A systematic evaluation of explainable AI methods for high-dimensional transcriptome-based cancer survival prediction
Open accessAbstract
Explainable Artificial Intelligence (XAI) holds the promise to compensate for the “black-box” nature of deep learning which impedes transcriptome-based cancer survival prediction. However, there is a lack of systematic benchmarking XAI frameworks tailored for high-dimensional survival data. To bridge this gap, we systematically evaluated six representative XAI methods in three main categories: gradient-based, propagation-based, and perturbation-based approaches by using a Self-Normalizing Neural Network (SNN) as the baseline survival model. 6,248 samples across 15 cancer types from The Cancer Genome Atlas (TCGA) was analysed in this evaluation with a unified framework we developed. The evaluation metrics encompassed three key dimensions: prognostic factor enrichment (univariate Cox regression significance), biological consistency (supported by four authoritative databases, including OpenTargets), and explanation stability (Kuncheva Index). Among the six XAI methods, we find that DeepSHAP achieved the best overall performance, identifying the highest number of statistically significant prognostic factors while maintaining superior explanation stability; LRP (Layer-wise Relevance Propagation) showed slightly lower prognostic specificity but the highest consensus with biological databases in capturing general cancer genes, making it suitable for validating biological plausibility. In contrast, the perturbation-based method, PFI (Permutation Feature Importance) exhibited systematic failure and extremely low stability due to its inability to handle feature collinearity in high-dimensional transcriptomic data. Furthermore, we identified explanation stability as a robust proxy for the biological validity of the XAI. Collectively, t his study establishes an empirical framework for selecting trustworthy AI explanation tools for precision medicine.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
A pan-cancer benchmarking study across 15 cancer types found that explanation stability across cancer types remained a major unresolved challenge for DeepSHAP in transcriptomic survival models.
“explanation stability across cancer types remained a major unresolved challenge”
From A generative AI multi-agent framework with integrated XAI governance for cancer diagnostics: from multi-omics interpretation to lifestyle risk stratification (2026) · cites this paper as Zuo et al., 2026
The Results (§3.3; page unavailable) report that DeepSHAP was among the most stable methods in the cancer cohorts examined. The composite evaluation (§3.4; page unavailable) ranks it first in explanation stability, contrary to the claim that stability remained a major unresolved challenge for DeepSHAP. Stability was measured across validation folds within each cancer cohort, not as consistency of explanations between cancer types. (LLM, conf=0.86)
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredA Graph Neural Network paired with GNNExplainer identifies network subgraphs whose activation most strongly influences cancer risk prediction.
“a Graph Neural Network architecture is paired with GNNExplainer to identify the network subgraphs whose activation most strongly influences the cancer risk prediction”
From A generative AI multi-agent framework with integrated XAI governance for cancer diagnostics: from multi-omics interpretation to lifestyle risk stratification (2026) · cites this paper as Zuo et al., 2026
The Model construction section describes a self-normalizing neural network for cancer survival prediction, not a graph neural network. The Discussion says the study focused exclusively on that architecture and leaves graph neural networks for future investigation; it does not report pairing one with GNNExplainer to identify influential subgraphs. Page numbers were not available in our copy. (LLM, conf=0.90)
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Subramanian, A., Tamayo, P., Mootha, V. K., Mukherjee, S., Ebert, B. L., Gillette, M. A., Paulovich, A., Pomeroy, S. L., Golub, T. R., Lander, E. S., et al. (2005) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.0506580102 · Cited by 53,625 · see citing works
Bray, F., Laversanne, M., Sung, H., Ferlay, J., Siegel, R. L., Soerjomataram, I., & Jemal, A. (2024) · CA: a cancer journal for clinicians · doi:10.3322/caac.21834 · Cited by 27,064 · cited 1× · see citing works
“Cancer remains a leading cause of morbidity and mortality worldwide (Bray et al., 2024).”
Dormann, C. F., Elith, J., Bacher, S., Buchmann, C. M., Carl, G., Carré, G., Marquéz, J. R. G., Gruber, B., Lafourcade, B., Leitão, P. J., et al. (2012) · Ecography · doi:10.1111/j.1600-0587.2012.07348.x · Cited by 11,116 · see citing works
Self-Normalizing Neural NetworksPreprint
Klambauer, G., Unterthiner, T., Mayr, A., & Hochreiter, S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1706.02515 · Cited by 2,031 · see citing works
Explainable artificial intelligence (XAI) in deep learning-based medical image analysis
van der Velden, B. H. M., Kuijf, H. J., Gilhuijs, K. G. A., & Viergever, M. A. (2022) · Medical Image Analysis · doi:10.1016/j.media.2022.102470 · Cited by 1,328 · see citing works
Deep learning in cancer diagnosis, prognosis and treatment selection
Tran, K. A., Kondrashova, O., Bradley, A., Williams, E. D., Pearson, J. V., & Waddell, N. (2021) · Genome Medicine · doi:10.1186/s13073-021-00968-x · Cited by 932 · see citing works
Yang, G., Ye, Q., & Xia, J. (2022) · Information Fusion · doi:10.1016/j.inffus.2021.07.016 · Cited by 717 · see citing works
Emerging role of deep learning‐based artificial intelligence in tumor pathology
Jiang, Y., Yang, M., Wang, S., Li, X., & Sun, Y. K. (2020) · 癌症：英文版 · doi:10.1002/cac2.12012 · Cited by 445 · see citing works
Obtaining genetics insights from deep learning via explainable artificial intelligence
Novakovsky, G., Dexter, N., Libbrecht, M. W., Wasserman, W. W., & Mostafavi, S. (2022) · Nature Reviews Genetics · doi:10.1038/s41576-022-00532-2 · Cited by 387 · see citing works
Hooker, G., Mentch, L., & Zhou, S. (2021) · Statistics and Computing · doi:10.1007/s11222-021-10057-z · Cited by 264 · see citing works
Tun, H. M., Rahman, H. A., Naing, L., & Malik, O. A. (2025) · Journal of Medical Internet Research · doi:10.2196/69678 · Cited by 210 · see citing works
Associations of meningioma molecular subgroup and tumor recurrence
Youngblood, M. W., Miyagishima, D. F., Jin, L., Gupte, T. P., Li, C., Durán, D., Montejo, J. D., Zhao, A., Sheth, A. H., Tyrtova, E., et al. (2020) · Neuro-Oncology · doi:10.1093/neuonc/noaa226 · Cited by 167 · see citing works
Molnar, C., König, G., Bischl, B., & Casalicchio, G. (2023) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-022-00901-9 · Cited by 139 · see citing works
Towards multimodal foundation models in molecular cell biology
Cui, H., Tejada-Lapuerta, A., Brbić, M., Sáez-Rodríguez, J., Cristea, S., Goodarzi, H., Lotfollahi, M., Theis, F. J., & Wang, B. (2025) · Nature · doi:10.1038/s41586-025-08710-y · Cited by 131 · see citing works
Deep learning in cancer genomics and histopathology
Unger, M., & Kather, J. N. (2024) · Genome Medicine · doi:10.1186/s13073-024-01315-6 · Cited by 124 · see citing works
End-to-end prognostication in colorectal cancer by deep learning: a retrospective, multicentre study
Jiang, X., Hoffmeister, M., Brenner, H., Muti, H. S., Yuan, T., Foersch, S., West, N. P., Brobeil, A., Jonnagaddala, J., Hawkins, N., et al. (2024) · The Lancet Digital Health · doi:10.1016/s2589-7500(23)00208-x · Cited by 108 · see citing works
Explainable artificial intelligence for omics data: a systematic mapping study
Toussaint, P. A., Leiser, F., Thiebes, S., Schlesner, M., Brors, B., & Sunyaev, A. (2023) · Briefings in Bioinformatics · doi:10.1093/bib/bbad453 · Cited by 103 · see citing works
Multimodal Deep Learning-Based Prognostication in Glioma Patients: A Systematic Review
Alleman, K., Knecht, E. M., Huang, J., Zhang, L., Lam, S. K., & DeCuypere, M. G. (2023) · Cancers · doi:10.3390/cancers15020545 · Cited by 47 · see citing works
A historical perspective of biomedical explainable AI research
Malinverno, L., Barros, V., Ghisoni, F., Visonà, G., Kern, R., Nickel, P. J., Ventura, B. E., Šimić, I., Stryeck, S., Manni, F., et al. (2023) · Patterns · doi:10.1016/j.patter.2023.100830 · Cited by 42 · see citing works
Pancancer outcome prediction via a unified weakly supervised deep learning model
Yuan, W. Q., Chen, Y. M., Zhu, B., Yang, S., Zhang, J., Mao, N., Xiang, J., Ji, Y., Ji, Y., Luo, X., et al. (2025) · Signal Transduction and Targeted Therapy · doi:10.1038/s41392-025-02374-w · Cited by 40 · see citing works
Self-citationHou, J., Zhang, R., Xie, Y., Li, C., & Qin, W. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-02257-y · Cited by 38 · see citing works
Chen, Y., Cai, F., Cao, J., Gao, F., Lv, Y., Tang, Y., Zhang, A., Yan, W. X., Wang, Y., Hu, X., et al. (2021) · Frontiers in Endocrinology · doi:10.3389/fendo.2021.793337 · Cited by 28 · see citing works
Deep profiling of gene expression across 18 human cancers
Qiu, W., Dincer, A. B., Janizek, J. D., Çelik, S., Pittet, M. J., Naxerova, K., & Lee, S. (2024) · Nature Biomedical Engineering · doi:10.1038/s41551-024-01290-8 · Cited by 25 · see citing works
Vidya, S., Gupta, K., Aly, A., Wills, A. J., Ifeachor, E., & Shankar, R. (2025) · EClinicalMedicine · doi:10.1016/j.eclinm.2025.103452 · Cited by 23 · see citing works
Šimić, I., Veas, E. E., & Sabol, V. (2025) · Scientific Reports · doi:10.1038/s41598-025-09538-2 · Cited by 17 · see citing works
Self-Normalizing Multi-Omics Neural Network for Pan-Cancer Prognostication
Waqas, A., Tripathi, A. G., Ahmed, S., Mukund, A., Farooq, H., Johnson, J., Stewart, P. A., Naeini, M., Schabath, M. B., & Rasool, G. (2025) · International Journal of Molecular Sciences · doi:10.3390/ijms26157358 · Cited by 11 · see citing works
Zhang, X., Liu, J., Zhou, W., Lü, J., Wu, L., Li, Y., Wang, Y., Wang, Z., & Cai, J. (2025) · Translational Lung Cancer Research · doi:10.21037/tlcr-2025-187 · Cited by 5 · see citing works
Uncertainty Propagation in XAI: A Comparison of Analytical and Empirical Estimators
Chiaburu, T., Bießmann, F., & Haußer, F. (2025) · Communications in computer and information science · doi:10.1007/978-3-032-08333-3_18 · Cited by 5 · see citing works
Systematic Benchmarking of Local and Global Explainable AI Methods for Tabular Healthcare Data
Karagoz, G. N., Özçelebi, T., & Meratnia, N. (2025) · Communications in computer and information science · doi:10.1007/978-3-032-08317-3_16 · Cited by 4 · see citing works
Flagged first, then by citations
Research works printed but not linked: 5
- Fisher, A., Rudin, C., and Dominici, F. (2019). All models are wrong, but are useful: learning a variable's importance by studying an entire class of prediction models simultaneously. J. Mach. Learn. Res. 20, 1-81. Available online at: https://arxiv.org/abs/ 1801.01489.
- Lundberg, S. M., and Lee, S. I. (2017). A unified approach to interpreting model predictions. Adv. Neur In 30, 4768-4777. Available online at: https://arxiv.org/abs/1705. 07874.
- Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., et al. (2019). PyTorch: an imperative style, high-performance deep learning library. Adv. Neural Inf. Process. Syst. 32, 8026-8037. Available online at: https://arxiv.org/abs/1912.01703. · arXiv
- Shrikumar, A., Greenside, P., and Kundaje, A. (2017). Learning important features through propagating activation differences. Pr. Mach. Learn. Res. 70, 3145-3153. Available online at: https://arxiv.org/abs/1704.02685. · arXiv
- Pr. Mach. Learn. Res. 70, 3319-3328. Available online at: https://arxiv.org/abs/1703.01365. · arXiv
Cited by
Sorted by citations
Barki, C., Chouchen, M., Sediri, A., Ceylan, H. İ., Rahmouni, H. B., Muntean, R. I., El-Seedi, H. R., Bragazzi, N. L., & Dergaa, I. (2026) · Frontiers in Systems Biology · doi:10.3389/fsysb.2026.1892418 · Cited by 1 · see citing works
“A pan-cancer benchmarking study across 15 cancer types found that while DeepSHAP performed well for identifying prognostic biomarkers in transcriptomic survival models, explanation stability across cancer types remained a major unresolved challenge (Zuo et al., 2026).”
Yaşar, Ş., Yağın, F. H., Alzakari, S. A., Alkhalifa, A. K., Al-Hashem, F. H., & Tabnjh, A. K. (2026) · International Journal of Molecular Sciences · doi:10.3390/ijms27156989
- Authors
- Zuo, Y., Yang, S., & Zhao, W.
- Affiliations
- Shenzhen Campus of Sun Yat-sen University, Molecular Cancer Research Center, School of Medicine
- Sun Yat-sen University Sixth Affiliated Hospital, Department of Neurosurgery, Graceland Medical Center
- Sun Yat-sen University (CN)
- Corresponding author
- Zhao, Wenxue
- Venue
- Frontiers in Physiology
- Publisher
- Frontiers Media SA
- Published
- 2026-04-22
- Volume
- 17
- Pages
- 1830956-1830956
- ISSN
- 1664-042X (ISSN-L)
- OpenAlex
- W7155191053
- Cited by
- 2
- References
- 29
- Citations by year
- 2026: 2
- FWCI
- 6.79
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Frontiers in Physiology (publishedVersion, cc-by)
- PubMed (publishedVersion)
- PubMed Central (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Basic and Applied Basic Research Foundation of Guangdong Province
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene expression and cancer classification
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Keywords
- explainable AI, cancer survival prediction, transcriptome, high-dimensional survival data, DeepSHAP, layer-wise relevance propagation, permutation importance, explanation stability, prognostic factors, biological plausibility, self-normalizing neural networks, precision medicine
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
