posted on 2024-01-27, 04:40authored byChen Ye, Qi Wu, Shuxia Chen, Xuemei Zhang, Wenwen Xu, Yunzhi Wu, Youhua Zhang, Yi Yue
Additional file 1: Figure S1. Dynamic PPI Networks. Figure S2. Comparison with Centrality Methods. Figure S3. AUC and RP curves of ECDEP compared on S. cerevisiae (BioGRID) dataset. Figure S4. AUC and RP curves of ECDEP compared on S. cerevisiae (Krogan) dataset. Figure S5. AUC and RP curves of ECDEP compared on M. musculus dataset. Figure S6. AUC and RP curves of ECDEP compared on C. elegans dataset. Figure S7. AUC and RP curves of ECDEP compared on S. cerevisiae (DIP) dataset. Figure S8. Comparison with machine learning and deep learning methods. Figure S9. Ablation study of features in ECDEP across six datasets evaluated with AUC score. Figure S10. Process of detaching each snapshot. Figure S11. Evaluate the results of detaching each snapshot with F1, AUC, and AP scores. Figure S12. Comparison of information from static network and dynamic network. Figure S13. Generate the intersection set of ECDEP and EP-EDL methods. Figure S14. Comparison of ECDEP with RNN-based methods. Figure S15. Compare ECDEP with canonical Graph Convolutional Network (GCN). Table S1. Version and sources of databases. Table S2. Download links of methods for comparison. Table S3. Process of essential proteins for different species. Table S4. Process of gene expression profiles. Table S5. PPI network details for different species and datasets. Table S6. Environment, package, and version requirements. Table S7. Hyperparameter settings of ECDEP model. Table S8. Experiment on different selections of M. musculus essential protein.
Funding
the Anhui Provincial Department of Education University Natural Science Research Project the Open Fund of State Key Laboratory of Tea Plant Biology and Utilization