AIBullisharXiv – CS AI · Jun 197/10
🧠Emyx, a 140M-parameter conditional flow matching model, achieves superior protein generation performance while requiring 4x less training compute than existing systems like RFdiffusion3. The model demonstrates that enzyme design generators can operate efficiently without inheriting expensive architectures from structure prediction systems, outperforming larger competitors on strict geometric accuracy and structural diversity benchmarks.
AINeutralarXiv – CS AI · Jun 236/10
🧠Researchers demonstrate that protein contact prediction can be extracted from language model attention heads in a single forward pass, outperforming the computationally expensive Categorical Jacobian method on clean test data. The findings reveal that contact information is concentrated in a small subset of attention heads, requiring only 10 labeled proteins for head selection.
AINeutralarXiv – CS AI · Jun 236/10
🧠Researchers propose Manifold Restore Mixing (MRM), a novel data augmentation method that addresses structural degradation issues in protein representation learning by mixing hidden representations of original and augmented protein data. The approach combines manifold mixup techniques with a difficulty scheduler to generate training samples that preserve protein structure while introducing beneficial variations.
AINeutralarXiv – CS AI · Jun 196/10
🧠Researchers introduce SSProNet, a graph neural network that improves protein representation learning by incorporating secondary structure information and energy-filtered hydrogen-bond interactions. The approach demonstrates consistent improvements over existing graph-based methods while offering enhanced biological interpretability aligned with established structural motifs.
AINeutralarXiv – CS AI · Jun 46/10
🧠Researchers introduce MuCO, a generative AI method for modeling cyclic peptide structures through multi-stage conformation optimization. The approach outperforms existing methods in stability, diversity, and efficiency, offering significant implications for computational drug discovery and peptide-based therapeutic development.
AINeutralarXiv – CS AI · Jun 26/10
🧠Researchers introduce SGAP-PPIS, a graph neural network model that uses adaptive propagation guided by protein structure geometry to predict protein-protein interaction sites more accurately. The model dynamically adjusts how information flows between residues based on their local geometric environment, outperforming fixed propagation approaches in distinguishing true interaction sites from similar non-interacting regions.
AINeutralarXiv – CS AI · Jun 26/10
🧠Researchers have developed a method to enhance generative AI models that simulate protein dynamics by introducing a history-dependent bias that steers sampling toward undiscovered molecular states. The technique achieves 37× faster coverage of low-energy protein configurations compared to standard approaches, significantly improving the practical utility of AI-accelerated molecular simulation.
AINeutralarXiv – CS AI · May 126/10
🧠Researchers introduce EDMolGPT, a generative AI model that uses electron density data from protein binding pockets to design novel drug molecules. The approach improves upon existing methods by incorporating physically grounded density information rather than empty pocket structures, enabling more accurate molecular generation with realistic 3D conformations.
AIBullisharXiv – CS AI · Mar 55/10
🧠Researchers developed Cryo-SWAN, a new AI autoencoder network that uses wavelet decomposition to better represent 3D molecular structures from cryo-electron microscopy data. The model outperforms existing 3D autoencoders on multiple datasets and can integrate with diffusion models for molecular shape generation and denoising.
AIBullisharXiv – CS AI · Mar 36/103
🧠Researchers have developed GeoBPE, a new protein structure tokenization method that converts protein backbone structures into discrete geometric tokens, achieving over 10x compression and data efficiency improvements. The approach uses geometry-grounded byte-pair encoding to create hierarchical vocabularies of protein structural primitives that align with functional families and enable better multimodal protein modeling.
AINeutralarXiv – CS AI · Mar 27/1012
🧠Researchers developed a novel method to represent local protein environments using atomistic foundation models (AFMs), creating embeddings that capture both structural and chemical features. The approach enables construction of data-driven priors for biomolecular environments and achieves state-of-the-art accuracy in physics-informed chemical shift prediction for NMR spectroscopy.
