Automated Reverse Transcription Polymerase Chain Reaction Data Analysis for Sars-CoV-2 Detection

Abstract

Background: The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic is a current public health concern. Rapid diagnosis is crucial, and reverse transcription polymerase chain reaction (RT-PCR) is presently the reference standard for SARS-CoV-2 detection. Objective: Automated RT-PCR analysis (ARPA) is a software designed to analyze RT-PCR data for SARSCoV-2 detection. ARPA loads the RT-PCR data, classifies each sample by assessing its amplification curve behavior, evaluates the experiment’s quality, and generates reports. Methods: ARPA was implemented in the R language and deployed as a Shiny application. We evaluated the performance of ARPA in 140 samples. The samples were manually classified and automatically analyzed using ARPA. Results: ARPA had a true-positive rate = 1, true-negative rate = 0.98, positive-predictive value = 0.95, and negative-predictive value = 1, with 36 samples correctly classified as positive, 100 samples correctly classified as negative, and two samples classified as positive even when labeled as negative by manual inspection. Two samples were labeled as invalid by ARPA and were not considered in the performance metrics calculation. Conclusions: ARPA is a sensitive and specific software that facilitates the analysis of RT-PCR data, and its implementation can reduce the time required in the diagnostic pipeline.

Keywords

• Severe acute respiratory syndrome coronavirus-2 detection. Reverse transcription polymerase chain reaction. Automatic analysis. Amplification curves.