DISTEMA: distance map-based estimation of single protein model accuracy with attentive 2D convolutional neural network.
DISTEMA: distance map-based estimation of single protein model accuracy with attentive 2D convolutional neural network.
复制标题
DOI：
10.1186/s12859-022-04683-1
复制
发表时间：
2022-04-19
影响因子：
3
通讯作者：
Cheng, Jianlin
中科院分区：
文献类型：
--
作者：
Chen, Xiao;Cheng, Jianlin
蛋白质结构模型的准确性（质量）的估计对于蛋白质结构模型的预测和使用都很重要。深度学习方法已用于整合蛋白质结构特征以预测蛋白质模型的质量。残基间距离是预测蛋白质三级结构的关键信息，在预测蛋白质结构模型的质量方面具有很好的潜力。然而，很少有方法已经开发出充分利用预测的残基间距离图来估计单个蛋白质结构模型的准确性。我们开发了一个专注的2D卷积神经网络（CNN），它具有通道式的注意力，只需要从单个蛋白质模型计算的残基间距离图和从蛋白质序列预测的距离图之间的原始差异图作为输入来预测模型的质量。该网络包括多个卷积层、批量归一化层、密集层和挤压和激励块，并注意从原始输入中自动提取与蛋白质模型质量相关的特征，而不使用任何专家策划的特征。我们评估了DISTEMA在GDT-TS评分排名损失方面为CASP 13靶标选择最佳模型的能力。DISTEMA的排序损失为0.079，低于几种最先进的单模型质量评估方法。这项工作表明，将原始残基间距离信息与深度学习结合使用，可以相当好地预测蛋白质结构模型的质量。DISTEMA可免费访问https://github.com/jianlin-cheng/DISTEMA
Estimation of the accuracy (quality) of protein structural models is important for both prediction and use of protein structural models. Deep learning methods have been used to integrate protein structure features to predict the quality of protein models. Inter-residue distances are key information for predicting protein’s tertiary structures and therefore have good potentials to predict the quality of protein structural models. However, few methods have been developed to fully take advantage of predicted inter-residue distance maps to estimate the accuracy of a single protein structural model. We developed an attentive 2D convolutional neural network (CNN) with channel-wise attention to take only a raw difference map between the inter-residue distance map calculated from a single protein model and the distance map predicted from the protein sequence as input to predict the quality of the model. The network comprises multiple convolutional layers, batch normalization layers, dense layers, and Squeeze-and-Excitation blocks with attention to automatically extract features relevant to protein model quality from the raw input without using any expert-curated features. We evaluated DISTEMA’s capability of selecting the best models for CASP13 targets in terms of ranking loss of GDT-TS score. The ranking loss of DISTEMA is 0.079, lower than several state-of-the-art single-model quality assessment methods. This work demonstrates that using raw inter-residue distance information with deep learning can predict the quality of protein structural models reasonably well. DISTEMA is freely at https://github.com/jianlin-cheng/DISTEMA
登录
查看更多内容
影响因子：
3
作者：
Ray A;Lindahl E;Wallner B
通讯作者：
Wallner B
影响因子：
16.6
作者：
Hiranuma N;Park H;Baek M;Anishchenko I;Dauparas J;Baker D
通讯作者：
Baker D
影响因子：
4.6
作者：
Uziela K;Shu N;Wallner B;Elofsson A
通讯作者：
Elofsson A
影响因子：
5.8
作者：
Jing, Xiaoyang;Xu, Jinbo
通讯作者：
Xu, Jinbo
影响因子：
3
作者：
Cao R;Bhattacharya D;Hou J;Cheng J
通讯作者：
Cheng J