Towards a massive COVID-19 diagnostics strategy for supporting the deconfinement process

Marco Antonio Mendoza-Parra; PhD /HDR

UMR 8030 Genomique Metabolique, Genoscope, Institut François Jacob, CEA, CNRS, Evry, France

Background: One of the major bottle-necks for fighting against the current Covid-19 pandemia is the capacity of a country to count on a large number of diagnostic tests. Currently, the use of RT-qPCR assays-driven diagnostics remains the main tool to track the spread of the disease, but this strategy has shown its limitations, notably by the lack of reagents during the first weeks of major Covid-19 crisis in France and abroad. Indeed, the methodology on its own is a sequential process of molecular biology steps, thus requiring major laborious work per patient sample, hence becoming a limiting factor when aiming to perform large population diagnostics assays (e.g. including asymptomatic candidates).

Considering our current needs in further tracing the spread of the disease, notably at the stage of the progressive release of the confinement, herein I propose the use of a novel strategy, relying on the use of massive parallel DNA sequencing, for population-scale diagnostics. In fact, with people going back to work after confinement, recurrent diagnostics assays will be required to screen for potential infection taking place over time.

Strategy: The Currently proposed protocols ([1] & [2]) follow the same initial steps used for RT-qPCR diagnostics, namely the collection of nasopharyngeal swabs, on which a reverse transcription assay is performed for generating cDNA. The major difference here is the incorporation of a DNA molecular barcode within the RT primers, such that the patient source is labeled. When large amounts of patients-labeled cDNA samples are available, they are combined into a single reaction for generating libraries for massive parallel DNA sequencing. Next-generation sequencing assay is then performed to a sequencing coverage (i.e. number of sequenced molecules) in agreement to the number of combined patient samples, such that the sensitivity of the assay is not compromised. Finally, a bioinformatics pipeline is used for recovering patients’ associated sequenced reads and delivering their related diagnostics status. This proposed strategy has been validated by Howard M. Salis (Pennsylvania State University) for the analysis of up to 19 200 patients’ samples per run [1]. A further enhanced strategy – described by Feng. Zhang (Broad Institute), replaced the cDNA step by the Reverse-Transcription Loop-mediated Isothermal Amplification combined with massive parallel sequencing (LAMP-Seq) [2]. This last study also estimated the cost of this diagnostic strategy to less than 7 dollars per patient; thus, supporting this methodology at a competitive price, notably when required to test several thousands of patients.

Practical requirements: The proposed strategy fits within the regular practices of research centers in France and Europe, taking advantage of Massive parallel DNA sequencing. Furthermore, both described methodologies (Salis proposal and LAMP-seq) were made available with detailed protocols and bioinformatics pipelines. Keeping in mind the potential needs of "rapid tests" or the portability of the diagnostics solutions to regions where a massive parallel DNA sequencing platform is not available, the use of the Oxford Nanopore MinION solutions will be explored as part of this Hackathon.

Potential action Plan: Considering that all over France and Europe, the collection of swabs is still ongoing for classical RT-qPCR diagnostic assays, it would be essential to coordinate a first pilot project, in which collected samples could also join a massive parallel DNA sequencing assay, such that we could count with a first proof of principle done in Europe with the corresponding RT-qPCR readouts as gold standards. Following such outcome, we could quite easily expand the strategy to different research centers.


[1] A Massively Parallel COVID-19 Diagnostic Assay for Simultaneous Testing of 19200 Patient Samples; Ayaan Hossain1, Alexander C. Reis2, Sarthok Rahman5, and Howard M. Salis1-4

[2] LAMP-Seq: Population-Scale COVID-19 Diagnostics Using a Compressed Barcode Space; Jonathan L. Schmid-Burgk, David Li, David Feldman, Mikołaj Słabicki, Jacob Borrajo, Jonathan Strecker, Brian Cleary, Aviv Regev, Feng Zhang

Share this project: