A Review on Recent Computational Methods for Predicting Noncoding RNAs

Table 2

De novo ncRNA function prediction methods using RNA sequence features.

Name	URL	Feature	Prediction algorithm
RNAcon [27]	http://crdd.osdd.net/raghava/rnacon/	3-mer of nucleotides	SVM with parameters and kernels optimized by model training
CNCI [28]	https://github.com/www-bioinfo-org/CNCI	Frequency of adjoining nucleotide triplets (6-mer), the length and -score of most-like CDS, length-percentage, score-distance, and codon-bias	SVM using the standard radial basis function kernel
PLEK [29]	https://sourceforge.net/projects/plek/	Normalized frequencies of 1–5 mers of RNA sequences	SVM
CONC [30]		Peptide length, amino acid composition, nucleotide frequencies, predicted secondary structure content, predicted percentage of exposed residues, compositional entropy, number of homologs from database searches, and alignment entropy	SVM
CPC [31]	http://cpc.cbi.pku.edu.cn/	The longest reading frame in the three forward frames, log-odds score, coverage of the predicted ORF, and integrity of the predicted ORF	SVM