A tumor-driven pathway prediction system from collective cellular mutation data and protein networks

Abstract

The present disclosure proposes a tumor-driven pathway prediction system for collective cell mutation data and protein networks, including: a weighted undirected network graph building module configured to: construct a weighted undirected network graph using vertex weighting and edge weighting methods; weighted directed network A graph building block, configured to: perform a restart random walk on , using the principle of thermal diffusion, to construct a weighted directed graph; an initial candidate module set building block, configured to: use the strong connectivity principle of a directed network graph to create a graph with certain The initial candidate module set of the number of genes; the optimal driving module set building module, which is configured as: splitting the large modules in the initial candidate module set using the derived subgraph method, and using the greedy strategy to expand the small modules to obtain the optimal Driver module set. The constructed biological network not only integrates the biological correlation between genes, but also reflects the topological correlation between genes. Apply complex network topology features to biological networks.

Description

Technical field [0001] This disclosure belongs to the field of medical and biological data processing technology, especially the tumor -driven pathway prediction system that involves collective cell mutation data and protein networks. Background technique [0002] The statement of this part is only to provide background technical information related to this disclosure, and it is not necessarily composed of first technology. [0003] In order to explore the production and pathogenesis of cancer, the research of cancer at the molecular level has been further promoted. Large -scale sequencing data such as TCGA and ICGC of cancer genomes have emerged, and cancer genome data has been continuously improved. Researchers found that the differences between the genetic data of normal cells and cancer cells found that although there are thousands of gene mutations in a cancer cell, not every gene mutation will affect its cell function. How to detect a functional high -frequency mutation gen...

AI Technical Summary

Problems solved by technology

[0008] In order to overcome the deficiencies of the above-mentioned existing technologies, the present disclosure provides a tumor-driven pathway prediction system based on collective cell mutation data and protein networks, which not only combines the mutual exclusivity and coverage among genes widely present in genome maps, but also utilizes complex network nodes Inter-similarity, effectively avoiding the problem of mutation heterogeneity, improving the accuracy and completeness of oncogenic module mining

Images