{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,30]],"date-time":"2026-10-30T03:52:20Z","timestamp":1793332340187,"version":"4.5.0"},"reference-count":55,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,2,26]],"date-time":"2026-02-26T00:00:00Z","timestamp":1772064000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Pharmacol."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Post-marketing safety surveillance of Chinese Materia Medica (CMM) is challenged by multi-component chemical heterogeneity and the limited mechanistic interpretability of signals derived solely from spontaneous reports. The FDA Adverse Event Reporting System (FAERS) provides large-scale pharmacovigilance evidence, yet it is noisy, susceptible to reporting bias, and weakly linked to underlying biological mechanisms. We aimed to develop an FAERS-informed, clinically oriented framework to predict CMM-associated adverse drug reactions (ADRs).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We constructed an evidence-rich heterogeneous graph integrating CMMs, compounds, protein targets, and ADRs. To differentiate pharmacovigilance-derived statistical associations from binary molecular interactions, we augmented each CMM\u2013ADR edge with a six-dimensional evidence feature vector (including semantic similarity, FAERS evidence as log-transformed report counts, source provenance, and topology-derived structural metrics) and used it to condition attention during message passing. We propose MSAT, a multi-scale heterogeneous graph neural network comprising: (i) an Evidence-Semantic Adaptive Gate to inject evidence-conditioned attention bias, (ii) a Hierarchical Signal Propagation layer to model cross-scale transduction from molecular mechanisms to clinical phenotypes, and (iii) a Hub-Calibrated Inference module to mitigate hub-driven bias. We evaluated MSAT using stratified 10-fold cross-validation, stress-tested robustness under increasing class imbalance up to a 1:10 positive:negative ratio, and assessed cold-start generalization. High-confidence predicted results were further examined via external database concordance and literature support.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      In stratified 10-fold cross-validation on 27,062 curated CMM\u2013ADR associations, MSAT achieved strong performance (AUC = 0.9792, AUPRC = 0.9766) and outperformed representative heterogeneous GNN baselines. MSAT remained robust under severe class imbalance (up to 1:10) and demonstrated favorable generalization in cold-start settings. Among the top 15 high-confidence predicted results absent from the labeled positives, 13\/15 (86.7%) were supported by independent database or literature evidence. For example, MSAT prioritized a potential liver-injury signal for Aiye (\n                      <jats:italic>Artemisia argyi<\/jats:italic>\n                      ) (predicted ADR: drug-induced liver injury, DILI), consistent with external evidence.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>By unifying FAERS pharmacovigilance evidence with multi-scale biomedical mechanisms in a heterogeneous graph learning framework, MSAT enables robust prediction and prioritization of CMM-associated ADR risks. This framework can support hypothesis generation and risk triage for post-marketing safety surveillance of complex Chinese Materia Medica products.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fphar.2026.1774128","type":"journal-article","created":{"date-parts":[[2026,2,26]],"date-time":"2026-02-26T06:55:52Z","timestamp":1772088952000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["MSAT: a FAERS-informed heterogeneous graph neural network for pharmacovigilance prediction of Chinese materia medica\u2013associated adverse drug reactions"],"prefix":"10.3389","volume":"17","author":[{"given":"Bowen","family":"Shi","sequence":"first","affiliation":[{"name":"School of Medical Information and Engineering, Guangdong Pharmaceutical University","place":["Guangzhou, Guangdong, China"]},{"name":"Guangdong Province Precise Medicine Big Data of Traditional Chinese Medicine Engineering Technology Research Center","place":["Guangzhou, Guangdong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaojie","family":"Li","sequence":"additional","affiliation":[{"name":"Guangdong Province Precise Medicine Big Data of Traditional Chinese Medicine Engineering Technology Research Center","place":["Guangzhou, Guangdong, China"]},{"name":"Key Specialty of Clinical Pharmacy, The First Affiliated Hospital of Guangdong Pharmaceutical University","place":["Guangdong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinghao","family":"Fang","sequence":"additional","affiliation":[{"name":"School of Medical Information and Engineering, Guangdong Pharmaceutical University","place":["Guangzhou, Guangdong, China"]},{"name":"Guangdong Province Precise Medicine Big Data of Traditional Chinese Medicine Engineering Technology Research Center","place":["Guangzhou, Guangdong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jisheng","family":"Chen","sequence":"additional","affiliation":[{"name":"Key Specialty of Clinical Pharmacy, The First Affiliated Hospital of Guangdong Pharmaceutical University","place":["Guangdong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jin","family":"Yang","sequence":"additional","affiliation":[{"name":"School of Medical Information and Engineering, Guangdong Pharmaceutical University","place":["Guangzhou, Guangdong, China"]},{"name":"Guangdong Province Precise Medicine Big Data of Traditional Chinese Medicine Engineering Technology Research Center","place":["Guangzhou, Guangdong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,2,26]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"109","DOI":"10.2165\/00002018-199920020-00002","article-title":"The medical dictionary for regulatory activities (MedDRA)","volume":"20","author":"Brown","year":"1999","journal-title":"Drug Saf."},{"key":"B2","doi-asserted-by":"publisher","first-page":"D907","DOI":"10.1093\/nar\/gku1066","article-title":"ADReCS: an ontology database for aiding standardization and hierarchical classification of adverse drug reaction terms","volume":"43","author":"Cai","year":"2015","journal-title":"Nucleic Acids Res."},{"key":"B3","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1038\/s41597-023-01960-3","article-title":"Building a knowledge graph to enable precision medicine","volume":"10","author":"Chandak","year":"2023","journal-title":"Sci. Data"},{"key":"B4","doi-asserted-by":"crossref","first-page":"785","DOI":"10.1145\/2939672.2939785","article-title":"XGBoost: a scalable tree boosting system","volume-title":"Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining","author":"Chen","year":"2016"},{"key":"B5","doi-asserted-by":"publisher","first-page":"183","DOI":"10.1016\/j.jchromb.2015.08.024","article-title":"Yuanhuapine-induced intestinal and hepatotoxicity were correlated with disturbance of amino acids, lipids, carbohydrate metabolism and gut microflora function: a rat urine metabonomic study","volume":"1026","author":"Chen","year":"2016","journal-title":"J. Chromatogr. B"},{"key":"B6","first-page":"1725","article-title":"Simple and deep graph convolutional networks","volume-title":"Proceedings of the 37th international conference on machine learning (ICML 2020). vol. 119 of proceedings of machine learning research","author":"Chen","year":"2020"},{"key":"B7","volume-title":"Pharmacopoeia of the people\u2019s Republic of China, 2025 edition","year":"2025"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2010.09885","article-title":"ChemBERTa: large-scale self-supervised pretraining for molecular property prediction","author":"Chithrananda","year":"2020","journal-title":"arXiv preprint ArXiv:2010.09885"},{"key":"B9","first-page":"1","article-title":"HES: edge sampling for heterogeneous graphs","volume-title":"Proceedings of the 2023 international joint conference on neural networks (IJCNN) (IEEE)","author":"Fang","year":"2023"},{"key":"B10","doi-asserted-by":"publisher","first-page":"baab068","DOI":"10.1093\/database\/baab068","article-title":"Mining drug\u2013target and drug\u2013adverse drug reaction databases to identify target\u2013adverse drug reaction relationships","volume":"2021","author":"Galletti","year":"2021","journal-title":"Database"},{"key":"B11","doi-asserted-by":"publisher","first-page":"74","DOI":"10.1016\/j.eng.2018.11.019","article-title":"Safety research in traditional Chinese medicine\u2014review: methods, applications, and outlook","volume":"5","author":"Gao","year":"2019","journal-title":"Engineering"},{"key":"B12","first-page":"407","article-title":"The research of oxidative damage mechanism in hepatical toxical injury caused by different components from folium artemisiae argyi in mice","volume":"8","author":"Gong","year":"2011","journal-title":"Chin. J. Pharmacovigil."},{"key":"B13","doi-asserted-by":"publisher","first-page":"1010","DOI":"10.1038\/clpt.2012.50","article-title":"Novel data-mining methodologies for adverse drug event discovery and analysis","volume":"91","author":"Harpaz","year":"2012","journal-title":"Clin. Pharmacol. Ther."},{"key":"B14","first-page":"2704","article-title":"Heterogeneous graph transformer","volume-title":"Proceedings of the web conference (WWW)","author":"Hu","year":"2020"},{"key":"B15","doi-asserted-by":"publisher","first-page":"235","DOI":"10.1163\/156856005774423827","article-title":"Modulation of gastric hemorrhage and ulceration by oxidative stress and histamine release in Salmonella typhimurium-infected rats","volume":"13","author":"Hung","year":"2005","journal-title":"Inflammopharmacology"},{"key":"B16","doi-asserted-by":"publisher","first-page":"339","DOI":"10.1007\/s10787-998-0017-1","article-title":"Roles of histamine receptors and oxyradicals in aggravation of acid-induced gastric haemorrhagic ulcers in endotoxaemic rats","volume":"6","author":"Hung","year":"1998","journal-title":"Inflammopharmacology"},{"key":"B17","article-title":"Semi-supervised classification with graph convolutional networks","volume-title":"5th international conference on learning representations, ICLR 2017","author":"Kipf","year":"2017"},{"key":"B18","doi-asserted-by":"publisher","first-page":"4053","DOI":"10.7501\/j.issn.0253-2670.2023.12.031","article-title":"Research progress on chemical components, pharmacological effects and clinical application of curcuma zedoary oil injection","volume":"54","author":"Kong","year":"2023","journal-title":"Chin. Tradit. Herb. Drugs"},{"key":"B19","doi-asserted-by":"publisher","first-page":"2395","DOI":"10.1161\/CIRCULATIONAHA.106.682658","article-title":"Logistic regression","volume":"117","author":"LaValley","year":"2008","journal-title":"Circulation"},{"key":"B20","doi-asserted-by":"publisher","first-page":"1234","DOI":"10.1093\/bioinformatics\/btz682","article-title":"BioBERT: a pre-trained biomedical language representation model for biomedical text mining","volume":"36","author":"Lee","year":"2020","journal-title":"Bioinformatics"},{"key":"B21","volume-title":"Common drug adverse reactions and treatment: Chinese medicine volume","author":"Lin","year":"2013"},{"key":"B22","doi-asserted-by":"publisher","first-page":"491","DOI":"10.1186\/s12906-017-1997-4","article-title":"Molecular network-based analysis of the mechanism of liver injury induced by volatile oils from Artemisiae argyi folium","volume":"17","author":"Liu","year":"2017","journal-title":"BMC Complement. Altern. Med."},{"key":"B23","first-page":"348","article-title":"Analysis of clinical application and adverse reactions of Polygonum cuspidatum based on ancient and modern literature of traditional Chinese medicine","volume":"15","author":"Liu","year":"2018","journal-title":"Chin. J. Pharmacovigil."},{"key":"B24","doi-asserted-by":"publisher","first-page":"883","DOI":"10.1007\/s11655-012-1290-6","article-title":"Bridging the traditional Chinese medicine pattern classification and biomedical disease diagnosis with systems biology","volume":"18","author":"Lu","year":"2012","journal-title":"Chin. J. Integr. Med."},{"key":"B25","doi-asserted-by":"crossref","first-page":"1150","DOI":"10.1145\/3447548.3467350","article-title":"Are we really making much progress? Revisiting, benchmarking and refining heterogeneous graph neural networks","volume-title":"Proceedings of the 27th ACM SIGKDD conference on knowledge discovery and data mining","author":"Lv","year":"2021"},{"key":"B26","doi-asserted-by":"publisher","first-page":"10684","DOI":"10.1039\/D3SC02139D","article-title":"TCMBank: bridges between the largest herbal medicines, chemical ingredients, target proteins, and associated diseases with intelligence text mining","volume":"14","author":"Lv","year":"2023","journal-title":"Chem. Sci."},{"key":"B27","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1186\/s13020-023-00741-9","article-title":"Machine learning in TCM with natural products and molecules: current status and future perspectives","volume":"18","author":"Ma","year":"2023","journal-title":"Chin. Med."},{"key":"B28","doi-asserted-by":"publisher","first-page":"24","DOI":"10.4103\/0971-5851.151771","article-title":"Harmful effects of nicotine","volume":"36","author":"Mishra","year":"2015","journal-title":"Indian J. Med. Paediatr. Oncol."},{"key":"B29","doi-asserted-by":"publisher","first-page":"D1353","DOI":"10.1093\/nar\/gkac1046","article-title":"The next-generation open targets platform: reimagined, redesigned, rebuilt","volume":"51","author":"Ochoa","year":"2023","journal-title":"Nucleic Acids Res."},{"key":"B30","first-page":"191","article-title":"Experimental studies on developmental mechanism of acute hemorrhagic gastric ulcers with emphasis on the effects of histamine","volume":"30","author":"Ogawa","year":"1981","journal-title":"Hiroshima J. Med. Sci."},{"key":"B31","first-page":"287","article-title":"Efficacy and safety of black cohosh (Actaea\/Cimicifuga racemosa) in the treatment of vasomotor symptoms: review of clinical trials","volume":"79","author":"Oleszczuk","year":"2008","journal-title":"Ginekol. Pol."},{"key":"B32","doi-asserted-by":"publisher","first-page":"2090","DOI":"10.1016\/j.bmc.2010.12.043","article-title":"Structure\u2013activity relationships of flavonoids as inhibitors of breast cancer resistance protein (bcrp)","volume":"19","author":"Pick","year":"2011","journal-title":"Bioorg. Med. Chem."},{"key":"B33","doi-asserted-by":"publisher","first-page":"9368","DOI":"10.1609\/aaai.v35i11.17129","article-title":"Relation-aware graph attention model with adaptive self-adversarial training","volume":"35","author":"Qin","year":"2021","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"B34","doi-asserted-by":"publisher","first-page":"31","DOI":"10.17849\/insm-47-01-31-39.1","article-title":"Random forest","volume":"47","author":"Rigatti","year":"2017","journal-title":"J. Insur Med."},{"key":"B35","first-page":"593","article-title":"Modeling relational data with graph convolutional networks","volume-title":"Proceedings of the European Semantic Web Conference (ESWC)","author":"Schlichtkrull","year":"2018"},{"key":"B36","unstructured":"Traditional Chinese medicine toxic side effects database, chemistry database\n          \n          \n          2024"},{"key":"B37","first-page":"79","article-title":"Review of salicylate-induced hearing loss, neurotoxicity, tinnitus and neuropathophysiology","volume":"34","author":"Sheppard","year":"2014","journal-title":"Acta Otorhinolaryngol. Ital."},{"key":"B38","first-page":"103","article-title":"Formaldehyde toxicity\u2014part II. Reviews of acute and chronic effects on health","volume":"66","author":"Solomons","year":"1984","journal-title":"S Afr. Med. J."},{"key":"B39","doi-asserted-by":"publisher","first-page":"1","DOI":"10.13313\/j.issn.1673-4890.20220325004","article-title":"Pharmacological effects and attenuating toxicity of dichroae radix and febrifugine","author":"Sun","year":"2022","journal-title":"Mod. Chin. Med."},{"key":"B40","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/CIBCB49929.2021.9562802","article-title":"Predicting drug-drug interactions using meta-path based similarities","volume-title":"2021 IEEE conference on computational intelligence in bioinformatics and computational biology (CIBCB)","author":"Tanvir","year":"2021"},{"key":"B41","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2207.05672","article-title":"Ddi prediction via heterogeneous graph attention networks","author":"Tanvir","year":"2022"},{"key":"B42","first-page":"208","article-title":"HeTAN: heterogeneous graph triplet attention network for drug repurposing","volume-title":"2024 IEEE 11th international conference on data science and advanced analytics (DSAA)","author":"Tanvir","year":""},{"key":"B43","doi-asserted-by":"crossref","DOI":"10.1109\/TCBB.2024.3417715","volume-title":"DDI prediction with heterogeneous information network - meta-path based approach","author":"Tanvir","year":""},{"key":"B44","doi-asserted-by":"publisher","first-page":"476","DOI":"10.1093\/toxres\/tfab026","article-title":"Epidermal growth factor receptor inhibitor-induced diarrhea: clinical incidence, toxicological mechanism, and management","volume":"10","author":"Tao","year":"2021","journal-title":"Toxicol. Res."},{"key":"B45","doi-asserted-by":"publisher","first-page":"D939","DOI":"10.1093\/nar\/gkaa980","article-title":"Genenames.org: the HGNC and VGNC resources in 2021","volume":"49","author":"Tweedie","year":"2021","journal-title":"Nucleic Acids Res."},{"key":"B46","volume-title":"Huang Di Nei Jing Su Wen: an Annotated Translation of Huang Di\u2019s Inner Classic\u2014Basic Questions","author":"Unschuld","year":"2011"},{"key":"B47","unstructured":"FDA adverse event reporting system (FAERS)\n          \n          \n          2024"},{"key":"B48","article-title":"Graph attention networks","volume-title":"6th international conference on learning representations, ICLR 2018","author":"Veli\u010dkovi\u0107","year":"2018"},{"key":"B49","doi-asserted-by":"publisher","first-page":"119","DOI":"10.1007\/bf02966882","article-title":"On the use of salicylic acid and the salicylates","volume":"68","author":"Woods","year":"1879","journal-title":"Dublin J. Med. Sci."},{"key":"B50","doi-asserted-by":"publisher","first-page":"D976","DOI":"10.1093\/nar\/gky987","article-title":"ETCM: an encyclopaedia of traditional Chinese medicine","volume":"47","author":"Xu","year":"2019","journal-title":"Nucleic Acids Res."},{"key":"B51","doi-asserted-by":"publisher","first-page":"761097","DOI":"10.3389\/fphar.2022.761097","article-title":"Safety of using traditional Chinese medicine injections in primary medical institutions: based on the spontaneous reporting system 2016\u20132020 in Henan province, China","volume":"13","author":"Yan","year":"2022","journal-title":"Front. Pharmacol."},{"key":"B52","doi-asserted-by":"publisher","first-page":"5807","DOI":"10.1016\/j.csbj.2023.11.030","article-title":"ccTCM: a quantitative component and compound platform for promoting the research of traditional Chinese medicine","volume":"21","author":"Yang","year":"2023","journal-title":"Comput. Struct. Biotechnol. J."},{"key":"B53","first-page":"793","article-title":"Heterogeneous graph neural network","volume-title":"Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery and data mining","author":"Zhang","year":"2019"},{"key":"B54","doi-asserted-by":"publisher","first-page":"bbad518","DOI":"10.1093\/bib\/bbad518","article-title":"Network pharmacology: towards the artificial intelligence-based precision traditional Chinese medicine","volume":"25","author":"Zhang","year":"2024","journal-title":"Brief. Bioinform"},{"key":"B55","doi-asserted-by":"publisher","first-page":"i457","DOI":"10.1093\/bioinformatics\/bty294","article-title":"Modeling polypharmacy side effects with graph convolutional networks","volume":"34","author":"Zitnik","year":"2018","journal-title":"Bioinformatics"}],"container-title":["Frontiers in Pharmacology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fphar.2026.1774128\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,26]],"date-time":"2026-02-26T06:55:59Z","timestamp":1772088959000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fphar.2026.1774128\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,26]]},"references-count":55,"alternative-id":["10.3389\/fphar.2026.1774128"],"URL":"https:\/\/doi.org\/10.3389\/fphar.2026.1774128","relation":{},"ISSN":["1663-9812"],"issn-type":[{"value":"1663-9812","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,26]]},"article-number":"1774128"}}