arXiv:2512.00521v1 Announce Type: new 
Abstract: In early stage drug discovery, bioactivity prediction of molecules against target proteins plays a crucial role. Trdaitional QSAR models that utilizes molecular descriptor based data often struggles to predict bioactivity of molecules effectively due to its limitation in capturing structural and contextual information embedded within each compound. To address this challenge, we propose Rep3Net, a unified deep learning architecture that not only incorporates descriptor data but also includes spatial and relational information through graph-based represenation of compounds and contextual information through ChemBERTa generated embeddings from SMILES strings. Our model employing multimodal concatenated features produce reliable bioactivity prediction on Poly [ADP-ribose] polymerase 1 (PARP-1) dataset. PARP-1 is a crucial agent in DNA damage repair and has become a significant theraputic target in malignancies that depend on it for survival and growth. A comprehensive analysis and comparison with conventional standalone models including GCN, GAT, XGBoost, etc. demonstrates that our architecture achieves the highest predictive performance. In computational screening of compounds in drug discovery, our architecture provides a scalable framework for bioactivity prediction.

تم اقتراح بنية جديدة للتعلم العميق تُدعى Rep3Net لتحسين التنبؤ بالنشاط الحيوي للجزيئات في المراحل المبكرة من اكتشاف الأدوية. يدمج هذا النموذج بيانات الوصف الجزيئي التقليدية مع المعلومات المكانية والعلاقية من خلال تمثيلات رسومية وبيانات سياقية تم إنشاؤها بواسطة ChemBERTa من سلاسل SMILES، مع التركيز بشكل خاص على مجموعة بيانات بولي [ADP-ribوز] بوليميراز 1 (PARP-1).

Se ha propuesto una nueva arquitectura de aprendizaje profundo llamada Rep3Net para mejorar la predicción de la bioactividad molecular en las primeras etapas del descubrimiento de fármacos. Este modelo integra datos de descriptores moleculares tradicionales con información espacial y relacional a través de representaciones gráficas y embeddings contextuales generados por ChemBERTa a partir de cadenas SMILES, enfocándose específicamente en el conjunto de datos de la poliamina [ADP-ribosa] polimerasa 1 (PARP-1).

Une nouvelle architecture d'apprentissage profond nommée Rep3Net a été proposée pour améliorer la prédiction de la bioactivité moléculaire dans la découverte de médicaments à un stade précoce. Ce modèle intègre des données de descripteurs moléculaires traditionnels avec des informations spatiales et relationnelles provenant de représentations graphiques et des embeddings contextuels générés par ChemBERTa à partir de chaînes SMILES, ciblant spécifiquement le jeu de données de la poly [ADP-ribose] polymérase 1 (PARP-1).

A new deep learning architecture named Rep3Net has been proposed to enhance molecular bioactivity prediction in early-stage drug discovery. This model integrates traditional molecular descriptor data with spatial and relational information from graph representations and contextual embeddings generated by ChemBERTa from SMILES strings, specifically targeting the Poly [ADP-ribose] polymerase 1 (PARP-1) dataset.

Rep3Net: An Approach Exploiting Multimodal Representation for Molecular Bioactivity Prediction

arXiv:2512.01333v1 Announce Type: cross 
Abstract: Stroke is a major cause of death and permanent impairment, making it a major worldwide health concern. For prompt intervention and successful preventative tactics, early risk assessment is essential. To address this challenge, we used ensemble modeling and explainable AI (XAI) techniques to create an interpretable machine learning framework for stroke risk prediction. A thorough evaluation of 10 different machine learning models using 5-fold cross-validation across several datasets was part of our all-inclusive strategy, which also included feature engineering and data pretreatment (using Random Over-Sampling (ROS) to solve class imbalance). Our optimized ensemble model (Random Forest + ExtraTrees + XGBoost) performed exceptionally well, obtaining a strong 99.09% accuracy on the Stroke Prediction Dataset (SPD). We improved the model's transparency and clinical applicability by identifying three important clinical variables using LIME-based interpretability analysis: age, hypertension, and glucose levels. Through early prediction, this study highlights how combining ensemble learning with explainable AI (XAI) can deliver highly accurate and interpretable stroke risk assessment. By enabling data-driven prevention and personalized clinical decisions, our framework has the potential to transform stroke prediction and cardiovascular risk management.

تم تطوير إطار جديد للتعلم الآلي لتحسين توقعات مخاطر السكتة الدماغية، باستخدام تقنيات النمذجة الجماعية والذكاء الاصطناعي القابل للتفسير. حقق هذا الإطار دقة ملحوظة تبلغ 99.09% على مجموعة بيانات توقع السكتة الدماغية من خلال استخدام إعادة العينة العشوائية لمعالجة عدم توازن الفئات وتحديد المتغيرات السريرية الرئيسية من خلال طرق قائمة على LIME.

Se ha desarrollado un nuevo marco de aprendizaje automático para optimizar la predicción del riesgo de accidente cerebrovascular, utilizando técnicas de modelado en conjunto y de inteligencia artificial explicable. Este marco logró una notable precisión del 99.09% en el Conjunto de Datos de Predicción de Accidentes Cerebrovasculares al emplear Sobremuestreo Aleatorio para abordar el desequilibrio de clases e identificar variables clínicas clave a través de métodos basados en LIME.

Un nouveau cadre d'apprentissage automatique a été développé pour optimiser la prédiction du risque d'accident vasculaire cérébral, en utilisant des techniques de modélisation par ensembles et d'intelligence artificielle explicable. Ce cadre a atteint une précision remarquable de 99,09 % sur le jeu de données de prédiction des AVC en utilisant le suréchantillonnage aléatoire pour traiter le déséquilibre des classes et en identifiant des variables cliniques clés grâce à des méthodes basées sur LIME.

A new machine learning framework has been developed to optimize stroke risk prediction, utilizing ensemble modeling and explainable AI techniques. This framework achieved a remarkable accuracy of 99.09% on the Stroke Prediction Dataset by employing Random Over-Sampling to address class imbalance and identifying key clinical variables through LIME-based methods.

Rep3Net: An Approach Exploiting Multimodal Representation for Molecular Bioactivity Prediction

Was this article worth reading? Share it

LucidQuery AI

Augmeta

ZeroGPT.org