University of Hawai‘i at Manoa Homepage

Events

Upcoming Events

Date: Tuesday, June 13, 2017
Time: 12:00 Noon to 1:00 pm
Place: University of Hawaii Cancer Center 701 Ilalo Street, Sullivan Conference Room
Title: Creating a Cancer Disparities Network for Pacific Islanders: Lessons Learned and Future Directions
Details:
Sora Park Tanjasiri, DrPH, MPH
Professor, Department of Health
Director, Health Promotion Research Institute
Chair, Department of Health Science
California State University, Fullerton

Date: Monday, June 19, 2017
Time: 12:00 Noon to 1:00 pm
Place: University of Hawaii Cancer Center 701 Ilalo Street, Sullivan Conference Room
Title: Next-gen mHealth: Integrating Body Sensors With Smart Technology to Motivate Health Behavior Change
Details:
Susan M. Schembre, PhD, RD
Assistant Professor
Director, Bionutrition Research Core
Co-Founder and Co-Chair of Focus on Junior Faculty
Department of Health Science
UT MD Anderson Cancer Center

Date: Friday, June 30, 2017
Time: 1:15 pm
Place: University of Hawaii Cancer Center 701 Ilalo Street, Sullivan Conference Room
Title: Finding Novel Genetic Patterns in The Cancer Genome Atlas Using Contemporary Cyberinfrastructure
Details:
F. Alex Feltus, Clemson University Department of Genetics & Biochemistry
With the availability of larger datasets, networks (graphs) are increasingly used in the life-sciences to model systems-level relationships at the molecular level. However these networks can be computationally challenging to construct and be quite noisy affecting their impact on new discoveries. We will describe the use of Gaussian Mixture Models (GMMs) on a gene pair-wise basis during network construction as an approach to address the challenge of noisy networks. We will show that traditional methods underperform in network construction and that use of GMMs improves the quality of the network. Additionally, we demonstrate the ability to separate condition-specific edges (i.e. relationships that occur only in a subset of experimental conditions) within the network. To demonstrate the applicability of this method, we constructed gene co-expression networks using the Open Science Grid and Clemson Palmetto cluster using a RNAseq dataset containing five different cancers types obtained for The Cancer Genome Atlas. This network was parsed into tumor specific subnetworks a 22 gene cluster that may regulate GBM invasiveness and checkpoint immunotherapy targets.