Bioinformatics and Systems Biology: Integrative Analysis of Multi-Scale, Multi-Modality Data

Genetic Feature Selection Using Dimensionality Reduction Approaches

The recent decade has witnessed great advances in microarray and genotyping technologies which allow genome-wide single nucleotide polymorphism (SNP) data to be captured on a single chip. As a consequence, genome-wide association studies require the development of algorithms capable of manipulating ultra-large-scale SNP datasets. Two SNP selection methods were proposed towards this goal. The first is a filtering technique using Independent Component Analysis (ICA) and the second is a multivariate regression approach based on a modified version of Fast Orthogonal Search (mFOS). 

                                                    

Figure 1: The workflow of SNP selection using ICA          Figure 2: Flowchart of the application of mFOS for modeling the SNP values                                                                                                                                                                               

Integrative Analysis of Gene Expression in Prostate Cancer

Prostate is the most common non-dermatological cancer amongst men in the West. The diseases is highly curable if detected early; although with a considerable risk of side effects that decrease quality of life. Treatment is thus becoming highly individualized, placing emphasis on early detection and prediction of disease prognosis. The main concern of this project is the computational integration and analysis of gene expression datasets, derived from prostate cancer, with information about cancer stage and progression. Integrating gene expression data from multiple sources will provide more robust and accurate characterizations of gene expression signatures related to prostate cancer grade and progression as well as prediction of future outcomes.

integrated Complex Traits Network (iCTNet)

                     

integrated Complex Traits Network (iCTNet) is a large-scale network, assembling human disease-gene association, tissue-gene association, disease-tissue associations, protein-DNA interactions, protein-protein interactions and drug-target information. This network provides a new and comprehensive perspective for human genetic diseases. iCTNet works as a plugin for Cytoscape and can be extended and combined using other Cytoscape plugins.

Related Publications: 

Modeling of Gene Regulatory Networks Using Dynamic Bayesian Networks

Modeling of Gene Regulatory Networks Using Neural Networks

Related Publications: 

Predictive Models for Identification of Molecular Signatures in Disease

Subscribe to RSS - Bioinformatics and Systems Biology: Integrative Analysis of Multi-Scale, Multi-Modality Data