Training distribution determines the ceiling of drug-blind cancer sensitivity prediction

• URL: http://arxiv.org/abs/2605.20885v1
• Date: Wed, 20 May 2026 08:24:56 GMT
• Title: Training distribution determines the ceiling of drug-blind cancer sensitivity prediction
• Abstract summary: Precision oncology requires predicting which drugs will suppress a specific tumor from its molecular profile.<n>We show that drug-blind sensitivity prediction has plateaued despite increasingly complex drug representations.
• Score: 0.304585143845864
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Precision oncology requires predicting which drugs will suppress a specific tumor from its molecular profile, but drug-blind sensitivity prediction has plateaued despite increasingly complex drug representations. Here we show that this stagnation reflects a metric artifact rather than a representational bottleneck. The standard benchmark, global Pearson r, is dominated by between-drug potency differences that a trivial drug-mean predictor captures without any cell-specific learning. Per-drug Pearson r, which isolates within-drug cell ranking, reveals that no drug encoding improves over cell-only features across four independent datasets. A controlled experiment channeling mechanism-of-action identity as either a drug feature or a training-distribution constraint identifies the cause. Supplying MoA as a feature yields negligible benefit, whereas using it to stratify training raises per-drug r substantially for targeted kinase inhibitors, because pan-cancer co-training suppresses pathway-specific sensitivity signals. Mechanism-stratified training and response matching from pilot observations provide two deployable strategies that together recover the principal sources of predictive gain in drug-blind sensitivity prediction.

Related papers

• Conformal Prediction for Molecular Properties under Label Shift [2.2400646874013037]
  We develop a conformal prediction framework tailored to label shift.<n>By weighting conformal scores using marginal label probability ratios, our method produces statistically rigorous prediction intervals without retraining.<n>By moving beyond accuracy alone to provide actionable confidence measures, our approach enhances the trustworthiness of AI-driven predictions.
  arXiv  Detail & Related papers  (2026-08-18T11:50:38Z)
• MARD: Mirror-Augmented Reasoning Distillation for Mechanism-Level Drug-Drug Interaction Prediction [0.0]
  Mechanism-level drug-drug interaction (DDI) prediction requires identifying which enzyme or pharmacodynamic axis is implicated.<n>We introduce a reproducible mechanism-level DDI labelling and evaluation protocol with a structured 7-family/147-subtype taxonomy.<n>We propose a pipeline that produces a 7B reasoning MARD (Mirror-Augmented Reasoning Distillation)
  arXiv  Detail & Related papers  (2026-06-10T18:26:11Z)
• LaPro-DTA: Latent Dual-View Drug Representations and Salient Protein Feature Extraction for Generalizable Drug--Target Affinity Prediction [6.065808462922399]
  Existing methods suffer from significant performance degradation in realistic cold-start scenarios.<n>We propose LaPro-DTA, a framework designed to achieve robust and generalizable DTA prediction.
  arXiv  Detail & Related papers  (2026-03-16T03:46:51Z)
• The Eminence in Shadow: Exploiting Feature Boundary Ambiguity for Robust Backdoor Attacks [51.468144272905135]
  Deep neural networks (DNNs) underpin critical applications yet remain vulnerable to backdoor attacks.<n>We provide a theoretical analysis targeting backdoor attacks, focusing on how sparse decision boundaries enable disproportionate model manipulation.<n>We propose Eminence, an explainable and robust black-box backdoor framework with provable theoretical guarantees and inherent stealth properties.
  arXiv  Detail & Related papers  (2025-12-11T08:09:07Z)
• Dual-Pathway Fusion of EHRs and Knowledge Graphs for Predicting Unseen Drug-Drug Interactions [5.481037938702276]
  Drug-drug interactions (DDIs) remain a major source of preventable harm.<n>We introduce, to our knowledge, the first system that conditions KG relation scoring on patient-level EHR context.<n>A fusion "Teacher" learns mechanism-specific relations for drug pairs represented in both sources.<n>A distilled "Student" generalizes to new or rarely used drugs without KG access at inference.
  arXiv  Detail & Related papers  (2025-11-10T03:18:16Z)
• GramSeq-DTA: A grammar-based drug-target affinity prediction approach fusing gene expression information [1.2289361708127877]
  We propose GramSeq-DTA, which integrates chemical perturbation information with the structural information of drugs and targets. Our approach outperforms the current state-of-the-art DTA prediction models when validated on widely used DTA datasets.
  arXiv  Detail & Related papers  (2024-11-03T03:17:09Z)
• Regressor-free Molecule Generation to Support Drug Response Prediction [83.25894107956735]
  Conditional generation based on the target IC50 score can obtain a more effective sampling space. Regressor-free guidance combines a diffusion model's score estimation with a regression controller model's gradient based on number labels.
  arXiv  Detail & Related papers  (2024-05-23T13:22:17Z)
• drGT: Attention-Guided Gene Assessment of Drug Response Utilizing a Drug-Cell-Gene Heterogeneous Network [8.664433151000255]
  drGT is a graph deep learning model that predicts sensitivity and aids in biomarker identification using attention coefficients (ACs)<n>The model is trained and evaluated using major benchmark datasets: Sanger GDSC, NCI60, and Broad CTRP, which cover a wide range of drugs and cancer cell lines.
  arXiv  Detail & Related papers  (2024-05-14T22:16:52Z)
• SSM-DTA: Breaking the Barriers of Data Scarcity in Drug-Target Affinity Prediction [127.43571146741984]
  Drug-Target Affinity (DTA) is of vital importance in early-stage drug discovery. wet experiments remain the most reliable method, but they are time-consuming and resource-intensive. Existing methods have primarily focused on developing techniques based on the available DTA data, without adequately addressing the data scarcity issue. We present the SSM-DTA framework, which incorporates three simple yet highly effective strategies.
  arXiv  Detail & Related papers  (2022-06-20T14:53:25Z)
• Taming Overconfident Prediction on Unlabeled Data from Hindsight [50.9088560433925]
  Minimizing prediction uncertainty on unlabeled data is a key factor to achieve good performance in semi-supervised learning. This paper proposes a dual mechanism, named ADaptive Sharpening (ADS), which first applies a soft-threshold to adaptively mask out determinate and negligible predictions. ADS significantly improves the state-of-the-art SSL methods by making it a plug-in.
  arXiv  Detail & Related papers  (2021-12-15T15:17:02Z)
• Relational graph convolutional networks for predicting blood-brain barrier penetration of drug molecules [12.041672273431994]
  The evaluation of the BBB penetrating ability of drug molecules is a critical step in brain drug development. We employ the relational graph convolutional network (RGCN) to handle the drug-protein relations as well as the features of each individual drug. The performance was already promising, demonstrating the significant role of the drug-protein/drug relations in the prediction of BBB permeability.
  arXiv  Detail & Related papers  (2021-07-04T15:56:02Z)
• Weakly-Supervised Cross-Domain Adaptation for Endoscopic Lesions Segmentation [79.58311369297635]
  We propose a new weakly-supervised lesions transfer framework, which can explore transferable domain-invariant knowledge across different datasets. A Wasserstein quantified transferability framework is developed to highlight widerange transferable contextual dependencies. A novel self-supervised pseudo label generator is designed to equally provide confident pseudo pixel labels for both hard-to-transfer and easy-to-transfer target samples.
  arXiv  Detail & Related papers  (2020-12-08T02:26:03Z)
• Deep Learning for Virtual Screening: Five Reasons to Use ROC Cost Functions [80.12620331438052]
  deep learning has become an important tool for rapid screening of billions of molecules in silico for potential hits containing desired chemical features. Despite its importance, substantial challenges persist in training these models, such as severe class imbalance, high decision thresholds, and lack of ground truth labels in some datasets. We argue in favor of directly optimizing the receiver operating characteristic (ROC) in such cases, due to its robustness to class imbalance.
  arXiv  Detail & Related papers  (2020-06-25T08:46:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.