A Review on Recent Computational Methods for Predicting Noncoding RNAs

Table 1

Homology-based ncRNA function prediction methods.

Name	URL	Feature	Prediction algorithm
BLAST [18]	https://blast.ncbi.nlm.nih.gov/Blast.cgi	Sequence only	BLAST -value
BLAT [21]	https://genome.ucsc.edu/cgi-bin/hgBlat	Sequence only	Pairwise alignment algorithm
CSHMM [38]		Structure only	A discriminant function based on likelihood score for a hidden Markov model (CSHMM)
Infernal [20, 24]	http://infernal.janelia.org/	Sequence and RNA secondary structure	Stochastic context-free grammars called covariance models (CMs), HMM
ERPIN [39]	http://rna.igmors.u-psud.fr/Software/erpin.php	Sequence and RNA secondary structure	Profile-based dynamic programming algorithm and -value
QRNA [19]		Sequence only	Pair hidden Markov model
RNAz [23]	https://www.tbi.univie.ac.at/~wash/RNAz/	RNA secondary structure and thermodynamic stability	Support vector machine regression
Evofold [40]	https://github.com/bowhan/kent/blob/master/src/hg/makeDb/trackDb/drosophila/evofold.html	A log-odds score	Phylogenetic stochastic context-free grammars
MASTR [25]	http://mastr.binf.ku.dk/	Mutual information with gap penalty, six canonical base pairs, stacking of adjacent base pairs, and the score combining the log-likelihood of the alignment, a covariation term, and the base-pair probabilities	Sampling approach by Markov chain Monte Carlo in a simulated annealing framework