Molecular Mechanisms and Function Prediction of Long Noncoding RNA

Table 1

Machine-learning methods for identifying lncRNAs.

Method	Features	Algorithm	References
	Peptide length
	Amino acid composition
	Hydrophobicity
CONC	Secondary structure content	SVM	[18]
	Percentage of residues exposed to solvent
	Sequence compositional entropy
	Number of homologs obtained by PSI-BLAST
	Alignment entropy
	ORF prediction quality
CPC	Number of homologs obtained by BLASTX	SVM	[19, 20]
	Alignment quality
	Segment distribution
Lu et al.	RNA-seq experiments; Tilling arrays; poly-A + RNA-seq experiments; poly-A + tilling arrays; GC content; DNA conservation; Predicted protein sequence conservation; Predicted secondary structure free energy; Predicted secondary structure conservation	Naïve Bayes; Bayes Net; Decision Tree; Random Forest; Logistic Regression; SVM	[21]