Algorithm

Sequence encoding

Before using the sequence dataset as input to machine learning classifiers, they need to be mapped onto numeric vectors. Here, compositions of k-mer frequencies were used to transform the sequences into numeric form. In particular, combination of frequencies for k-mer size 1, 2, 3 and 4 were used by which each sequence was converted to a 340-dimentional numeric vector.

SVM predictor

For prediction purpose, we used support vector machine with radial kernel. Actually the prediction was made in two stages by constructing two classifiers: (i) a binary classifier was constructed for classification of resistant and non-resistant categories, and (ii) a multiclass classifier for categorization of seven classes of resistant categories. Initially, the test sequences were predicted as resistant or non-resistant types and then, the sequences predicted as resistant types were subjected to the second phase, where they were classified into any one of the seven resistant categories of target sites. In the first stage of prediction the SVM was trained with resistant and non-resistant sequences, whereas in the second stage only the seven classes of resistant sequences were used to train the SVM prediction model.

Our other Prediction servers

• dSSPred
• MaLDoSS
• PreDOSS
• HSplice
• SPIDBAR
• DCDNC
• iAMPpred
• DIRprot

Our recent publications

Meher PK , Sahu TK , Banchariya A, Rao AR (2017). DIRProt: A computational approach for discriminating insecticide resistant proteins from non-resistant proteins. BMC Bioinformatics, 18: 190.

Meher PK , Sahu TK , Saini V, and Rao AR (2017). Predicting antimicrobial peptides with improved accuracy by incorporating the compositional, physico-chemical and structural features into Chou’s general PseAAC. Scientific Reports, 7: 42362.

Meher PK , Sahu TK , Rao AR, Wahi, SD(2016). Discriminating coding from non-coding regions based on codon structure and methylation-mediated substitution: An application in rice and cattle. Computers and Electronics in Agriculture, 129: 66–73.

Meher PK , Sahu TK , Rao AR(2016). Prediction of donor splice sites using random forest with a new sequence encoding approach. BioData mining, 9:4.

Meher PK , Sahu TK , Rao AR, Wahi, SD(2016). Identification of donor splice sites using support vector machine: a computational approach based on positional, compositional and dependency features. Algorithms for Molecular Biology, 11:16.

Meher PK , Sahu TK , Rao AR, Wahi, SD(2016). A computational approach for prediction of donor splice sites with improved accuracy. Journal of Theoretical Biology, 404: 285–294.

Meher PK , Sahu TK , Rao AR(2016). Identification of species based on DNA barcode using k-mer feature vector and Random forest classifier. Gene, 592: 316–324.

Meher PK , Sahu TK , Rao AR, Wahi, SD(2015). Determination of window size and identification of suitable method for prediction of donor splice sites in rice (Oryza sativa) genome. Journal of Plant Biochemistry and Biotechnology, 24(4): 385-392.

Meher PK , Sahu TK , Rao AR, Wahi, SD(2014). A statistical approach for 5’ splice site prediction using short sequence motifs and without encoding sequence data. BMC Bioinformatics, 15: 362.