Research On The Method Of Identifying Essential Proteins Based On Multi Data Source

The biological functions of essential proteins and the transportation between them are combined to complete various biochemical reactions in biological systems.Most dysfunction of biological systems and even death are caused by the absence of such proteins.Scholars from various fields have shown great interest in discovering essential proteins.At the same time,it is of great significance to identify the key proteins in the cell system,whether it is to explain the life activities in theory or the application in the production.The continuous efforts of scholars and advances in science and technology have accumulated a large amount of available data for biological information research.These rich data enable researchers to move from molecular level to system network level in vitial protein studies.Currently,there are two types of methods used to detect proteins in the network,depending on whether or not the biological attributes of proteins are taken into account.However,due to the variety of data types related to essentiality,how to fuse these data effectively to improve the detection effect of key proteins is further studied.Aiming at the above problems,this paper mainly obtains the following results on the basis of studying the topological structure,biological properties and protein detection algorithm on PPI network.(1)Based on the close links between various data and protein nodes in the network,this paper proposes a detection method based on ELM algorithm.This method not only considers the location of protein in the network,but also takes into account the biological properties of protein in cell system.(2)Considering the dependence of the existing methods of PPI network topology,based on the protein in the system dynamics and network edge of unreliability,building dynamic weighted PPI network is put forward.The construction process is divided into two steps: dynamic PPI network construction and network weight calculation.First of all,according to the protein gene expression data changes over time to judge each protein in the active state of time point in the network,combined with static PPI network and the activity of the protein to build dynamic PPI network.Then use the GO terms and edge aggregation coefficient to evaluate the reliability of the edge in the network,and use the value as the weight of the edge in the dynamic PPI network.(3)Finally,based on the dynamic weighted PPI network,DWE algorithm is proposed to detect the essential proteins in the network.In order to verify the effect of dynamic weighted PPI network and to evaluate the recognition efficiency of DWE method,two groups of experiments are designed.The experimental results show that the dynamic weighted PPI network can achieve better recognition effect,and the DWE method is outperform than several other existing prediction methods.