Identification of coenzyme-binding proteins with machine learning algorithms. (April 2019)

• Record Type: Journal Article
• Authors: Liu, Yong; Munteanu, Cristian R.; Kong, Zhiwei; Ran, Tao; Sahagún-Ruiz, Alfredo; He, Zhixiong; Zhou, Chuanshe; Tan, Zhiliang
• Is Part Of: Computational biology and chemistry. Volume 79(2019)
• Page Start: 185
• Page End: 192
• Publication Date: 2019-04
• Subjects: Coenzyme-binding -- Protein sequence -- Topological indices -- Classification model -- Random Forest; Chemistry -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; Biochemistry -- Data processing; Biology -- Data processing; Molecular biology -- Data processing; Periodicals; Electronic journals; 542.85
• DOI: 10.1016/j.compbiolchem.2019.01.014
• Languages: English
• ISSNs: 1476-9271

Abstract

Highlights: CoEnBPs play a vital role in fatty acid biosynthesis, gene regulation, and lipid synthesis, etc . In silico screening of a new peptide with coenzyme binding activity might be an efficient process. A dataset of 42 SGTI features for 2, 897 proteins (456 CoEnBPs) was used as inputs to construct classification models. Best model obtained with 3 SGTI features in RF model with 0.971 (AUROC), 91.7% (TP rate), and 7.6% (FP rate). ML models for predicting new CoEnBPs could be useful for future drug development or enzyme catalysis metabolism research. Abstract: The coenzyme-binding proteins play a vital role in the cellular metabolism processes, such as fatty acid biosynthesis, enzyme and gene regulation, lipid synthesis, particular vesicular traffic, and β-oxidation donation of acyl-CoA esters. Based on the theory of Star Graph Topological Indices (SGTIs) of protein primary sequences, we proposed a method to develop a first classification model for predicting protein with coenzyme-binding properties. To simulate the properties of coenzyme-binding proteins, we created a dataset containing 2897 proteins, among 456 proteins functioned as coenzyme-binding activity. The SGTIs of peptide sequence were calculated with Sequence to Star Network (S2SNet) application. We used the SGTIs as inputs to several classification techniques with a machine learning software - Weka. A Random Forest classifier based on 3 features of the embedded and non-embedded graphs was identified as the best predictive model for coenzyme-binding proteins. This model developed was with the true positive (TP) rate of 91.7%, false positive (FP) rate of 7.6%, and Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.971. The prediction of new coenzyme-binding activity proteins using this model could be useful for further drug development or enzyme metabolism researches.