Accuracy (5-fold): 80.4%Recall (5-fold): 71.0%Precision (5-fold): 80.1%Selectivity (5-fold): 87.9%
Papers with Code · CC BY-SA 4.0
Antibody-antigen binding prediction
Antibody-antigen binding prediction involves using computational or experimental methods to assess how well an antibody interacts with its target antigen. This can be done by analyzing features such as sequence, structure, and physicochemical properties. These predictions aid in designing therapeutic antibodies, vaccines, and diagnostic tests. Validation through experimental assays ensures the accuracy of the predictions and their applicability in biomedical research and development.
The binding site prediction task is used to predict which antibody residues interact with an antigen. A residue is considered part of the paratope’s binding site if any of its heavy atoms (non-hydrogen atoms) is located within 4.5˚A of any antigen-heavy atom.
개요는 각 항목의 일부입니다. 위 탭에서 전체 기록을 페이지별로 탐색할 수 있습니다.
벤치마크
Papers with Code 아카이브의 데이터셋별 기록 예시입니다. 평가 조건과 지표 방향이 검증되지 않아 순위를 의미하지 않습니다.
데이터셋
4개논문
AUC-PR: 0.650AUC-ROC: 0.927
AUC-PR: 0.781AUC-ROC: 0.967
Accuracy (5-fold): 65.8%Recall (5-fold): 35.3%Precision (5-fold): 82.9%Selectivity (5-fold): 93.4%
AUC-PR: 0.612AUC-ROC: 0.883
AUC-PR: 0.741AUC-ROC: 0.927
Learning context-aware structural representations to predict antigen and antibody binding interfaces
AUC-PR: 0.713AUC-ROC: 0.915
AUC-PR: 0.652AUC-ROC: 0.868
AUC-PR: 0.724AUC-ROC: 0.856
라이브러리
Papers with Code 리더보드 row에 연결된 구현 저장소입니다.