Disease relevance and replicability of deep learning gene expression prediction
- 1SUNY Upstate
Abstract
Recent deep learning (DL) models predict average gene expression levels from DNA sequences with high overall correlation to measured values. We examine these DL models through the lens of disease research. The seemingly high overall performance of DL models is largely due to capturing whether genes are "On" or "Off", and to a lesser extent, disease-relevant expression level changes for "On" genes. Indeed, the more bimodal the gene expression distribution, the better the reported performance. We track the extent of this issue across tissues, molecular systems, and cancers. Compounding the model evaluation issue, we find inconsistencies between the published code and reported performance, highlighting the importance of versioning and publishing performance evaluation code. These findings indicate that the high reported performance of popular DL models falls unexpectedly short in disease applications, and that the problem of personalized genomic prediction remains far from solved in a disease context.
Related
- Integrating Multisource Target Evidence with Machine-Learning Potency Prediction for Alzheimer's Disease Drug Discovery
- The relevance of GGCC methylation for cyanobacterial gene expression regulation
- MarkerScout: A Disease-Agnostic Machine Learning Framework for Biomarker Prediction from Multi-Scale Mechanistic Models
- Heterogeneous Graph Contrastive Learning for Drug-Gene-Disease Motif Prediction
- A Curated Phenotype-Linked HIV-1 Sequence Dataset for Machine Learning Approaches to Tropism Prediction, Disease Progression and Proviral Reservoir Studies.
BibTeX
@misc{zhang2026disease,
title = {{Disease relevance and replicability of deep learning gene expression prediction}},
author = {A. Zhang and S. Tasaki and D. Connell and B. Ng and C. Gaiteri},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.24.753534},
url = {https://www.biorxiv.org/content/10.64898/2026.09.24.753534v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Disease relevance and replicability of deep learning gene expression prediction
AU - A. Zhang
AU - S. Tasaki
AU - D. Connell
AU - B. Ng
AU - C. Gaiteri
PY - 2026
DA - 2026/09/24
PB - bioRxiv
DO - 10.64898/2026.09.24.753534
UR - https://www.biorxiv.org/content/10.64898/2026.09.24.753534v1
AB - Recent deep learning (DL) models predict average gene expression levels from DNA sequences with high overall correlation to measured values. We examine these DL models through the lens of disease research. The seemingly high overall performance of DL models is largely due to capturing whether genes are "On" or "Off", and to a lesser extent, disease-relevant expression level changes for "On" genes. Indeed, the more bimodal the gene expression distribution, the better the reported performance. We track the extent of this issue across tissues, molecular systems, and cancers. Compounding the model evaluation issue, we find inconsistencies between the published code and reported performance, highlighting the importance of versioning and publishing performance evaluation code. These findings indicate that the high reported performance of popular DL models falls unexpectedly short in disease applications, and that the problem of personalized genomic prediction remains far from solved in a disease context.
ER -