abstract: Immunosignature is a technology that retrieves information from the immune system. The technology is based on microarrays with peptides chosen from random sequence space. My thesis focuses on improving the Immunosignature platform and using Immunosignatures to improve diagnosis for diseases. I first contributed to the optimization of the immunosignature platform by introducing scoring metrics to select optimal parameters, considering performance as well as practicality. Next, I primarily worked on identifying a signature shared across various pathogens that can distinguish them from the healthy population. I further retrieved consensus epitopes from the disease common signature and proposed that most pathogens could share the signature by studying the enrichment of the common signature in the pathogen proteomes. Following this, I worked on studying cancer samples from different stages and correlated the immune response with whether the epitope presented by tumor is similar to the pathogen proteome. An effective immune response is defined as an antibody titer increasing followed by decrease, suggesting elimination of the epitope. I found that an effective immune response usually correlates with epitopes that are more similar to pathogens. This suggests that the immune system might occupy a limited space and can be effective against only certain epitopes that have similarity with pathogens. I then participated in the attempt to solve the antibiotic resistance problem by developing a classification algorithm that can distinguish bacterial versus viral infection. This algorithm outperforms other currently available classification methods. Finally, I worked on the concept of deriving a single number to represent all the data on the immunosignature platform. This is in resemblance to the concept of temperature, which is an approximate measurement of whether an individual is healthy. The measure of Immune Entropy was found to work best as a single measurement to describe the immune system information derived from the immunosignature. Entropy is relatively invariant in healthy population, but shows significant differences when comparing healthy donors with patients either infected with a pathogen or have cancer. / Dissertation/Thesis / Doctoral Dissertation Molecular and Cellular Biology 2018
| Identifer | oai:union.ndltd.org:asu.edu/item:48458 |
| Date | January 2018 |
| Contributors | Wang, Lu (Author), Johnston, Stephen (Advisor), Stafford, Phillip (Committee member), Buetow, Kenneth (Committee member), McFadden, Grant (Committee member), Arizona State University (Publisher) |
| Source Sets | Arizona State University |
| Language | English |
| Detected Language | English |
| Type | Doctoral Dissertation |
| Format | 173 pages |
| Rights | http://rightsstatements.org/vocab/InC/1.0/, All Rights Reserved |
Page generated in 0.0019 seconds