3D Deep CNN for Amino Acid Environment Similarity Analysis

Central to protein biology is the understanding of how structural elements give rise to observed function. The surfeit of protein structural data enables development of computational methods to systematically derive rules governing structural-functional relationships. However, performance of these methods depends critically on the choice of protein structural representation. Most current methods rely on features that are manually selected based on knowledge about protein structures. These are often general-purposed but not optimized for the specific problem of interest. In this project, we develop a general framework that applies 3D convolutional neural network (3DCNN) technology to structure-based protein analysis. The framework automatically extracts task-specific features from the raw atom distribution, driven by supervised labels. As a pilot study, we use our network to analyze local protein microenvironments surrounding the 20 amino acids, and predict the amino acids most compatible with environments within a protein structure. To further validate the power of our method, we construct two amino acid substitution matrices from the prediction statistics and use them to predict effects of mutations in T4 lysozyme structures. <br/><br/>This project includes the following software/data packages: <br/> <ul> <li> <a href="https://simtk.org/frs?group_id=1244#pack_1931">Datasets </a> </li> <li> <a href="https://simtk.org/frs?group_id=1244#pack_1932">Models </a> </li> </ul>

蛋白质生物学研究的核心在于理解结构元素如何引致观察到的功能。蛋白质结构数据的丰富性为开发计算方法，以系统性地推导出调控结构-功能关系的规则提供了可能。然而，这些方法的表现力在很大程度上取决于蛋白质结构表示的选择。大多数现有方法依赖于基于对蛋白质结构知识的人工选定的特征，这些特征通常为通用目的而设计，但并未针对特定问题进行优化。在本项目中，我们开发了一个通用框架，该框架将三维卷积神经网络（3DCNN）技术应用于基于结构的蛋白质分析。该框架能够从原始原子分布中自动提取特定任务的特性，并由监督标签驱动。作为先行研究，我们利用我们的网络分析围绕20种氨基酸的局部蛋白质微环境，并预测与蛋白质结构内部环境最为兼容的氨基酸。为了进一步验证我们方法的力量，我们构建了两个基于预测统计的氨基酸替换矩阵，并利用它们预测T4溶菌酶结构中突变的影响。本项目包括以下软件/数据包：

<ul>
<li><a href="https://simtk.org/frs?group_id=1244#pack_1931">数据集</a></li>
<li><a href="https://simtk.org/frs?group_id=1244#pack_1932">模型</a></li>
</ul>