Table 2. Open in new tab Performance... | Oxford Academic

Table 2.

Open in new tab

Performance comparison of RNALoc-LM and deep-learning baseline models using 5-fold CV.^a

RNA type	Deep-learning baseline model	ACC	Macro F1	Macro precision	Macro recall
lncRNA	RNA-FM + Transformer + MLP	0.649	0.491	0.713	0.551
	one-hot + TextCNN + Bi-LSTM + MLP	0.633	0.388	0.316	0.500
	word2vec + TextCNN + Bi-LSTM + MLP	0.633	0.388	0.316	0.500
	RNALoc-LM	0.677	0.593	0.596	0.613
miRNA	RNA-FM + Transformer + MLP	0.913	0.903	0.901	0.907
	one-hot + TextCNN + Bi-LSTM + MLP	0.878	0.860	0.869	0.854
	word2vec + TextCNN + Bi-LSTM + MLP	0.881	0.866	0.868	0.867
	RNALoc-LM	0.913	0.903	0.899	0.909
circRNA	RNA-FM + Transformer + MLP	0.795	0.787	0.792	0.785
	one-hot + TextCNN + Bi-LSTM + MLP	0.797	0.789	0.794	0.786
	word2vec + TextCNN + Bi-LSTM + MLP	0.802	0.795	0.799	0.793
	RNALoc-LM	0.804	0.797	0.802	0.794

RNA type	Deep-learning baseline model	ACC	Macro F1	Macro precision	Macro recall
lncRNA	RNA-FM + Transformer + MLP	0.649	0.491	0.713	0.551
	one-hot + TextCNN + Bi-LSTM + MLP	0.633	0.388	0.316	0.500
	word2vec + TextCNN + Bi-LSTM + MLP	0.633	0.388	0.316	0.500
	RNALoc-LM	0.677	0.593	0.596	0.613
miRNA	RNA-FM + Transformer + MLP	0.913	0.903	0.901	0.907
	one-hot + TextCNN + Bi-LSTM + MLP	0.878	0.860	0.869	0.854
	word2vec + TextCNN + Bi-LSTM + MLP	0.881	0.866	0.868	0.867
	RNALoc-LM	0.913	0.903	0.899	0.909
circRNA	RNA-FM + Transformer + MLP	0.795	0.787	0.792	0.785
	one-hot + TextCNN + Bi-LSTM + MLP	0.797	0.789	0.794	0.786
	word2vec + TextCNN + Bi-LSTM + MLP	0.802	0.795	0.799	0.793
	RNALoc-LM	0.804	0.797	0.802	0.794

The best performance values are highlighted in bold.

Table 2.

Open in new tab

Performance comparison of RNALoc-LM and deep-learning baseline models using 5-fold CV.^a

RNA type	Deep-learning baseline model	ACC	Macro F1	Macro precision	Macro recall
lncRNA	RNA-FM + Transformer + MLP	0.649	0.491	0.713	0.551
	one-hot + TextCNN + Bi-LSTM + MLP	0.633	0.388	0.316	0.500
	word2vec + TextCNN + Bi-LSTM + MLP	0.633	0.388	0.316	0.500
	RNALoc-LM	0.677	0.593	0.596	0.613
miRNA	RNA-FM + Transformer + MLP	0.913	0.903	0.901	0.907
	one-hot + TextCNN + Bi-LSTM + MLP	0.878	0.860	0.869	0.854
	word2vec + TextCNN + Bi-LSTM + MLP	0.881	0.866	0.868	0.867
	RNALoc-LM	0.913	0.903	0.899	0.909
circRNA	RNA-FM + Transformer + MLP	0.795	0.787	0.792	0.785
	one-hot + TextCNN + Bi-LSTM + MLP	0.797	0.789	0.794	0.786
	word2vec + TextCNN + Bi-LSTM + MLP	0.802	0.795	0.799	0.793
	RNALoc-LM	0.804	0.797	0.802	0.794

RNA type	Deep-learning baseline model	ACC	Macro F1	Macro precision	Macro recall
lncRNA	RNA-FM + Transformer + MLP	0.649	0.491	0.713	0.551
	one-hot + TextCNN + Bi-LSTM + MLP	0.633	0.388	0.316	0.500
	word2vec + TextCNN + Bi-LSTM + MLP	0.633	0.388	0.316	0.500
	RNALoc-LM	0.677	0.593	0.596	0.613
miRNA	RNA-FM + Transformer + MLP	0.913	0.903	0.901	0.907
	one-hot + TextCNN + Bi-LSTM + MLP	0.878	0.860	0.869	0.854
	word2vec + TextCNN + Bi-LSTM + MLP	0.881	0.866	0.868	0.867
	RNALoc-LM	0.913	0.903	0.899	0.909
circRNA	RNA-FM + Transformer + MLP	0.795	0.787	0.792	0.785
	one-hot + TextCNN + Bi-LSTM + MLP	0.797	0.789	0.794	0.786
	word2vec + TextCNN + Bi-LSTM + MLP	0.802	0.795	0.799	0.793
	RNALoc-LM	0.804	0.797	0.802	0.794

The best performance values are highlighted in bold.