Events Calendar

30 Mar
Biostatistics Seminar: Neural Network Models for Sequence-Based TCR and HLA Association Prediction
Event Type

Lectures, Symposia, Etc.

Topic

Research

Target Audience

Undergraduate Students, Faculty, Graduate Students, Postdocs

University Unit
Department of Biostatistics
Subscribe
Google Calendar iCal Outlook

Biostatistics Seminar: Neural Network Models for Sequence-Based TCR and HLA Association Prediction

Wei Sun, PhD

Professor, Biostatistics Program
Public Health Sciences Division, Fred Hutch
 

ABSTRACT: T cell is a critical component of human immune system. A T cell relies on its T cell receptor to recognize foreign antigens presented by a human leukocyte antigen (HLA), which is the human version of major histocompatibility complex (MHC). HLA is the most polymorphic locus in human genome and there are thousands of major HLA alleles in human population. Most of the public TCRs that are shared across subjects are restricted to certain HLA alleles. Therefore it is important to study TCR-HLA associations to understand the impact of HLA alleles on human immune response. Currently, TCR-HLA associations are mainly assessed through co-occurrence patterns. In this work we explore the capacity of neural networks to predict the association between HLA and TCR, based on their amino acid sequences. Our model can make predictions on HLA and TCR that are not seen during training. The predictions of TCR-HLA associations allow us to quantify the functional similarities of HLA alleles, which can be used in many applications. In this paper, we  demonstrate one application to use such similarities are associated with survival outcome of cancer patients who received immune checkpoint clockade (ICB) treatment.

Thursday, March 30 at 4:00 p.m.

Public Health, G23
130 Desoto Street, Pittsburgh, 15261

Biostatistics Seminar: Neural Network Models for Sequence-Based TCR and HLA Association Prediction

Wei Sun, PhD

Professor, Biostatistics Program
Public Health Sciences Division, Fred Hutch
 

ABSTRACT: T cell is a critical component of human immune system. A T cell relies on its T cell receptor to recognize foreign antigens presented by a human leukocyte antigen (HLA), which is the human version of major histocompatibility complex (MHC). HLA is the most polymorphic locus in human genome and there are thousands of major HLA alleles in human population. Most of the public TCRs that are shared across subjects are restricted to certain HLA alleles. Therefore it is important to study TCR-HLA associations to understand the impact of HLA alleles on human immune response. Currently, TCR-HLA associations are mainly assessed through co-occurrence patterns. In this work we explore the capacity of neural networks to predict the association between HLA and TCR, based on their amino acid sequences. Our model can make predictions on HLA and TCR that are not seen during training. The predictions of TCR-HLA associations allow us to quantify the functional similarities of HLA alleles, which can be used in many applications. In this paper, we  demonstrate one application to use such similarities are associated with survival outcome of cancer patients who received immune checkpoint clockade (ICB) treatment.

Thursday, March 30 at 4:00 p.m.

Public Health, G23
130 Desoto Street, Pittsburgh, 15261

Topic

Research

University Unit
Department of Biostatistics

Powered by the Localist Community Events Calendar ©