Chen Song

This study shows that AlphaFold2 distance maps can reveal multi-state protein structures, enabling prediction of alternative conformations with high accuracy.

The paper develops selective fluorescent sensors to detect endocannabinoid dynamics in vivo and shows that Δ9-THC induces sustained 2-AG signaling in the nucleus accumbens.

The paper identifies CLCC1 as a client of Torsin ATPases, showing they organize CLCC1 to maintain ER bilayer and lipid homeostasis, with implications for MASH development.

This paper introduces AMPainterV2, a model that designs membrane-lytic antimicrobial peptides using generative adversarial imitation learning, achieving active peptides with improved activity and miniaturization.

PepMCP predicts membrane contact probability for short peptides using graph-based modeling, achieving high correlation with experimental data.

SteerAF improves AlphaFold2's ability to predict alternative protein conformations by using distogram information from MSAs, achieving better results than existing methods.

The paper determines the structure of the poxvirus portal complex and its pore, revealing their role in molecular transport and antiviral target potential.

CTGoMartini simulates protein conformational transitions using G[o]-Martini models, enabling the study of functional state changes in membrane proteins.

MemConverter reprograms protein localization between membrane and aqueous environments using MCP-guided sequence design and AlphaFold2 refinement, achieving stable integration or solubility.

This paper introduces improved Anisotropic Network Models that incorporate Membrane Contact Probability to better predict membrane protein dynamics and mechanosensitivity, aligning with experimental findings.

This study uses molecular dynamics simulations to show that lipid asymmetry, specifically LPC and PIP2 conversion, affects membrane structure and curvature.

This study screens antimicrobial peptides using a mechanism-driven approach, identifying seven peptides with membrane-targeting and pore-forming activity that show antimicrobial efficacy with low cytotoxicity.

This study identifies torque on the TRP domain as the primary driver of NOMPC channel opening, revealing a twist-to-open mechanism facilitated by AR domain compression-twist coupling.

This paper introduces AMPainter, a deep reinforcement learning model for designing antimicrobial peptides, which outperforms existing models and generates effective AMPs from signal peptides and random sequences.

This study uses molecular dynamics simulations to reveal the three-ion knock-on mechanism governing calcium ion permeation and valence selectivity in CaV1 channels.

This paper introduces a new computational method for simulating protein conformational transitions in explicit environments and demonstrates its application across multiple protein types.

This study uses molecular dynamics simulations to show that specific THz frequencies enhance conductance of Kv1.2 and Nav1.5 ion channels by affecting ion motion in selectivity filters.

The paper outlines a simulation protocol to identify THz frequencies affecting ion channels and finds a broad frequency band between 1 to 20 THz impacts CaV channels.

The paper introduces ProtRAP-LM, a transformer-based model using language model embeddings to predict membrane contact probability and relative accessibility of proteins, achieving faster and more accurate results than MSA-based methods.

The paper introduces a standardized NRPS architecture using conserved motifs, leading to new subtype classifications, novel motifs, and insights into reengineering challenges.