Majallah-i Dānishgāh-i ̒Ulūm-i Pizishkī-i Qum (Oct 2015)
Designing Prediction Models to Determine the Structure of Stearoyl-Acyl Carrier Protein Desaturase 1
Abstract
Background and Objectives: Essential fatty acids, such as α-linolenic acid (ALA), as an omega-3 fatty acid, and linoleic acid (LA), as an omega-6 fatty acid cannot be synthesized by human cells. Therefore, they should be provided from food sources, such as fish, soybean oil, flaxseed, and sunflower seeds. The process of α-linolenic acid synthesis is performed by 6 enzymes, one of which is stearoyl-acyl-carrier-protein-desaturase (S-ACP-DES). This study was carried out aiming at designing prediction models to determine the structure of stearoyl-acyl carrier protein desaturase 1. Methods:Some bioinformatics tools were used to determine the most important gene features of S-ACP-DES enzyme in order to develop some organism prediction models. Then, data mining techniques, such as feature selection, decision tree, and classification models, were used to develop efficient and accurate prediction algorithms based on gene features of enzyme (S-ACP-DES) from various organisms. Results: The most important gene feature in the identification of organism structures of the enzyme (S-ACP-DES) was frequency of length. Also, the designed prediction models indicated that Naive Bayes model with FCdb criteria and accuracy of 97.83% can predict new organism enzymes based on the gene features. The above findings have been reported for the first time in this study. Conclusion:The results of this study showed that with the use of bioinformatics tools, we can classify DESA1 enzyme based on its organism, and the variable of the length of genes is the best indicator for this classification. Also, the best learning machine model in Naive Bayes was Bayesian with high accuracy of 38.97% for determination of DESA1 enzyme, which has been reported for the first time in this study.
