Tianqi Wu

This study identifies dimer opening in CK-BB as a mechanism for sensing membrane curvature and regulating ATP regeneration under acidic conditions and substrate binding.

The paper identifies an enzymatic pathway for isoaspartate formation in bacterial ribosomal protein uS11, showing its conservation and structural importance.

The paper investigates the mechanism of proton-coupled transport in the mycobacterial Spns transporter using spectroscopy, modeling, and mutagenesis, revealing how protonation and substrate binding influence conformational states and transport directionality.

The paper uses AlphaFold3 to predict ligand-dependent conformational changes in ABC transporters and shows these predictions correlate with experimental data and energetics.

This paper introduces TransPross, a transformer-based model for protein secondary structure prediction that outperforms existing methods by leveraging attention mechanisms and deep learning.

The paper modifies AlphaFold2 to incorporate distance distributions from DEER spectroscopy, improving conformational prediction with fewer experimental constraints.

The paper introduces MULTICOM, a system that enhances AlphaFold-Multimer for protein complex structure prediction, achieving higher TM-scores in CASP15.

The paper improves AlphaFold2-based protein structure prediction using MULTICOM_refine, achieving higher accuracy in CASP15 compared to standard AlphaFold2.

The paper introduces EnQA, a 3D-equivariant graph neural network for assessing protein model quality, achieving state-of-the-art performance and outperforming AlphaFold2's own scores.

ATOMRefine improves protein structure prediction by refining all-atom coordinates using a SE(3)-equivariant graph neural network, enhancing both backbone and side-chain accuracy.

The paper improves protein tertiary structure prediction by enhancing deep learning and distance prediction methods in the MULTICOM system, achieving better performance in CASP14.

The paper develops deep learning-based EMA predictors for protein model accuracy estimation in CASP14, achieving top rankings in GDT-TS score loss.

The paper develops deep learning-based protein distance predictors for CASP14, achieving top rankings in contact prediction and showing the importance of MSAs and features in accuracy.

The paper evaluates 43 methods for predicting intrinsically disordered regions and binding regions in proteins, finding that deep learning methods outperform older physicochemical approaches but binding prediction remains challenging.

This study introduces an attention-based convolutional neural network for protein contact prediction, showing improved interpretability and predictive performance on CASP13 targets.

DeepDist predicts real-value inter-residue distances using a deep residual network, improving contact prediction and tertiary structure reconstruction compared to multi-class methods.

Cascaded CNNs segmenting secondary structure, backbone and Cα voxels, plus tabu-search tracing, place 88.5% of Cα atoms within 3Å across 50 cryo-EM maps versus Phenix's 66.8%.

The paper describes improvements to the MULTICOM system for protein structure prediction using deep learning and contact distance prediction in CASP13, achieving top rankings in both template-free and template-based modeling.