当前位置：
X-MOL 学术
›
Chem. Asian J.
›
论文详情
Our official English website, www.x-mol.net, welcomes your
feedback! (Note: you will need to create a separate account there.)
Rotational Variance-Based Data Augmentation in 3D Graph Convolutional Network
Chemistry - An Asian Journal ( IF 3.5 ) Pub Date : 2021-08-09 , DOI: 10.1002/asia.202100789 Jihoo Kim 1 , Yeji Kim 1 , Eok Kyun Lee 1 , Chong Hak Chae 2 , Kwangho Lee 2 , Won June Kim 3 , Insung S Choi 1
Chemistry - An Asian Journal ( IF 3.5 ) Pub Date : 2021-08-09 , DOI: 10.1002/asia.202100789 Jihoo Kim 1 , Yeji Kim 1 , Eok Kyun Lee 1 , Chong Hak Chae 2 , Kwangho Lee 2 , Won June Kim 3 , Insung S Choi 1
Affiliation
|
The prediction performance of 3D graph convolutional network is greatly enhanced, in the ligand-binding task to β-secretase-1, by data augmentation that is achieved by rotating known active ligands in the 3D Cartesian coordinate and assigning the rotated ones as inactive data.
中文翻译：
3D 图卷积网络中基于旋转方差的数据增强
在 β-secretase-1 的配体结合任务中，通过在 3D 笛卡尔坐标中旋转已知的活性配体并将旋转后的配体分配为非活动数据来实现数据增强，3D 图形卷积网络的预测性能大大增强。
更新日期：2021-09-20
中文翻译：
3D 图卷积网络中基于旋转方差的数据增强
在 β-secretase-1 的配体结合任务中，通过在 3D 笛卡尔坐标中旋转已知的活性配体并将旋转后的配体分配为非活动数据来实现数据增强，3D 图形卷积网络的预测性能大大增强。
京公网安备 11010802027423号