LEARNING SPARSE MODELS FOR A DYNAMIC BAYESIAN NETWORK CLASSIFIER OF PROTEIN SECO

学习蛋白质 SECO 动态贝叶斯网络分类器的稀疏模型

• 批准号: 8365898
• 负责人: William Noble
• 金额: $ 2.14万
• 依托单位: UNIVERSITY OF WASHINGTON
• 依托单位国家: 美国
• 财政年份: 2011
• 资助国家: 美国
• 起止时间: 2011-09-01 至 2012-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8365898
• 关键词: Algorithms; Amino Acids; Binding; Biological; Biology; Data; Elements; Funding; Fungal Genome; Grant; Label; Learning; Modeling; National Center for Research Resources; Performance; Principal Investigator; Protein Databases; Proteins; Research; Research Infrastructure; Resources; Secondary Protein Structure; Simulate; Source; Structure; Testing; United States National Institutes of Health; computer based statistical methods; cost; insight; protein function; three dimensional structure

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. Primary support for the subproject and the subproject's principal investigator may have been provided by other sources, including other NIH sources. The Total Cost listed for the subproject likely represents the estimated amount of Center infrastructure utilized by the subproject, not direct funding provided by the NCRR grant to the subproject or subproject staff. Protein secondary structure prediction provides insight into protein function and is a valuable preliminary step for predicting the 3D structure of a protein. Dynamic Bayesian networks (DBNs) have been shown to provide state-of-the-art performance in secondary structure prediction. As the size of the protein database grows, it becomes feasible to use a richer model in an effort to capture subtle correlations among the amino acids and the predicted labels. In this context, it is beneficial to derive sparse models that discourage over-fitting and provide biological insight. Results: We introduce an algorithm for sparsifying the parameters of a DBN. Using this algorithm, we can automatically remove up to 80% of the parameters of a DBN while maintaining the same level of predictive accuracy. We also prove an upper bound for the test error difference between the sparse and fully dense models. Finally, we demonstrate, using simulated data, that the algorithm is able to recover true sparse structures with high accuracy, and using real data, that the sparse model identifies known correlation structure related to different classes of secondary structure elements.

这个子项目是许多利用资源的研究子项目之一 由NIH/NCRR资助的中心拨款提供。子项目的主要支持 而子项目的主要调查员可能是由其他来源提供的， 包括其它NIH来源。 列出的子项目总成本可能 代表子项目使用的中心基础设施的估计数量， 而不是由NCRR赠款提供给子项目或子项目工作人员的直接资金。 蛋白质二级结构预测提供了对蛋白质功能的深入了解，并且是预测蛋白质3D结构的有价值的初步步骤。 动态贝叶斯网络（DBN）已被证明在二级结构预测中提供最先进的性能。 随着蛋白质数据库规模的增长，使用更丰富的模型来捕获氨基酸和预测标签之间的微妙相关性变得可行。 在这种情况下，导出稀疏模型是有益的，该稀疏模型阻止过度拟合并提供生物学见解。 结果：我们介绍了一种算法稀疏的DBN的参数。使用这种算法，我们可以自动删除高达80%的DBN参数，同时保持相同的预测精度。 我们还证明了一个上界的测试误差之间的差异稀疏和完全密集的模型。 最后，我们证明，使用模拟数据，该算法是能够恢复真正的稀疏结构具有高精度，并使用真实的数据，稀疏模型识别已知的相关结构相关的不同类别的二级结构元素。