Accurate identification of mutation-driven cancer driver genes is vital for cancer biology and targeted therapy.To address noise in protein–protein interaction(PPI)networks,we propose DGCL_DWA,a novel method utilizin...Accurate identification of mutation-driven cancer driver genes is vital for cancer biology and targeted therapy.To address noise in protein–protein interaction(PPI)networks,we propose DGCL_DWA,a novel method utilizing graph diffusion and contrastive learning.DGCL_DWA first employs personalized PageRank to generate a diffusion graph,revealing hidden biological connections.Chebyshev graph convolution extracts features from both the PPI and diffusion networks,and neighborhood contrastive learning harmonizes gene representations,reducing noise.The network-specific features are refined via Chebyshev graph convolutions,which are constrained via node classification and link prediction.A dynamic weight adjustment strategy balances task-specific losses during training.Finally,logistic regression is used to predict driver genes.The experimental results demonstrate the superior performance in pan-cancer and specific cancer driver gene identification compared with state-of-the-art methods.Ablation studies confirm the positive impact of the diffusion graph,contrastive learning,and dynamic weight adjustment on predictive accuracy.The source codes are available at https://doi.org/10.57760/sciencedb.31933.展开更多
Cellular networks are overloaded due to the mobile traffic surge,and mobile social networks(MSNets) can be leveraged for traffic offloading.In this paper,we study the issue of choosing seed users for maximizing the mo...Cellular networks are overloaded due to the mobile traffic surge,and mobile social networks(MSNets) can be leveraged for traffic offloading.In this paper,we study the issue of choosing seed users for maximizing the mobile traffic offloaded from cellular networks.We introduce a gossip-style social cascade(GSC) model to model the epidemic-like information diffusion process in MSNets.For static-case and mobile-case networks,we establish an equivalent view and a temporal mapping of the information diffusion process,respectively.We further prove the submodularity in the information diffusion and propose a greedy algorithm to choose the seed users for traffic offloading,yielding a sub-optimal solution to the NP-hard traffic offloading maximization(TOM) problem.Experiments are carried out to study the offloading performance,illustrating that the greedy algorithm significantly outperforms the heuristic and random algorithms,and user mobility can help further reduce cellular load.展开更多
基金supported in part by the National Natural Science Foundation of China under Grant Nos.62472202 and 61972185the Yunnan Ten Thousand Talents Plan for Young Professionals.
文摘Accurate identification of mutation-driven cancer driver genes is vital for cancer biology and targeted therapy.To address noise in protein–protein interaction(PPI)networks,we propose DGCL_DWA,a novel method utilizing graph diffusion and contrastive learning.DGCL_DWA first employs personalized PageRank to generate a diffusion graph,revealing hidden biological connections.Chebyshev graph convolution extracts features from both the PPI and diffusion networks,and neighborhood contrastive learning harmonizes gene representations,reducing noise.The network-specific features are refined via Chebyshev graph convolutions,which are constrained via node classification and link prediction.A dynamic weight adjustment strategy balances task-specific losses during training.Finally,logistic regression is used to predict driver genes.The experimental results demonstrate the superior performance in pan-cancer and specific cancer driver gene identification compared with state-of-the-art methods.Ablation studies confirm the positive impact of the diffusion graph,contrastive learning,and dynamic weight adjustment on predictive accuracy.The source codes are available at https://doi.org/10.57760/sciencedb.31933.
基金supported by the National Basic Research Program of China(973 Program) through grant 2012CB316004the Doctoral Program of Higher Education(SRFDP)+1 种基金Research Grants Council Earmarked Research Grants(RGC ERG) Joint Research Scheme through Specialized Research Fund 20133402140001National Natural Science Foundation of China through grant 61379003
文摘Cellular networks are overloaded due to the mobile traffic surge,and mobile social networks(MSNets) can be leveraged for traffic offloading.In this paper,we study the issue of choosing seed users for maximizing the mobile traffic offloaded from cellular networks.We introduce a gossip-style social cascade(GSC) model to model the epidemic-like information diffusion process in MSNets.For static-case and mobile-case networks,we establish an equivalent view and a temporal mapping of the information diffusion process,respectively.We further prove the submodularity in the information diffusion and propose a greedy algorithm to choose the seed users for traffic offloading,yielding a sub-optimal solution to the NP-hard traffic offloading maximization(TOM) problem.Experiments are carried out to study the offloading performance,illustrating that the greedy algorithm significantly outperforms the heuristic and random algorithms,and user mobility can help further reduce cellular load.
