Back to search

University of Washington

Immunologic Correlates Analysis of RhCMV/SIV Vaccine Efficacy - Applying Machine Learning Techniques to Model Vaccine Elicited T Cell Responses

Abstract

dc:description.abstract

In the past 30 years, HIV vaccine studies on traditional CD8+ T cell-targeted HIV vaccines were frustrated by the ineffectiveness of mediating immediate vaccinal interception upon infection acquisition prior to the explosive viral amplification. As the most important lesson of past HIV vaccine researches, the first hours to days immediately after viral infection might be the only vulnerable time period for immunologic interceptions.[1, 2] With this regard, immunologists started a novel research on employing Cytomegelovirus (CMV) as vaccine vector in early 2000s, to exploit CMV vectors’ unique ability on eliciting and maintaining abundant functional T cell responses at all potential HIV infection sites.[3-6] Recent CMV-based vaccine research, demonstrated by Louis Picker and colleagues, with statistical support by Dr. Edlefsen, manifests a remarkable infection control and clearance on ~50% of HIV-acquired rhesus macaques (RM) vaccinated by Simian immunodeficiency virus (SIV) inserted rhesus cytomegalovirus (CMV) vaccine.[7, 8] This promising protection pattern motivates further immunologic correlates analysis on vaccine efficacy to investigate potential immunological mechanisms of the partial protection. As part of vaccine efficacy analysis, this project was conducted to inspect the relation between CD4+, CD8+ T cell responses elicited by vaccine and the unique protection outcome via interpretability attached machine learning techniques. Interpretability has been regarded as the driven feature of this immunologic correlates machine learning process. After stringent data screening and statistical modeling along with strategic informatics interpretation, I preliminarily identified two immunologic features which correlates with the protection pattern by potentially corresponding to formation of germinal centers that act as HIV virus’s shelters. Keywords: CMV-based SIV Vaccine, T Cell Immune Responses, Vaccine Efficacy, Immunologic Correlates Analysis, Machine Learning, Informatics Interpretation

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Song, Wenjun
Advisors dc:contributor.advisor
  • Edlefsen, Paul T.
  • Tarczy-Hornoch, Peter

Subjects

dc:subject × 10

Rights

dc:rights
Statement dc:rights
  • CC BY-SA
Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
http://hdl.handle.net/1773/43262

Chain of custody

source
Harvested from
University of Washington
Base URL
digital.lib.washington.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Song, Wenjun. Immunologic Correlates Analysis of RhCMV/SIV Vaccine Efficacy - Applying Machine Learning Techniques to Model Vaccine Elicited T Cell Responses. 2019. http://hdl.handle.net/1773/43262