HMCDA: a novel method based on the heterogeneous graph neural network and metapath for circRNA-disease associations prediction

Table 4 Different combination of metapaths

	c–d–c	d–c–d	c–m–c	d–m–d	d–d	c–m–d–m–c
metapath2	✓	✓
metapath3	✓	✓	✓
metapath4	✓	✓	✓	✓
metapath5	✓	✓	✓	✓	✓
metapath6	✓	✓	✓	✓	✓	✓

Metapath2: Combination of 2 metapaths: \(c-d-c, d-c-d\). Metapath3: Combination of 3 metapaths: \(c-d-c, d-c-d, c-m-c\), Metapath4: Combination of 4 metapaths: \(c-d-c, d-c-d, c-m-c, d-m-d\). Metapath5: Combination of 5 metapaths: \(c-d-c, d-c-d, c-m-c, d-m-d, d-d\). Metapath6: Combination of 6 metapaths: \(c-d-c, d-c-d, c-m-c, d-m-d, d-d, c-m-d-m-c\)

ISSN: 1471-2105