Research On Probability Distribution Embedding Method Of Individual Specific Network For Mining Critical Signals Of System

The detection of critical signals in complex systems has been a hot topic of research.Recently,many scholars have applied critical signals to complex disease systems.Noise is often present in the collection of biological disease data,which makes data mining inaccurate.At this time,some traditional methods for mining critical point signals fail.In this paper,the threenode biological simulation system,time series data of liver cancer and brain tumor were studied based on the method of individual-specific network probability distribution embedding,and the critical signal of mutation was detected.The theoretical basis of this work is to transform the large noise data of the sample state of the original system into the small noise data of the sample probability distribution by the probability distribution embedded transform,and then establish an individual-specific network.It is found that the noise interference of the data can be well reduced,and the problem of less sample data is solved.The signal of the critical mutation was then detected based on the dynamic network biomarkers and found to be effective.The paper is based on the method of individual-specific network probability distribution embedding,mainly doing the following two aspects:(1)Firstly,the concept of critical point is described.For deterministic systems,critical signals can be detected by CSD.The critical signal of disease system is detected by dynamic network biomarkers.Based on the concept of the moment equation,the probability distribution embedding method and the individual-specific network method are further explained.Combined with the probability distribution embedding method and the individual-specific network method,the individual-specific network probability distribution is embedded into the dynamic network biomarker comprehensive index.(2)Apply the above method to the three-node bio-simulation system,liver cancer data and brain tumor data,and use MATLAB software to simulate the trend graph with time,and successfully detected the critical signal.In addition,according to the dynamic network markers of liver cancer and brain tumors,STRING was used to obtain the protein interaction network.Cytoscape was used to draw the flip diagram of the whole gene disc map and markers,and it was found to be consistent with the experimental results.Finally,the dynamic network biomarkers of liver cancer and brain tumors were analyzed by KEGG enrichment analysis and survival analysis.Some signal pathways were found and many of them were found to be closely related to the induction and deterioration of cancer.