. 2021 Nov 18;12:773882. doi: 10.3389/fgene.2021.773882

TABLE 7.

Five methods to identify Ψ sites are summarized in all aspects.

Method	Feature extraction	Number of features	Classifiers
iRNA-PseU	PseKNC	${\begin{matrix} H . s a p i e n s 84 \\ M . m u s c u l u s 84 \\ S . c e r e v i s i a e 124 \end{matrix}$	SVM
PseUI	NC + DC + pseDNC + PSNP + PSDP	${\begin{matrix} H . s a p i e n s 1045 \\ M . m u s c u l u s 1045 \\ S . c e r e v i s i a e 1526 \end{matrix}$	SVM
XG-PseU	One-hot + TNC + NCP + ND + DNC	${\begin{matrix} H . s a p i e n s 1848 \\ M . m u s c u l u s 1848 \\ S . c e r e v i s i a e 2728 \end{matrix}$	XGBoost
EnsemPseU	Kmer + Binary + ENAC + NCP + ND	>1700	SVM + XGBoost + NB + KNN + RF
PseUdeep	One-hot + PSNP + KNFP	${\begin{matrix} H . s a p i e n s 109 \\ M . m u s c u l u s 109 \\ S . c e r e v i s i a e 119 \end{matrix}$	Deep learning network