Related Experiment Video
Updated: Aug 5, 2025
08:08
qPCR Is a Sensitive and Rapid Method for Detection of Cytomegaloviral DNA in Formalin-fixed, Paraffin-embedded Biopsy Tissue
Published on: July 9, 2014
15.8K
Applying T-classifier, binary classifiers, upon high-throughput TCR sequencing output to identify cytomegalovirus
Kaiyue Zhou1, Jiaxin Huo1, Caixia Gao1
1Department of Mathematics, School of Mathematical Sciences, Inner Mongolia University, Hohhot, China.
Scientific Reports
|March 28, 2023
Summary
This study introduces a novel method for detecting Cytomegalovirus (CMV) infection by analyzing T cell receptor beta chain (TCRβ) sequencing data. This approach offers a potential new diagnostic tool for CMV and other viral infections.
Area of Science:
- Medical data analysis
- Immunoinformatics
- Virology
Background:
- Cytomegalovirus (CMV) is a widespread virus with a high infection rate in adults, often asymptomatic.
- Accurate detection of CMV infection is crucial for public health, despite challenges with current diagnostic methods.
- The increasing volume of medical data necessitates advanced analytical techniques, including artificial intelligence.
Purpose of the Study:
- To develop and evaluate a novel method for detecting CMV infection status using T cell receptor beta chain (TCRβ) high-throughput sequencing data.
- To compare the performance of different machine learning algorithms for classifying CMV infection status.
- To explore the potential of TCRβ sequencing as a diagnostic biomarker for viral infections.
Main Methods:
- Analysis of high-throughput TCRβ sequencing data from 640 subjects (cohort 1) and validation in cohort 2.
- Application of Fisher's exact test to identify TCRβ sequences correlated with CMV status.
- Development and comparison of four binary classification models: Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), and Linear Discriminant Analysis (LDA).
Main Results:
- Optimal classification models were identified based on performance metrics at specific Fisher's exact test thresholds.
- Logistic Regression achieved 87.5% sensitivity and 96.88% specificity at a threshold of 10^-5.
- Support Vector Machine demonstrated 85.42% sensitivity and 96.88% specificity at 10^-5, while Linear Discriminant Analysis showed 95.83% sensitivity and 90.63% specificity at 10^-4.
Conclusions:
- TCRβ sequencing analysis presents a promising new method for detecting CMV infection status.
- The study highlights the potential of machine learning algorithms in analyzing complex biological data for viral diagnostics.
- This approach may be extendable to the detection of other viral infections, including historical infections.
