End-to-End Data Automation for Pooled Sample SARS-CoV-2 Using R and Other Open-Source Tools

Abstract

Background: Due to supply chain shortages of reagents for real-time (RT)-PCR for SARS-CoV-2 and increasing demand on technical staff, an end-to-end data automation strategy for SARS-CoV-2 sample pooling and singleton analysis became necessary in the summer of 2020. Methods: Using entirely open source software tools - Linux, bash, R, RShiny, ShinyProxy, and Docker - we developed a modular software application stack to manage the preanalytical, analytical, and postanalytical processes for singleton and pooled testing in a 5-week time frame. Results: Pooling was operationalized for 81 days, during which time 64 pooled runs were performed for a total of 5320 sample pools and approximately 21 280 patient samples in 4:1 format. A total of 17 580 negative pooled results were released in bulk. After pooling was discontinued, the application stack was used for singleton analysis and modified to release all viral RT-PCR results from our laboratory. To date, 236 109 samples have been processed avoiding over 610 000 transcriptions. Conclusions: We present an end-to-end data automation strategy connecting 11 devices, one network attached storage, 2 Linux servers, and the laboratory information system.