Deep-learning model maps disease-related targets of human metabolites

A research team has developed DeepETD, an attention-based deep-learning model that predicts protein targets of endogenous metabolites by matching their disease associations, cellular phenotypes, and subcellular locations. The model achieved a validation accuracy of 82.49% and an area under the receiver operating characteristic curve (AUC-ROC) of 0.8470, outperforming several conventional machine-learning and deep-learning approaches. It also guided the experimental identification of three metabolite-protein interactions. By combining DeepETD with a freely accessible database covering 3,382 human endogenous metabolites, the work provides a practical resource for investigating metabolite functions, identifying disease-associated targets, and supporting phenotype-driven drug discovery and precision medicine.

Endogenous metabolites participate in nearly every fundamental biological process, serving not only as metabolic intermediates but also as signaling molecules that influence physiology and disease. Advances in metabolomics and chemical proteomics have uncovered unexpected functions and protein partners for many familiar metabolites. However, systematically identifying functional targets remains difficult because metabolite-protein binding affinities vary widely, biological effects depend on disease and cellular context, and experimental methods cannot always connect physical binding with a specific phenotype. Existing computational approaches often emphasize chemical structures or omics data while underusing information about diseases, phenotypes, and cellular locations. These limitations create a need for computational strategies that integrate multiple types of biomedical evidence to prioritize biologically meaningful metabolite targets.

A study (DOI: 10.48130/targetome-0026-0024) published in Targetome on 17 June 2026 by Hao Zhang's team, Shanghai University of Traditional Chinese Medicine, reports that integrating multidimensional biological fingerprints with an attention mechanism can accurately prioritize disease-related protein targets and reveal previously unverified metabolite-protein interactions.

To construct DeepETD, the researchers assembled 15,872 metabolite-protein pairs from the Human Metabolome Database, the IUPHAR/BPS Guide to PHARMACOLOGY, ChEMBL, and BindingDB. Pairs with half-maximal inhibitory concentration (IC50) values of 100 nmol·L−1 or below were treated as positive samples, producing 2,565 positive and 13,307 negative examples. The team mined PubMed abstracts and incorporated Disease Ontology and Human Phenotype Ontology data to generate fingerprints describing each metabolite and protein through associated diseases, cellular phenotypes, and subcellular locations. An attention layer assigned different weights to these features before a deep neural network estimated interaction probabilities. The dataset was divided 80:20 for training and validation, while ten-fold cross-validation, ablation tests, and comparisons with XGBoost, CatBoost, DrugBAN, TransformerCPI, and MGNDTI assessed performance. DeepETD produced training and validation AUC-ROC values of 0.9771 and 0.8470, respectively, with corresponding accuracies of 92.74% and 82.49%. Ten-fold cross-validation yielded a mean AUC of 0.8484 and mean accuracy of 0.8344, indicating stable performance. Removing the attention mechanism significantly reduced accuracy, while eliminating disease-related features caused the greatest decline among the three fingerprint dimensions. In global ranking tests, the top 10 predictions reached a mean precision of 0.92 and represented a 5.73-fold enrichment over the validation set’s baseline positive rate. The model also ranked known targets of dopamine and estradiol highly within Parkinson’s disease and breast cancer contexts. The researchers then applied DeepETD to 3,382 metabolites across ten disease contexts, generating 33,820 records for the Endogenous Metabolites Target Discovery Database (EMTDD). Microscale thermophoresis experiments tested predicted targets for testosterone and leukotriene B4. The assays supported direct binding between testosterone and NR2F6, as well as between leukotriene B4 and PRMT2 and ACAT1; molecular docking provided compatible binding models. Not every tested prediction was experimentally confirmed, underscoring the need for further validation.

Overall, DeepETD offers a computational complement to chemical proteomics by transforming dispersed biomedical knowledge into disease-aware target predictions. Its combination of interpretable feature weighting, competitive predictive performance, experimental testing, and an open database may help researchers prioritize targets before undertaking resource-intensive laboratory studies. The approach nevertheless depends on the availability and quality of existing biological annotations, and its fixed affinity threshold and imbalanced training data may introduce uncertainty.

###

References

DOI

10.48130/targetome-0026-0024

Original Source URL

https://doi.org/10.48130/targetome-0026-0024

Funding information

We gratefully acknowledge the financial supports from the National Key R&D Program of China (2022YFC3400500 to Cheng Luo), the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB0830301 to Cheng Luo), the National Natural Science Foundation of China (81903538 to Hao Zhang), the Science and Technology Department of Guizhou Province (CXPTXM[2025]021 and KXJZ[2025]014 to Cheng Luo), the Applied Basic Research Foundation of Yunnan Province (202501BC070005 to Cheng Luo), the Shanghai Municipal Health Commission Medical New Technology Project (2025ZZ2060 to Hao Zhang), the Shanghai Oriental Talent Plan Youth Project (QNJY2025170 to Hao Zhang), the Shanghai Science and Technology Committee Computational Biology Special Project (YDZX20233100004032 to Cheng Luo), and (24JS2830200 to Heng Xu), the Lingang Laboratory (LG-QS-202204-07 to Hao Zhang), and the Shanghai Municipal Education Commission AI for Science Project (301-0406 to Hao Zhang).

About Targetome

Targetome refers to the complete collection of molecular targets (e.g., proteins, RNA or DNA) that interact with and mediate the effect of a specific biomolecule, such as a drug, toxin, metabolites, transcription factor or microRNA, within a biological system. Targetome is an open access journal publishing rigorously peer-reviewed original research articles, reviews, break-through methods, and perspectives that advance our understanding, identification and validation of molecular targets for new drug development.