Dokument: Drug response prediction: A critical systematic review of current datasets and methods

Titel:	Drug response prediction: A critical systematic review of current datasets and methods
URL für Lesezeichen:	https://docserv.uni-duesseldorf.de/servlets/DocumentServlet?id=71896
URN (NBN):	urn:nbn:de:hbz:061-20260115-104003-5
Kollektion:	Publikationen
Sprache:	Englisch
Dokumententyp:	Wissenschaftliche Texte » Artikel, Aufsatz
Medientyp:	Text
Autoren:	Tran, Nguyen Khoa [Autor] Klau, Gunnar [Autor]
Dateien:	[Dateien anzeigen] Adobe PDF [Details] 755,4 KB in einer Datei [ZIP-Datei erzeugen] Dateien vom 15.01.2026 / geändert 15.01.2026
Stichwörter:	Molecular graphs , Drug response prediction , Multi-omics , Multi-output regression , Graph neural network , Multilayer perceptron
Beschreibung:	Predicting drug response is a critical task in personalized medicine. Several recent studies have reported promising improvements in predictive performance with deep learning models trained on molecular characterizations of cell lines and drugs. However, our baseline tests suggest that little to no meaningful biological or chemical information is being learned from multi-omics data in the publicly available large-scale datasets GDSC and DepMap Public or molecular graphs, respectively. In our experiments, even gene expression data, commonly regarded as highly predictive, failed to deliver satisfactory drug response predictions. This raises the possibility that drug response measures or patterns observed in multi-omics data may not arise from underlying biological mechanisms. To investigate this, we identified and examined inconsistencies within and across the GDSC2 and DepMap Public 24Q2 datasets. We found that IC50 and AUC values of replicated experiments in GDSC2 had an average Pearson correlation coefficient of only 0.563±0.230 and 0.468±0.358, respectively. Additionally, somatic mutations shared between cell lines in the two datasets showed a Pearson correlation coefficient of only 0.180. Even in cases where TGSA, the current best-performing method to our knowledge, exceeded baseline performance, it still did not surpass a simple baseline multi-output multilayer perceptron (MMLP). Moreover, MMLP is not only more easily adaptable to new datasets but also significantly faster, making it a viable baseline for comparisons. In conclusion, our findings suggest that current cell-line and drug data are insufficient for existing modeling approaches to effectively uncover the biological and chemical mechanisms underlying drug response. Therefore, improving data quality or focusing on different data types is crucial before proposing novel methods.
Rechtliche Vermerke:	Originalveröffentlichung: Tran, N. K., & Klau, G. (2025). Drug response prediction: A critical systematic review of current datasets and methods. Pattern Recognition Letters, 199, 21–26. https://doi.org/10.1016/j.patrec.2025.10.016
Lizenz:	Dieses Werk ist lizenziert unter einer Creative Commons Namensnennung 4.0 International Lizenz
Fachbereich / Einrichtung:	Mathematisch- Naturwissenschaftliche Fakultät
Dokument erstellt am:	15.01.2026
Dateien geändert am:	15.01.2026