. C., Gross, S. M., Heiser, L. M., Feltus, F. A., et al. (2022) · Nature Communications · doi:10.1038/s41467-022-31138-1 · Cited by 26 · see citing works
Unveiling new disease, pathway, and gene associations via multi-scale neural network
Gaudelet, T., Malod-Dognin, N., Sánchez-Valle, J., Pancaldi, V., Valencia, A., & Pržulj, N. (2020) · PLoS ONE · doi:10.1371/journal.pone.0231059 · Cited by 25 · see citing works
Chou, A., Espin, A. T., Kyritsis, N., Huie, J. R., Khatry, S., Funk, J., Hay, J., Lofgreen, A., Shah, R. R., McCann, C., et al. (2022) · PLoS ONE · doi:10.1371/journal.pone.0265254 · Cited by 24 · see citing works
Medina-Ortiz, D., Contreras, S., Quiroz, C., & Olivera‐Nappa, Á. (2020) · Frontiers in Molecular Biosciences · doi:10.3389/fmolb.2020.00013 · Cited by 23 · see citing works
Li, Z., Jin, K., Chen, H., Zhang, L., Zhang, G., Jiang, Y., Zou, H., Wang, W., Qi, G., & Qu, X. (2022) · Nanoscale · doi:10.1039/d1nr07452k · Cited by 23 · see citing works
Bourgeais, V., Zehraoui, F., & Hanczar, B. (2022) · Bioinformatics · doi:10.1093/bioinformatics/btac147 · Cited by 22 · see citing works
Industrializing AI-powered drug discovery: lessons learned from the Patrimony computing platform
Guedj, M., Swindle, J., Hamon, A., Hubert, S., Desvaux, E., Laplume, J., Xuereb, L., Lefèbvre, C., Haudry, Y., Gabarroca, C., et al. (2022) · Expert Opinion on Drug Discovery · doi:10.1080/17460441.2022.2095368 · Cited by 23 · see citing works
How Computation Is Helping Unravel the Dynamics of Morphogenesis
Pastor-Escuredo, D., & del Álamo, J. C. (2020) · Frontiers in Physics · doi:10.3389/fphy.2020.00031 · Cited by 20 · see citing works
Using interpretable deep learning to model cancer dependencies
Lin, C.-H., & Lichtarge, O. (2021) · Bioinformatics · doi:10.1093/bioinformatics/btab137 · Cited by 20 · see citing works
Medina-Ortiz, D., Contreras, S., Quiroz, C., Asenjo, J. A., & Olivera‐Nappa, Á. (2020) · Information Systems · doi:10.1016/j.is.2020.101557 · Cited by 20 · see citing works
Smith, J. B., Shew, M. A., Karadaghy, O. A., Nallani, R., Sykes, K. J., Gan, G. N., Brant, J. A., & Bur, A. M. (2020) · Head & Neck · doi:10.1002/hed.26246 · Cited by 18 · see citing works
Multi-omics Cancer Prognosis Analysis Based on Graph Convolution Network
Wang, Y., Zhang, Z., Chai, H., & Yang, Y. (2021) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm52615.2021.9669797 · Cited by 18 · see citing works
Roy, G. G., Geard, N. L., Verspoor, K. M., & He, S. (2022) · Bioinformatics · doi:10.1093/bioinformatics/btac636 · Cited by 18 · see citing works
Yang, C., Su, X., Liu, D., Guo, Z., Wang, F., & Lu, Y. (2020) · Frontiers in Marine Science · doi:10.3389/fmars.2020.00032 · Cited by 17 · see citing works
Shew, M. A., Wichova, H., Bur, A. M., Koestler, D. C., Peter, M. S., Warnecke, A., & Staecker, H. (2020) · Otolaryngology · doi:10.1177/0194599820940649 · Cited by 15 · see citing works
Prediction of anticancer drug sensitivity using an interpretable model guided by deep learning
Pang, W., Chen, M., & Qin, Y. (2024) · BMC Bioinformatics · doi:10.1186/s12859-024-05669-x · Cited by 14 · see citing works
Machine learning-assisted synthetic biology of cyanobacteria and microalgae
Jin, W., Wang, F., Chen, L., & Zhang, W. (2025) · Algal Research · doi:10.1016/j.algal.2025.103911 · Cited by 14 · see citing works
Zhao, Y., Chen, Z., Zhang, Y., Zhou, J., Chen, J., Lee, K. E., Combs, F. J., Parajuli, R., Mehta, R. S., Wang, M., et al. (2021) · Frontiers in Oncology · doi:10.3389/fonc.2021.774248 · Cited by 14 · see citing works
Mechanism-Centric Approaches for Biomarker Detection and Precision Therapeutics in Cancer
Yu, C. Y., & Mitrofanova, A. (2021) · Frontiers in Genetics · doi:10.3389/fgene.2021.687813 · Cited by 14 · see citing works
Clark, T., Mohan, J., Schaffer, L. V., Obernier, K., Manir, S. A., Churas, C., Dailamy, A., Doctor, Y., Forget, A., Hansen, J. N., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.05.21.589311 · Cited by 14 · see citing works
Self-citationA REVIEW OF ARTIFICIAL INTELLIGENCE IN TREATMENT OF COVID-19
Gandla, K., Teja, K., Babu, P. V., Sagapola, R., & Sudhakar, P. (2022) · Journal of Pharmaceutical Negative Results · doi:10.47750/pnr.2022.13.s01.31 · Cited by 13 · see citing works
Machine learning directed sentinel lymph node biopsy in cutaneous head and neck melanoma
Oliver, J. R., Karadaghy, O. A., Fassas, S. N., Arambula, Z., & Bur, A. M. (2022) · Head & Neck · doi:10.1002/hed.26993 · Cited by 13 · see citing works
Yang, T., Hu, Y., Pan, X., Lou, S., Zou, J., Deng, Q., Zhang, Q. C., Zhou, J., & Zhu, J. (2023) · Brain Sciences · doi:10.3390/brainsci13040557 · Cited by 12 · see citing works
Alabi, R. O., Sjöblom, A., Carpén, T. P., Elmusrati, M. S., Leivo, I. V. V., Almangush, A., & Mäkitie, A. A. (2023) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2023.105064 · Cited by 12 · see citing works
Yang, J. Z., Dokpuang, D., Nemati, R., He, K. H., Zheng, A. B., Petrov, M. S., & Lu, J. (2020) · Biomedicines · doi:10.3390/biomedicines8060174 · Cited by 12 · see citing works
Challenges and Opportunities in Analyzing Cancer-Associated Microbiomes
Chia, M., Pop, M., Salzberg, S. L., & Nagarajan, N. (2025) · Cancer Research · doi:10.1158/0008-5472.can-24-3629 · Cited by 11 · see citing works
Zhang, H., Cao, D., Chen, Z., Zhang, X., Chen, Y., Sessions, C., Cruchaga, C., Payne, P., Li, G., Province, M. A., et al. (2024) · Bioinformatics Advances · doi:10.1093/bioadv/vbae151 · Cited by 11 · see citing works
Chest CT Image based Lung Disease Classification – A Review
Kondamuri, S. R., Thadikemalla, V. S. G., Suryanarayana, G., Karthik, C., Reddy, V. S., Sahithi, V. B., Anitha, Y., Yogitha, V., & Valli, P. R. (2023) · Current Medical Imaging Formerly Current Medical Imaging Reviews · doi:10.2174/0115734056248176230923143105 · Cited by 11 · see citing works
Singh, B., Doborjeh, M., Doborjeh, Z., Budhraja, S., Tan, S. C. W., Sumich, A. L., Goh, W. W. B., Lee, J., Lai, E. M.-K., & Kasabov, N. K. (2023) · Scientific Reports · doi:10.1038/s41598-022-27132-8 · Cited by 10 · see citing works
Erdem, C., & Birtwistle, M. R. (2023) · Frontiers in Systems Biology · doi:10.3389/fsysb.2023.1099413 · Cited by 9 · see citing works
DeepBend: An interpretable model of DNA bendability
Khan, S. R., Sakib, S., Rahman, M. S., & Samee, M. A. H. (2023) · iScience · doi:10.1016/j.isci.2023.105945 · Cited by 9 · see citing works
Can Genetic Programming Perform Explainable Machine Learning for Bioinformatics?
Hu, T. (2020) · Genetic and evolutionary computation · doi:10.1007/978-3-030-39958-0_4 · Cited by 9 · see citing works
Artificial intelligence, molecular subtyping, biomarkers, and precision oncology
Shen, J. P. (2021) · Emerging Topics in Life Sciences · doi:10.1042/etls20210212 · Cited by 8 · see citing works
Anderson, P., Gadgil, R., Johnson, W. A., Schwab, E., & Davidson, J. M. (2021) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2021.104850 · Cited by 7 · see citing works
GenNet framework: interpretable neural networks for phenotype predictionPreprint
van Hilten, A., Kushner, S. A., Kayser, M., Ikram, M. A., Adams, H. H. H., Klaver, C. C. W., Niessen, W. J., & Roshchupkin, G. V. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.06.19.159152 · Cited by 7 · see citing works
Ni, J., Li, B., Miao, S., Zhang, X., Yan, D., Jing, S., Lu, S., Xie, Z., Zhang, X., & Liu, Y. (2025) · Briefings in Bioinformatics · doi:10.1093/bib/bbaf131 · Cited by 6 · see citing works
MOBILE pipeline enables identification of context-specific networks and regulatory mechanisms
Erdem, C., Gross, S. M., Heiser, L. M., & Birtwistle, M. R. (2023) · Nature Communications · doi:10.1038/s41467-023-39729-2 · Cited by 7 · see citing works
- Authors
- Yu, M. K., Ma, J., Fisher, J., Kreisberg, J. F., Raphael, B. J., & Ideker, T.
- Affiliations
- University of California San Diego (US)
- University of Cambridge (GB)
- Princeton University (US)
- Corresponding author
- Raphael, Benjamin J.
- Ideker, Trey
- Venue
- Cell
- Publisher
- Elsevier BV
- Published
- 2018-06
- Volume
- 173
- Issue
- 7
- Pages
- 1562-1565
- ISSN
- 0092-8674 (ISSN-L)
- PMC
- PMC6483071
- OpenAlex
- W2808686685
- Semantic Scholar
- 2dc18f661b400033abd1086b917c451d3358aef2
- Cited by
- 162
- References
- 18
- Citations by year
- 2026: 8 · 2025: 10 · 2024: 20 · 2023: 23 · 2022: 24 · 2021: 23 · 2020: 29 · 2019: 22 · 2018: 2 · 2014: 1
- FWCI
- 10.71
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- Locations
- Cell (publishedVersion)
- PubMed (publishedVersion)
- Europe PMC (PubMed Central) (acceptedVersion)
- PubMed Central (acceptedVersion)
- Funders
- National Institutes of Health (R01 HG009979, R01 HG007069, U24 CA211000, P41 GM103504, OT3 TR002026)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Physical Sciences › Computer Science › Computational Theory and Mathematics › Computational Drug Discovery Methods
- Keywords
- model interpretability, data heterogeneity
- MeSH terms
- Machine Learning, Algorithms, Computational Biology, Biomedical Research
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-10.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-10.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
