Prediction of Antimicrobial Peptides Based on Sequence Alignment and Support Vector Machine-Pairwise Algorithm Utilizing LZ-Complexity

Abstract

This study concerns an attempt to establish a new method for predicting antimicrobial peptides (AMPs) which are important to the immune system. Recently, researchers are interested in designing alternative drugs based on AMPs because they have found that a large number of bacterial strains have become resistant to available antibiotics. However, researchers have encountered obstacles in the AMPs designing process as experiments to extract AMPs from protein sequences are costly and require a long set-up time. Therefore, a computational tool for AMPs prediction is needed to resolve this problem. In this study, an integrated algorithm is newly introduced to predict AMPs by integrating sequence alignment and support vector machine- (SVM-) LZ complexity pairwise algorithm. It was observed that, when all sequences in the training set are used, the sensitivity of the proposed algorithm is 95.28% in jackknife test and 87.59% in independent test, while the sensitivity obtained for jackknife test and independent test is 88.74% and 78.70%, respectively, when only the sequences that has less than 70% similarity are used. Applying the proposed algorithm may allow researchers to effectively predict AMPs from unknown protein peptide sequences with higher sensitivity.

AmPEP: Sequence-based prediction of antimicrobial peptides using distribution patterns of amino acid properties and random forest

“…Furthermore, to the best of our knowledge, our work contributed to the construction of the largest and diverse AMP dataset for the purpose of machine learning model evaluation. Our positive data were curated after retrieval from three major databases (CAMP, APD3, and LAMP), whereas most of the earlier methods involve a single database: either CAMP or APD 5 , 20 , 21 . Because experimental negative data are scarce, the negative data here were generated from UniProt sequences.…”

Predicting antimicrobial peptides with improved accuracy by incorporating the compositional, physico-chemical and structural features into Chou’s general PseAAC

“…The predictive ability of SVM, mainly depends upon the type of kernel function that maps the input data to a high-dimensional feature space, where the observations belong to different classes are linearly separable by a optimal separating hyper plane. In this work, the radial basis function (RBF) was used as kernel, due to its wide and successful application in most of the AMP prediction studies 1 9 10 33 . Further, in RBF kernel, default values of parameters gamma (gamma = 1/number of attributes) and cost (C = 1) were used to train and test the prediction model.…”

Predicting cancerlectins by the optimal g-gap dipeptides

“…Support Vector Machine (SVM) is a kind of learning machine method based on statistical learning theory and has been widely used in the field of bioinformatics 54 55 56 57 58 59 60 61 . The basic idea of applying SVMs to pattern classification can be summarized as follows.…”