Minkyung Baek

Thal-Kak combines five structure predictors to improve sampling but finds confidence scores and quality assessment methods hinder model selection.

The paper uses high-throughput antibody variant analysis and AI models to optimize antibody productivity by navigating fitness landscapes and identifying structurally compatible CDR variants.

The paper predicts 18,316 human protein-protein interactions with 90% precision using enhanced coevolutionary analysis and a new deep learning model.

The paper determines the structure of E. coli LetAB, identifies LetA as a novel transporter family, and proposes a mechanism for lipid transport using proton-motive force.

This study uses deep learning to predict protein-protein interactions in bacterial pathogens, identifying new complexes involving essential genes and virulence factors, with half of tested predictions validated.

This paper applies parameter-efficient fine-tuning methods to proteomics, achieving competitive or better performance than traditional fine-tuning with significantly reduced computational resources.

The paper introduces a method to map antibody-antigen interactions using SPID platform to enhance affinity prediction and optimize antibody design.

This paper shows that targeting FAM19A5 with a monoclonal antibody improves synaptic function and cognition in Alzheimer's disease models.

Fine-tuning RoseTTAFold as a diffusion denoiser gives RFdiffusion, a general protein backbone generator; a designed influenza haemagglutinin binder matched its model by cryo-EM.

The paper fine-tunes AlphaFold to predict peptide binding specificity for MHC and other domains, achieving performance comparable to sequence-based methods.

This paper introduces a top-down reinforcement learning approach to design protein nanomaterials, resulting in more compact and robust icosahedral assemblies than previously observed.

One RoseTTAFold-derived network models protein-DNA and protein-RNA complexes plus RNA tertiary structure end to end, returning confidence estimates alongside each 3D model.

The paper introduces a 3-track network for accurate protein structure and interaction prediction, achieving performance close to DeepMind's CASP14 results and enabling rapid modeling of protein complexes from sequence data.

The paper introduces DeepAccNet, a deep learning framework that improves protein structure refinement by estimating accuracy and guiding Rosetta refinement.