Extrication of Survival Associated Genes of Glioblastoma Multiforme Patient through Statistical Modeling

Abstract

Glioblastoma multiforme (GBM) is an aggressive type of brain tumour that progresses quickly and has a poor survival. Survival analysis is a statistical method used to examine how long an event is likely to take place before occurring, such as death in biological organisms or failure in mechanical systems. This analysis is called as time to event analysis as it is used to determine how long it will be before a specific event of interest, like death or recurrence, happens. There are numerous statistical techniques that make it possible to estimate the overall survival (OS) of GBM patients, which is highly useful for developing targeted medicines to slow the growth of a disease. It also helps to determine factors contributing to the survival of the patient. Based on the distribution of relevant events, these statistical techniques use a variety of non-parametric, semi-parametric, and parametric approaches. From TCGA, genomic datasets of GBM patients and information on their survival are gathered for this investigation. Three statistical techniques to evaluate the patterns of event timings and investigate how much some genetic characteristics influence the risk of an event of interest, Kaplan Meier, Cox Regression Model, and Accelerated Failure Time Model are used. Significant genes that are connected to the GBM patient's survival are identified in study.