Graph Neural Networks(GNNs)play a significant role in tasks related to homophilic graphs.Traditional GNNs,based on the assumption of homophily,employ low-pass filters for neighboring nodes to achieve information aggre...Graph Neural Networks(GNNs)play a significant role in tasks related to homophilic graphs.Traditional GNNs,based on the assumption of homophily,employ low-pass filters for neighboring nodes to achieve information aggregation and embedding.However,in heterophilic graphs,nodes from different categories often establish connections,while nodes of the same category are located further apart in the graph topology.This characteristic poses challenges to traditional GNNs,leading to issues of“distant node modeling deficiency”and“failure of the homophily assumption”.In response,this paper introduces the Spatial-Frequency domain Adaptive Heterophilic Graph Neural Networks(SFA-HGNN),which integrates adaptive embedding mechanisms for both spatial and frequency domains to address the aforementioned issues.Specifically,for the first problem,we propose the“Distant Spatial Embedding Module”,aiming to select and aggregate distant nodes through high-order randomwalk transition probabilities to enhance modeling capabilities.For the second issue,we design the“Proximal Frequency Domain Embedding Module”,constructing adaptive filters to separate high and low-frequency signals of nodes,and introduce frequency-domain guided attention mechanisms to fuse the relevant information,thereby reducing the noise introduced by the failure of the homophily assumption.We deploy the SFA-HGNN on six publicly available heterophilic networks,achieving state-of-the-art results in four of them.Furthermore,we elaborate on the hyperparameter selection mechanism and validate the performance of each module through experimentation,demonstrating a positive correlation between“node structural similarity”,“node attribute vector similarity”,and“node homophily”in heterophilic networks.展开更多
Cooperative multi-agent reinforcement learning( MARL) is an important topic in the field of artificial intelligence,in which distributed constraint optimization( DCOP) algorithms have been widely used to coordinate th...Cooperative multi-agent reinforcement learning( MARL) is an important topic in the field of artificial intelligence,in which distributed constraint optimization( DCOP) algorithms have been widely used to coordinate the actions of multiple agents. However,dense communication among agents affects the practicability of DCOP algorithms. In this paper,we propose a novel DCOP algorithm dealing with the previous DCOP algorithms' communication problem by reducing constraints.The contributions of this paper are primarily threefold:(1) It is proved that removing constraints can effectively reduce the communication burden of DCOP algorithms.(2) An criterion is provided to identify insignificant constraints whose elimination doesn't have a great impact on the performance of the whole system.(3) A constraint-reduced DCOP algorithm is proposed by adopting a variant of spectral clustering algorithm to detect and eliminate the insignificant constraints. Our algorithm reduces the communication burdern of the benchmark DCOP algorithm while keeping its overall performance unaffected. The performance of constraint-reduced DCOP algorithm is evaluated on four configurations of cooperative sensor networks. The effectiveness of communication reduction is also verified by comparisons between the constraint-reduced DCOP and the benchmark DCOP.展开更多
As an unsupervised learning method,stochastic competitive learning is commonly used for community detection in social network analysis.Compared with the traditional community detection algorithms,it has the advantage ...As an unsupervised learning method,stochastic competitive learning is commonly used for community detection in social network analysis.Compared with the traditional community detection algorithms,it has the advantage of realizing the time-series community detection by simulating the community formation process.In order to improve the accuracy and solve the problem that several parameters in stochastic competitive learning need to be pre-set,the author improves the algorithms and realizes improved stochastic competitive learning by particle position initialization,parameter optimization and particle domination ability self-adaptive.The experiment result shows that each improved method improves the accuracy of the algorithm,and the F1 score of the improved algorithm is 9.07%higher than that of original algorithm.展开更多
S-adenosyl-1-methionine(SAM)-dependent enzymes regulate various disease-related behaviors in all organisms.Recently,the leporin biosynthesis enzyme LepI,a SAM-dependent enzyme,was reported to catalyze pericyclic react...S-adenosyl-1-methionine(SAM)-dependent enzymes regulate various disease-related behaviors in all organisms.Recently,the leporin biosynthesis enzyme LepI,a SAM-dependent enzyme,was reported to catalyze pericyclic reactions in leporin biosynthesis;however,the mechanisms underlying LepI activation and catalysis remain unclear.This study aimed to investigate the molecular mechanisms of LepI.Here,we reported crystal structures of LepI bound to SAM/5′-deoxy-5′-(methylthio)adenosine(MTA),S-adenosyl-homocysteine(SAH),and SAM/substrate states.Structural and biochemical analysis revealed that MTA or SAH inhibited the enzyme activities,whereas SAM activated the enzyme.The analysis of the substrate-bound structure of LepI demonstrated that this enzymatic retro-Claisen rearrangement was primarily driven by three critical polar residues His133,Arg197,Arg295 around the active site and assisted by SAM with unclear mechanism.The present studies indicate that the unique mechanisms underlying regulatory and catalysis of the unusual SAM-dependent enzyme LepI,not only strengthening current understanding of the fundamentally biochemical catalysis,but also providing novel insights into the design of SAM-dependent enzyme-specific small molecules.展开更多
The Clustered Regularly Interspaced Short Palindromic Repeats(CRISPR)-CRISPR-associated(Cas) system is an adaptive immune system in bacteria and archaea that resists exogenous invasion through nucleic acid-mediated cl...The Clustered Regularly Interspaced Short Palindromic Repeats(CRISPR)-CRISPR-associated(Cas) system is an adaptive immune system in bacteria and archaea that resists exogenous invasion through nucleic acid-mediated cleavage. In the type III-A system, the Csm complex contains five effectors and a CRISPR RNA, which edits both single stranded RNA and double stranded DNA. It has recently been demonstrated that cyclic oligoadenylates(cOAs), which are synthesized by the Csm complex, act as second messengers that bind and activate Csm6. Here, we report the crystal structures of Staphylococcus epidermidis Csm3(SeCsm3) and an N-terminally truncated Csm6(SeCsm6 AN) at 2.26 and 2.0 , respectively. The structure of SeCsm3 highly resembled previously reported Csm3 structures from other species; however, it provided novel observations allowing further enzyme characterization. The homodimeric SeCsm6 AN folds into a compact structure. The dimerization of the HEPN domain leads to the formation of the ribonuclease active site, which is consistent with the reported Csm6 structures. Altogether, our studies provide a structural view of the ribonuclease activity mediated by Csm3 and Csm6 of the type Ⅲ-A CRISPR-Cas system.展开更多
基金supported by the Fundamental Research Funds for the Central Universities(Grant No.2022JKF02039).
文摘Graph Neural Networks(GNNs)play a significant role in tasks related to homophilic graphs.Traditional GNNs,based on the assumption of homophily,employ low-pass filters for neighboring nodes to achieve information aggregation and embedding.However,in heterophilic graphs,nodes from different categories often establish connections,while nodes of the same category are located further apart in the graph topology.This characteristic poses challenges to traditional GNNs,leading to issues of“distant node modeling deficiency”and“failure of the homophily assumption”.In response,this paper introduces the Spatial-Frequency domain Adaptive Heterophilic Graph Neural Networks(SFA-HGNN),which integrates adaptive embedding mechanisms for both spatial and frequency domains to address the aforementioned issues.Specifically,for the first problem,we propose the“Distant Spatial Embedding Module”,aiming to select and aggregate distant nodes through high-order randomwalk transition probabilities to enhance modeling capabilities.For the second issue,we design the“Proximal Frequency Domain Embedding Module”,constructing adaptive filters to separate high and low-frequency signals of nodes,and introduce frequency-domain guided attention mechanisms to fuse the relevant information,thereby reducing the noise introduced by the failure of the homophily assumption.We deploy the SFA-HGNN on six publicly available heterophilic networks,achieving state-of-the-art results in four of them.Furthermore,we elaborate on the hyperparameter selection mechanism and validate the performance of each module through experimentation,demonstrating a positive correlation between“node structural similarity”,“node attribute vector similarity”,and“node homophily”in heterophilic networks.
基金Supported by the National Social Science Foundation of China(15ZDA034,14BZZ028)Beijing Social Science Foundation(16JDGLA036)JKF Program of People’s Public Security University of China(2016JKF01318)
文摘Cooperative multi-agent reinforcement learning( MARL) is an important topic in the field of artificial intelligence,in which distributed constraint optimization( DCOP) algorithms have been widely used to coordinate the actions of multiple agents. However,dense communication among agents affects the practicability of DCOP algorithms. In this paper,we propose a novel DCOP algorithm dealing with the previous DCOP algorithms' communication problem by reducing constraints.The contributions of this paper are primarily threefold:(1) It is proved that removing constraints can effectively reduce the communication burden of DCOP algorithms.(2) An criterion is provided to identify insignificant constraints whose elimination doesn't have a great impact on the performance of the whole system.(3) A constraint-reduced DCOP algorithm is proposed by adopting a variant of spectral clustering algorithm to detect and eliminate the insignificant constraints. Our algorithm reduces the communication burdern of the benchmark DCOP algorithm while keeping its overall performance unaffected. The performance of constraint-reduced DCOP algorithm is evaluated on four configurations of cooperative sensor networks. The effectiveness of communication reduction is also verified by comparisons between the constraint-reduced DCOP and the benchmark DCOP.
基金This research was funded by National Natural Science Foundation of China(Grant No.2017YFC0820100)。
文摘As an unsupervised learning method,stochastic competitive learning is commonly used for community detection in social network analysis.Compared with the traditional community detection algorithms,it has the advantage of realizing the time-series community detection by simulating the community formation process.In order to improve the accuracy and solve the problem that several parameters in stochastic competitive learning need to be pre-set,the author improves the algorithms and realizes improved stochastic competitive learning by particle position initialization,parameter optimization and particle domination ability self-adaptive.The experiment result shows that each improved method improves the accuracy of the algorithm,and the F1 score of the improved algorithm is 9.07%higher than that of original algorithm.
基金We gratefully acknowledge the financial support from the National Natural Science Foundation of China(Grant No.31570842,No.31870836,and No.21702141)the National Young Thousand Talents Program and the Sichuan Province Thousand Talents program in China.
文摘S-adenosyl-1-methionine(SAM)-dependent enzymes regulate various disease-related behaviors in all organisms.Recently,the leporin biosynthesis enzyme LepI,a SAM-dependent enzyme,was reported to catalyze pericyclic reactions in leporin biosynthesis;however,the mechanisms underlying LepI activation and catalysis remain unclear.This study aimed to investigate the molecular mechanisms of LepI.Here,we reported crystal structures of LepI bound to SAM/5′-deoxy-5′-(methylthio)adenosine(MTA),S-adenosyl-homocysteine(SAH),and SAM/substrate states.Structural and biochemical analysis revealed that MTA or SAH inhibited the enzyme activities,whereas SAM activated the enzyme.The analysis of the substrate-bound structure of LepI demonstrated that this enzymatic retro-Claisen rearrangement was primarily driven by three critical polar residues His133,Arg197,Arg295 around the active site and assisted by SAM with unclear mechanism.The present studies indicate that the unique mechanisms underlying regulatory and catalysis of the unusual SAM-dependent enzyme LepI,not only strengthening current understanding of the fundamentally biochemical catalysis,but also providing novel insights into the design of SAM-dependent enzyme-specific small molecules.
基金supported by the National Natural Science Foundation of China(31570842 to W.C.)the National Young Thousand Talents Programthe Sichuan Province Thousand Talents Program in China
文摘The Clustered Regularly Interspaced Short Palindromic Repeats(CRISPR)-CRISPR-associated(Cas) system is an adaptive immune system in bacteria and archaea that resists exogenous invasion through nucleic acid-mediated cleavage. In the type III-A system, the Csm complex contains five effectors and a CRISPR RNA, which edits both single stranded RNA and double stranded DNA. It has recently been demonstrated that cyclic oligoadenylates(cOAs), which are synthesized by the Csm complex, act as second messengers that bind and activate Csm6. Here, we report the crystal structures of Staphylococcus epidermidis Csm3(SeCsm3) and an N-terminally truncated Csm6(SeCsm6 AN) at 2.26 and 2.0 , respectively. The structure of SeCsm3 highly resembled previously reported Csm3 structures from other species; however, it provided novel observations allowing further enzyme characterization. The homodimeric SeCsm6 AN folds into a compact structure. The dimerization of the HEPN domain leads to the formation of the ribonuclease active site, which is consistent with the reported Csm6 structures. Altogether, our studies provide a structural view of the ribonuclease activity mediated by Csm3 and Csm6 of the type Ⅲ-A CRISPR-Cas system.