Logical Analysis of Data as a Predictor of Protein Secondary Structures
Protein secondary structure prediction problem has been known for almost a quarter of century. The idea to create a tool to help molecular biologists was the main reason to choose the new rule-based method—Logical Analysis of Data with its high accuracy in various field of life. The LAD produced overall percentage accuracy of correctly predicted residues Q 3 = 71,6%, and segment overlap measure SOV = 70,9% on the set consist of 126 proteins. The goal of the analysis described in this paper is to create a system that allows for receiving as the output the protein secondary structure, based on its primary structure being an input, and for finding rules responsible for this effect.
Key wordslogical analysis of data protein prediction protein secondary structure machine learning
Unable to display preview. Download preview PDF.