EDLm6APred: ensemble deep learning approach for mRNA m6A site prediction

BMC Bioinformatics

Table 2 Evaluation results of single predictor and integrated predictor based on different species

Species	Classifiers	AUROC	MCC	ACC	Precision	Recall
Human	BiLSTM_One-hot	0.7810	0.4409	0.7159	0.7716	0.6095
	BiLSTM_Embedding	0.8504	0.5602	0.7739	0.8470	0.6661
	BiLSTM_Word2vec	0.8510	0.5497	0.7695	0.8361	0.6678
	EDLm⁶APred	0.8660	0.5819	0.7843	0.8617	0.6750
Mouse	BiLSTM_One-hot	0.7838	0.4354	0.7088	0.7901	0.5739
	BiLSTM_Embedding	0.8390	0.5394	0.7642	0.8296	0.6691
	BiLSTM_Word2vec	0.8464	0.5369	0.7604	0.8429	0.6442
	EDLm⁶APred	0.8588	0.5664	0.7754	0.8579	0.6639
Mix	BiLSTM_One-hot	0.8055	0.4758	0.7361	0.7687	0.6755
	BiLSTM_Embedding	0.8459	0.5670	0.7801	0.8313	0.7028
	BiLSTM_Word2vec	0.8463	0.5477	0.7707	0.8189	0.6952
	EDLm⁶APred	0.8605	0.5787	0.7862	0.8355	0.7128

The evaluation indexes with EDLm⁶ APred better than its any single predictor are show in bold

ISSN: 1471-2105