Abstract

= 104) and developed a predictive model for COVID-19 using partial least squares discriminant analysis. While high sensitivity of 93.48% was achieved through leave-one-out cross-validation, COVID-19 patients testing negative on follow-up on the day of saliva sampling using RT-PCR was poorly predicted in this model. Importantly, COVID-19 vaccination did not lead to the misclassification of COVID-19 negatives. Finally, meta-analysis revealed that SARS-CoV-2 induced increases in the amide II band in all arms of this study and in recently published cohort studies, indicative of altered ß-sheet structures in secreted proteins. In conclusion, this study reveals a consistent secretory pathophysiological response to SARS-CoV-2, as well as a simple, robust method for COVID-19 saliva screening using ATR-FTIR.

Authors	Kazmer, Seth T; Hartel, Gunter; Robinson, Harley; Richards, Renee S; Yan, Kexin; van Hal, Sebastiaan J; Chan, Raymond; Hind, Andrew; Bradley, David; Zieschang, Fabian; Rawle, Daniel J; Le, Thuy T; Reid, David W; Suhrbier, Andreas; Hill, Michelle M
Journal	Biomedicines
Pages
Volume	10
Date	1/01/2022
Grant ID
Funding Body
URL	http://www.ncbi.nlm.nih.gov/pubmed/?term=10.3390/biomedicines10020351