🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Multi-Modal Molecular Representation Learning via Structure Awareness １．This work presents MMSA, a novel self-supervised framework for molecular representation learning that fuses 2D/3D graphs and molecular images, enhanced by structure-aware hypergraph modeling and memory

Multi-Modal Molecular Representation Learning via Structure Awareness １．This work presents MMSA, a novel self-supervised framework for molecular representation learning that fuses 2D/3D graphs and molecular images, enhanced by structure-aware hypergraph modeling and memory

PyTDC: A multimodal machine learning training, evaluation, and inference platform for biomedical foundation models １．PyTDC is the first end-to-end machine learning platform designed to train, evaluate, and deploy biomedical foundation models that integrate multimodal

PyTDC: A multimodal machine learning training, evaluation, and inference platform for biomedical foundation models １．PyTDC is the first end-to-end machine learning platform designed to train, evaluate, and deploy biomedical foundation models that integrate multimodal

LLMs Outperform Experts on Challenging Biology Benchmarks １．This study systematically evaluates 27 frontier large language models (LLMs) on eight biology-specific benchmarks, revealing that recent models are now matching or surpassing domain experts in multiple challenging

LLMs Outperform Experts on Challenging Biology Benchmarks １．This study systematically evaluates 27 frontier large language models (LLMs) on eight biology-specific benchmarks, revealing that recent models are now matching or surpassing domain experts in multiple challenging

Multimodal Bonds Reconstruction Towards Generative Molecular Design １．This paper introduces YuelBond, a graph neural network framework designed to robustly reconstruct chemical bonds from diverse molecular data—accurate 3D structures, noisy generative outputs, and 2D

Multimodal Bonds Reconstruction Towards Generative Molecular Design １．This paper introduces YuelBond, a graph neural network framework designed to robustly reconstruct chemical bonds from diverse molecular data—accurate 3D structures, noisy generative outputs, and 2D

LightNobel: Improving Sequence Length Limitation in Protein Structure Prediction Model via Adaptive Activation Quantization １．This paper introduces LightNobel, the first hardware-software co-designed system specifically optimized to overcome sequence length limitations in

LightNobel: Improving Sequence Length Limitation in Protein Structure Prediction Model via Adaptive Activation Quantization １．This paper introduces LightNobel, the first hardware-software co-designed system specifically optimized to overcome sequence length limitations in

Assessment of nucleic acid structure prediction in CASP16 １．This paper presents a comprehensive assessment of RNA and DNA structure prediction in the CASP16 experiment, highlighting the stark contrast in predictive accuracy between nucleic acids and proteins, despite the rise

Assessment of nucleic acid structure prediction in CASP16 １．This paper presents a comprehensive assessment of RNA and DNA structure prediction in the CASP16 experiment, highlighting the stark contrast in predictive accuracy between nucleic acids and proteins, despite the rise

A 3D pocket-aware and evolutionary conserved interaction guided diffusion model for molecular optimization １．This paper introduces DiffDecip, a novel 3D diffusion model for molecular optimization that explicitly integrates both protein-ligand interaction priors and

A 3D pocket-aware and evolutionary conserved interaction guided diffusion model for molecular optimization １．This paper introduces DiffDecip, a novel 3D diffusion model for molecular optimization that explicitly integrates both protein-ligand interaction priors and

Discovery of a DNA-based Optical Nanotube Sensor for Glucose Using Clustering and Deep Learning Algorithms １．This study presents the first successful identification of a single-stranded DNA (ssDNA) sequence that enables optical glucose sensing via single-walled carbon

Discovery of a DNA-based Optical Nanotube Sensor for Glucose Using Clustering and Deep Learning Algorithms １．This study presents the first successful identification of a single-stranded DNA (ssDNA) sequence that enables optical glucose sensing via single-walled carbon

SBMLNetwork: a framework for standards-based visualization of biochemical models １．SBMLNetwork introduces a robust, open-source software library that finally makes the SBML Layout and Render standards practical for widespread, standards-compliant visualization of biological

SBMLNetwork: a framework for standards-based visualization of biochemical models １．SBMLNetwork introduces a robust, open-source software library that finally makes the SBML Layout and Render standards practical for widespread, standards-compliant visualization of biological

GeoFlow-V2: A Unified Atomic Diffusion Model for Protein Structure Prediction and De Novo Design １．GeoFlow-V2 introduces a unified atomic diffusion framework that bridges protein structure prediction and de novo design. It supports proteins, nucleic acids, and small molecules,

GeoFlow-V2: A Unified Atomic Diffusion Model for Protein Structure Prediction and De Novo Design １．GeoFlow-V2 introduces a unified atomic diffusion framework that bridges protein structure prediction and de novo design. It supports proteins, nucleic acids, and small molecules,