Kei Terayama

The paper introduces aBCC, a machine learning-based metric for evaluating ligand coordinate-density consistency in crystallographic structures across different resolutions.

This study evaluates a multimodal LLM workflow for extracting biological information from text and figures, achieving an F1-score of 0.847 using Gemini 2.5 Pro.

This paper develops a data-driven feed optimization system for aquaculture using Bayesian optimization and omics analysis to improve sustainability and growth efficiency.

The paper introduces cryoMDM, a method that estimates 3D atomic models from single cryoEM particle images by matching them to simulated structural models, revealing intermediate structures of the SARS-CoV-2 spike protein.

The paper builds a dataset and two models to predict protein-RNA condensate formation under various environmental conditions.

This paper introduces DEFMap, a deep learning method that extracts dynamics information from cryo-EM maps, achieving results comparable to molecular dynamics simulations and experimental approaches.