Predicting Protein Secondary Structure Using Consensus Data Mining (CDM) Based on Empirical Statistics and Evolutionary Information

Abstract

Predicting the secondary structure of a protein from its sequence still remains a challenging problem. The prediction accuracies remain around 80 %, and for very diverse methods. Using evolutionary information and machine learning algorithms in particular has had the most impact. In this chapter, we will first define secondary structures, then we will review the Consensus Data Mining (CDM) technique based on the robust GOR algorithm and Fragment Database Mining (FDM) approach. GOR V is an empirical method utilizing a sliding window approach to model the secondary structural elements of a protein by making use of generalized evolutionary information. FDM uses data mining from experimental structure fragments, and is able to successfully predict the secondary structure of a protein by combining experimentally determined structural fragments based on sequence similarities of the fragments. The CDM method combines predictions from GOR V and FDM in a hierarchical manner to produce consensus predictions for secondary structure. In other words, if sequence fragment are not available, then it uses GOR V to make the secondary structure prediction. The online server of CDM is available at http://gor.bb.iastate.edu/cdm/ .

Noninvasive Bioluminescence Imaging of Matrix Metalloproteinase-14 Activity in Lung Cancer Using a Membrane-Bound Biosensor

“…To ensure that CP Gluc could be successfully reconstituted, three different structurally tolerant Gluc sites were tested for the insertion of the MMP-14 substrate site according to the secondary structure prediction of GLuc using the CDM algorithm . The three cut sites were at 79, Gly93, and 106, which were all within the regions of unstructured Gluc proteins sequence (amino acids 65–109, Figure B).…”

“…To ensure that CP Gluc could be successfully reconstituted, three different structurally tolerant Gluc sites were tested for the insertion of the MMP-14 substrate site according to the secondary structure prediction of GLuc using the CDM algorithm. 18 The three cut sites were at 79, Gly93, and 106, which were all within the regions of unstructured Gluc proteins sequence (amino acids 65−109, Figure 1B). Accordingly, these three genetically encoded variants of MMP-14 biosensors, designated as MMP-CP79, MMP-CP93, and MMP-CP106, were constructed with linkage of the MMP substrate sequence (Figure 1C).…”

Prediction of Protein Secondary Structure based on Multi-scale Convolutional Neural Network

“…In the 21st century, various machine learning methods, especially artificial neural networks, have been used to improve performance, such as support vector machines [3], cyclic neural networks [4] (RNN), conditional neural field combination [5] and other probabilistic graphical models have been widely used. In recent years, big data, deep learning methods and other technologies have been widely applied, and it has become a research trend to predict the secondary structure of proteins combined with deep learning models.…”

Computational Methods for Predicting Functions at the mRNA Isoform Level

“…Previously, machine learning methods have been used to address a multitude of problems, some of which include, drug target discovery, gene function prediction, protein–protein interaction (PPI) prediction, protein structure and functional site prediction, and subcellular localization protein prediction [ 21 , 22 , 23 , 24 , 25 , 26 , 27 ]. More recently, several machine learning and recommendation system methods have also been developed to predict the biological functions of mRNA isoforms [ 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 ].…”