An Improved Approach for Rapidly Identifying Different Types of Gram-Negative Bacterial Secreted Proteins

• 作者: Fengjuan Liu Lezheng Yu Lixiao Du Yizhou Li
• 来源期刊: 自然科学期刊(英文)
• 学科: 医学
• 关键词: GRAM-NEGATIVE Bacteria SECRETED Protein Position-Specific SCORING Matrix Signal Peptide Support Vector Machine
• 年，卷（期）: zrkxqkyw_2018,（5）
• 页码范围: 168-177
• 页数: 10页
• 分类号: R73

摘要

Protein secretion plays an important role in bacterial lifestyles. In Gram-negative bacteria, a wide range of proteins are secreted to modulate the interactions of bacteria with their environments and other bacteria via various secretion systems. These proteins are essential for the virulence of bacteria, so it is crucial to study them for the pathogenesis of diseases and the development of drugs. Using amino acid composition (AAC), position-specific scoring matrix (PSSM) and N-terminal signal peptides, two different substitution models are firstly constructed to transform protein sequences into numerical vectors. Then, based on support vector machine (SVM) and the “one to one”?algorithm, a hybrid multi-classifier named SecretP v.2.2 is proposed to rapidly and accurately?distinguish different types of Gram-negative?bacterial secreted proteins. When performed on the same test set for a comparison with other methods, SecretP v.2.2 gets the highest total sensitivity of 93.60%. A public independent dataset is used to further test the power of SecretP v.2.2 for predicting NCSPs, it also yields satisfactory results.