Predicting abiotic stress-responsive miRNA in plants based on multi-source features fusion and graph neural network

Table 7 Results of the ablation experiment with the associated number under five-fold cross-validation

Models	Known associations	AUPR	AUC
GIN	200	0.9443	0.8881
	300	0.9604	0.9208
	400	0.9665	0.9324
	600	0.9716	0.9548
	823	0.9824	0.9743
	200	0.9415	0.8852
	300	0.9424	0.8944
GraphSAGE	400	0.9504	0.9043
	600	0.9606	0.9356
	823	0.9726	0.9612
	200	0.8933	0.8138
	300	0.9010	0.8327
GCN	400	0.9135	0.8503
	600	0.9227	0.8839
	823	0.9494	0.9306
	200	0.8832	0.8256
	300	0.8919	0.8345
GAT	400	0.8910	0.8675
	600	0.9009	0.8937
	823	0.9136	0.9127

ISSN: 1746-4811