Satpathy, Sitanshu (2017-10). Probabilistic Analysis of Pair Wise Gene Interactions Using Support Vector Clustering. Master's Thesis. Thesis uri icon

abstract

  • The most difficult challenge in genetic epidemiology is to characterize the gene interactions that affect a complex disease. DNA microarray has made it easier for engineers to study gene expression profiles of numerous genes by describing the complete genomic activity, but extraction of useful data without losing information poses a major challenge. Various clustering algorithms have been applied to these microarray profiles to identify the gene interactions based on various factors such as a stimuli or genes affecting a disease. However, only a few of them have been applied to find the interactions between the genes in the same cluster. Several methods have been used to predict complex gene networks, and they have been largely successful, but it cannot be inferred that the pair of genes interact every time. Gene interactions can be affected by various environmental factors, stimuli, or inactivating genes. This thesis aims to address this challenge by proposing a method that provides a probabilistic analysis of the interaction between a pair of genes. The proposed method uses Support Vector Clustering to classify a pair of genes, and the clusters formed are used to analyze their interaction. The algorithm is tested using yeast microarray data. The results found are validated using biological literature surveys.

publication date

  • October 2017