An active learning workflow that efficiently trains a deep learning model to learn energy functions for specific protein targets, combining the advantages of machine learning and physics-based computations to achieve more efficient antibody development.
항체 개발 시 현재 바이러스 변종뿐만 아니라 미래 변종에 대한 대응을 고려하는 것이 중요하다. 이를 위해 상대방 형성(opponent shaping) 기법을 활용하여 항체를 최적화하면 바이러스 진화 경로를 효과적으로 제한할 수 있다.
Antibody shapers optimized using opponent shaping principles can effectively limit viral escape by anticipating and influencing the evolutionary trajectories of viruses.
構造ベースの創薬において、原子核と電子雲の周りの変調を同時にモデル化することで、原子間の最小距離制約を満たしつつ、高い結合親和性を持つリガンドを生成することができる。
A manifold-constrained denoising diffusion model, NucleusDiff, is proposed to effectively generate ligands with high binding affinities and reduced separation violations for structure-based drug design.
최근 딥러닝 및 대규모 언어 모델(LLM)의 발전은 마이크로바이옴과 메타게놈 데이터 분석에 큰 영향을 미쳤다. 미생물 단백질 및 유전체 서열은 생명의 언어와 같아서 LLM을 활용하여 복잡한 미생물 생태계로부터 유용한 통찰을 얻을 수 있다.
Recent advancements in deep learning and large language models have significantly impacted the study of microbiomes and metagenomics, enabling researchers to extract valuable insights from the complex language of microbial genomic and protein sequences.
Eyelash mites, known as Demodex, are a common microscopic inhabitant found in the eyelashes of most people worldwide, and can cause issues if left unchecked.
The Krebs cycle dataset provides a standardized synthetic benchmark for evaluating causal learning methods using time series data from a real-world biochemical process.
Hypergraph diffusion wavelets provide an efficient and multiscale framework for representing complex cellular niches in spatial transcriptomics data, enabling the identification of disease-relevant cellular neighborhoods in Alzheimer's disease.