arXiv AI September 25, 2026 By Quang Minh Nguyen, Thuy Quynh Nguyen, Duc Minh Le, Ho Nhat Minh Nguyen, Thanh Long Dai Doan, Trong Nghia Nguyen

SMILESGNN: Interpretable Clinical Toxicity Prediction via SMILES-Graph Cross-Attention Fusion

Read the original on arXiv AI →

SMILESGNN is a multimodal architecture that fuses a SMILES Transformer encoder with a GATv2 graph encoder through cross‑attention, enabling interpretable clinical toxicity predictions. The model retains an explicit graph branch, allowing GNNExplainer to identify substructures linked to toxicity. On the ClinTox dataset it achieves an AUC‑ROC of 0.987 and F1 of 0.906 with only 0.4 M parameters, while on Tox21 it attains a mean AUC‑ROC of 0.750, comparable to strong single‑modality baselines.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI

Jul 7

Predicting Therapeutic Outcome via Aligning Patient-Specific Knowledge Graph and Gene-Level Perturbation Representations

arXiv:2607. 04557v1 Announce Type: cross 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.

By Dongmin Bang, Sugyun An, Inyoung Sung, Ilho Yun, Sun Kim, Sangseon Lee

rag diffusion benchmarks

Hugging Face Trending Papers

Jul 6

Predicting Therapeutic Outcome via Aligning Patient-Specific Knowledge Graph and Gene-Level Perturbation Representations

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.

rag diffusion benchmarks

arXiv Machine Learning

2d ago

Benchmarking graph-based models for in-silico toxicity prediction in drug discovery

arXiv:2609.37555v1 Announce Type: new Abstract: Drug discovery is a costly and high-risk process, where toxicity-related failures remain a major cause of attrition in both preclinical and clinical st...

By Noel Suarez-Barro, Manuel Lama, Juan C. Vidal

benchmarks safety

arXiv AI

Jun 9

Pharmacogenomic Knowledge Graph Augmentation for Graph Neural Network-Based Drug-Drug Interaction Prediction

arXiv:2606. 07698v1 Announce Type: cross Abstract: Graph neural networks (GNNs) applied to drug-drug interaction (DDI) prediction rely exclusively on molecular structure encoded as SMILES-derived graphs.

By Juergen Dietrich

rag multimodal benchmarks safety

arXiv Machine Learning

Jun 16

A polarity-aware multi-relational model for the signed interaction prediction in biological networks

arXiv:2407. 07357v3 Announce Type: replace Abstract: Predicting signed interactions in biological networks is crucial for understanding drug mechanisms and facilitating drug repurposing.

By Ziye Zhou, Meijie Wang, Lun Yu

arXiv Machine Learning

Jul 9

Asymmetric Focal Loss Improves Graph Neural Network Prediction of Drug-Drug Interactions

arXiv:2607. 07611v1 Announce Type: new Abstract: Background: Graph neural networks improve computational prediction of polypharmacy side effects, but standard binary cross-entropy training allocates equal capacity to well-classified and difficult examples, potentially missing clinically significant interactions.

By Faranak Hatami, Mousa Moradi

rag

One email a morning, machine-written

Related stories

Predicting Therapeutic Outcome via Aligning Patient-Specific Knowledge Graph and Gene-Level Perturbation Representations

Predicting Therapeutic Outcome via Aligning Patient-Specific Knowledge Graph and Gene-Level Perturbation Representations

Benchmarking graph-based models for in-silico toxicity prediction in drug discovery

Pharmacogenomic Knowledge Graph Augmentation for Graph Neural Network-Based Drug-Drug Interaction Prediction

A polarity-aware multi-relational model for the signed interaction prediction in biological networks

Asymmetric Focal Loss Improves Graph Neural Network Prediction of Drug-Drug Interactions