Identifying protein complexes based on node embeddings obtained from protein-protein interaction networks

BMC Bioinformatics

Table 9 Performance comparison using different vector generation strategies on HPRD and DIP datasets

Methods	No. of complexes	Precision	Recall	F-score
HPRD
Max	871	0.7107	0.3983	0.5105
Min	854	0.7037	0.3824	0.4956
Average	937	0.6126	0.354	0.4487
DIP
Max	710	0.8070	0.4473	0.5755
Min	701	0.8160	0.4368	0.5690
Average	698	0.8181	0.4354	0.5683

Bold value denotes the best score corresponding to F-score. Max denotes selecting the max value of each column of the matrix Z which is composed by the corresponding node embeddings in the complex. Min denotes selecting the min value of each column of the matrix Z. Average denotes getting the average value of each column of the matrix Z

ISSN: 1471-2105