A genetic feature mining method for depressive disorders based on multi-network fusion and multi-layer network diffusion

Abstract

The invention discloses a depressive disorder gene feature mining method based on multi-network fusion and multi-layer network diffusion. The mining method mainly includes the following steps: 1. Construct a multi-type gene association network; 2. Construct a standardized multi-layer gene network; 3. Stimulate the diffusion dynamics process of the multi-layer gene network driven by the depressive disorder gene; 4. Mining the diffusion dynamics of the multi-layer gene network driven by the depressive disorder gene. This mining method can effectively integrate different types of biomolecular networks, and mine effective disease gene characteristics from various diverse biomolecular networks, so as to identify depression-related genes more effectively.

Description

technical field [0001] The invention belongs to the field of bioinformatics analysis, and relates to a depressive disorder gene feature mining method based on multi-network fusion and multi-layer network diffusion. Background technique [0002] Depression has become a common disease in today's society, seriously affecting the quality of life of patients. Depression is closely related to the dysfunction of related genes. However, traditional methods such as genome-wide association studies are often difficult to accurately map depression genes. Due to the high cost and long time period of biomedical experiments, it is very important for the mechanism research, prevention, diagnosis and treatment of depression disorders to identify the genes related to depression disorders by developing computational methods to mine the genetic characteristics of depression disorders. [0003] With the rapid accumulation of biomolecular network data such as protein interaction networks, gene fe...

AI Technical Summary

Problems solved by technology

The causative genes of the same disease, due to their functional correlation, tend to gather in biomolecular networks, so algorithms based on network propagation become an effective strategy, for example, random walk with restart, random walk in heterogeneous networks walk, multiple network random walk, etc.; based on the rapid development of modern high-throughput experimental technology, the rapid growth of various types of biomolecular network data, the effective use of these biomolecular network data will help to more effectively mine depression Obstacle gene characteristics; the traditional algorithm based on aggregated network random walk can reduce the impact of network incompleteness, but it may lose the structural characteristics of individual networks; the ranking aggregation method of independent data sources utilizes the characteristics of individual networks, but ignores The relationship between different types of networks / layers; the multi-graph framework considers different types of networks, but the interaction between network layers is too strong

Heterogeneous networks help to integrate multi-source heterogeneous association data, but how to extract useful information from heterogeneous networks to deal with specific diseases is still a challenge; therefore, how to effectively integrate these different types of biomolecular networks, how to extract It is still an important issue to be studied to mine effective genetic characteristics of depression and other diseases in diverse biomolecular networks, so as to more effectively identify genes related to depression

Images