Related Experiment Video
Updated: Aug 12, 2025
08:26
Large-Scale SARS-CoV-2 Testing Utilizing Saliva and Transposition Sample Pooling
Published on: June 23, 2022
1.8K
Performance of methods for SARS-CoV-2 variant detection and abundance estimation within mixed population samples
Tunc Kayikcioglu1,2, Jasmine Amirzadegan1,3, Hugh Rand1
1Biostatistics and Bioinformatics Staff, Office of Analytics and Outreach, Center for Food Safety and Applied Nutrition, US Food and Drug Administration, College Park, MD, United States of America.
Peerj
|February 1, 2023
Summary
Accurate SARS-CoV-2 variant tracking in wastewater is crucial. Kallisto and Freyja showed the best performance in estimating variant abundance, though real-world data revealed significant method discrepancies.
Area of Science:
- Environmental microbiology
- Genomic epidemiology
- Bioinformatics
Background:
- Accurate identification and abundance estimation of SARS-CoV-2 (SC2) variants in mixed samples like wastewater are vital for public health surveillance.
- Evaluating the performance of SC2 variant composition estimators (VCEs) enhances confidence in public health decision-making.
Purpose of the Study:
- To introduce and compare a novel linear regression-based VCE against established methods.
- To assess the accuracy and agreement of different VCEs for SC2 variant abundance estimation in mixed samples.
Main Methods:
- Simulated DNA sequence datasets from Illumina and Oxford Nanopore Technologies (ONT) platforms were used to evaluate VCE performance.
- Empirical wastewater samples were analyzed using a custom NextFlow workflow (C-WAP).
- Performance was measured by relative root mean squared error (RRMSE) and concordance correlation coefficient (CCC).
Main Results:
- Kallisto demonstrated the highest accuracy (lowest RRMSE) on simulated data, followed closely by Freyja.
- Kallisto and Freyja showed the highest agreement (highest CCC) in their predictions.
- VCE performance was dependent on sequencing platform and amplicon panel; empirical data showed poor agreement among methods.
Conclusions:
- Understanding VCE performance differences is crucial for interpreting SC2 variant data from wastewater.
- This information aids public health officials in assessing the reliability of variant surveillance data for decision-making.
