Sheng Wang

The paper uses interspecific hybrid embryos to identify EIF1A-family proteins as conserved nuclear regulators of zygotic genome activation across mammals.

This paper introduces Pinal, a foundation model that designs proteins from natural language descriptions, achieving functional alignment and experimental success in four design tasks.

This paper introduces zPolaron, a graph neural network for functional protein design that integrates structural, interaction, and evolutionary information, improving fitness prediction beyond structural compatibility.

This study introduces AF-pred, an Alphafold-based tool for cross-species prediction of MHC-II restricted epitopes, showing better performance for non-human MHC-II than sequence-based methods.

This paper identifies the enzymatic mechanism for forming the gamma-thiolactone ring in thiolactomycin biosynthesis, revealing the roles of TlnCCy and TlnA.

This study benchmarks AlphaFold3-like methods for predicting protein-peptide complex structures, showing improved accuracy but challenges in binding site prediction.

The paper introduces CARD, a method that integrates complex-level information to improve tertiary structure-based RNA design by incorporating protein features and affinity evaluation.

This study benchmarks reverse docking pipelines using AlphaFold2 human proteome to predict drug-target interactions and off-target effects, finding that Glide_sfct (PS) performs best.

The paper curates a database of disulfide bonds and their effects on protein thermostability, develops machine learning models to predict beneficial disulfide bonds, and demonstrates the utility of AlphaFold2 for this task.

This paper introduces IGModel, a deep learning framework that predicts protein-ligand docking RMSD and binding strength, achieving state-of-the-art performance on multiple benchmarks.

The paper introduces Z-align, a molecular alignment method that improves binding pose prediction by using topological structural information and flexible optimization of bond lengths and angles.

The paper describes an RNA tertiary structure modeling method used in CASP15, combining secondary structure information, motif prediction, and the BRiQ potential for energy optimization.

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.

The paper develops a method to quantify rare genetic variants in human cells using long-read sequencing and shows CRISPR-Cas9 induces nonrandom large deletions and insertions.

PointSite identifies protein ligand binding atoms using point cloud segmentation, outperforming previous methods and reducing false positives through cascaded filtering.

This paper introduces two novel algorithms for efficiently searching and mapping genomic subsequences in nanopore raw signals, improving SNP detection and haplotype classification under low coverage and accuracy.

This paper evaluates RaptorX's performance in CASP13 using distance-based methods, showing strong contact prediction and fold prediction for some targets despite underperforming AlphaFold in 3D modeling.

PureseqTM predicts transmembrane topology from amino acid sequences, outperforming existing methods and correcting UniProt annotations in the human proteome.

The paper introduces cwDTW, a wavelet-based DTW algorithm for nanopore sequencing signal alignment, achieving high accuracy and speed compared to existing methods.

This paper introduces a hybrid method combining clustering and deep learning to predict real-valued protein backbone dihedral angles and their confidence bounds, achieving better performance than existing methods.

The paper shows deep learning outperforms DCA in predicting inter-protein contacts using only intra-protein data and improves alignment for better predictions.

Chiron is a deep learning model that directly translates nanopore raw signals into DNA sequences without segmentation, achieving high accuracy and speed with minimal training data.

DeepSimulator generates realistic Nanopore sequencing data using a context-dependent deep learning model, improving signal accuracy and enabling tool development.

The paper classifies binding and non-binding intra-protein peptide fragments with high accuracy using a feed-forward neural network.

This paper introduces a deep transfer learning method for predicting membrane protein structures, achieving improved contact prediction and accurate 3D models for many membrane proteins.

The paper evaluates a deep learning method for protein contact prediction in CASP12, achieving 47% and 44% accuracy for long- and medium-range contacts, respectively.

DeepBound identifies transcript boundaries from RNA-seq reads using deep convolutional neural fields, outperforming existing methods on simulated and biological datasets.

This paper introduces a deep learning method for predicting protein contact maps, achieving higher accuracy than existing methods, especially for proteins with few sequence homologs.