Predicting Therapeutic Outcome via Aligning Patient-Specific Knowledge Graph and Gene-Level Perturbation Representations
2026-07-06 • Machine Learning
Machine LearningArtificial Intelligence
AI summaryⓘ
The authors developed PREDIKTOR, a new method to better predict how individual patients will respond to cancer drugs using their tumor gene data before treatment. They combined personalized gene networks with drug effect simulations to create a more accurate and interpretable prediction. Their approach outperformed existing methods on several tests and worked well even on new clinical trial data without extra training. This method also helps identify important genes and pathways involved in drug response, which could guide personalized cancer treatment.
gene regulatory networktranscriptomicsdrug response predictiontransfer learninggraph neural networkcontrastive learningprecision oncologyTCGALINCS L1000I-SPY2 trial
Authors
Dongmin Bang, Sugyun An, Inyoung Sung, Ilho Yun, Sun Kim, Sangseon Lee
Abstract
Accurate prediction of patient-specific therapeutic response from pre-treatment transcriptomes is hindered by the scarcity of matched clinical response labels and post-treatment molecular profiles. Preclinical transfer-learning models can simulate drug-induced expression changes but are often hard to interpret and unstable, whereas knowledge-graph methods provide mechanistic context yet remain static and fail to capture drug-induced transcriptomic perturbation dynamics. We propose PREDIKTOR, a patient-centered multi-view framework that aligns a personalized network view with a transferable transcriptomic perturbation view to predict clinical drug response. For each patient, we construct an individualized gene regulatory network from tumor expression using DysRegNet and augment it with drug-target links from DrugBank; a graph neural encoder yields a drug-centric, mechanistically grounded embedding. In parallel, a frozen condition-specific gene-gene attention model pretrained on LINCS L1000 generates a simulated post-perturbation transcriptomic profile for the same patient-drug pair. We align the two views in a shared latent space via a CLIP-style contrastive objective with drug-context hard negatives, then concatenate the representations for end-to-end response classification. On TCGA, PREDIKTOR consistently outperforms state-of-the-art baselines under patient-, drug-, and tissue-split evaluations, and transfers zero-shot to the I-SPY2 trial, improving AUROC by 5.6% over competing methods. The aligned embeddings yield stable gene and pathway attributions that recover known mechanisms, supporting actionable and interpretable precision oncology.