Using Support Vector Machine and Evolutionary Profiles to Predict Antifreeze Protein Sequences

Xiaowei Zhao,Minghao Yin,Zhiqiang Ma

doi:10.3390/ijms13022196

Xiaowei Zhao, Minghao Yin + Show 1 more

Open Access

https://doi.org/10.3390/ijms13022196

Copy DOI

Abstract

Antifreeze proteins (AFPs) are ice-binding proteins. Accurate identification of new AFPs is important in understanding ice-protein interactions and creating novel ice-binding domains in other proteins. In this paper, an accurate method, called AFP_PSSM, has been developed for predicting antifreeze proteins using a support vector machine (SVM) and position specific scoring matrix (PSSM) profiles. This is the first study in which evolutionary information in the form of PSSM profiles has been successfully used for predicting antifreeze proteins. Tested by 10-fold cross validation and independent test, the accuracy of the proposed method reaches 82.67% for the training dataset and 93.01% for the testing dataset, respectively. These results indicate that our predictor is a useful tool for predicting antifreeze proteins. A web server (AFP_PSSM) that implements the proposed predictor is freely available.

Highlights

Antifreeze proteins (AFPs) are functional proteins in a cell
After a preliminary evaluation of different encoding schemes, we found that the evolutionary information in the form of position specific scoring matrix (PSSM) profiles is suitable for representing the antifreeze protein sequence
Four support vector machine (SVM) models based on amino acids composition, dipeptides composition, Chou’s pseudo amino acid composition (PseAAC) and PSSM-400 are constructed respectively

Summary

Introduction

Antifreeze proteins (AFPs) are functional proteins in a cell. With special antifreeze activity, AFPs make the organisms less sensitive to cold temperatures. AFPs bind to small ice crystals to inhibit growth and recrystallization of ice that would otherwise be fatal [1]. Relational analyses show that there is low sequence or structure similarity for an ice-binding domain, and lack of common features among different AFPs [7,8,9,10]. One reason for this phenomenon is that ice can present many different surfaces with different arrangements of oxygen atoms [8]. How to discriminate AFPs from other proteins is important in understanding protein-ice interactions and creating new ice-binding domains in other proteins

Methods

Results

Conclusion