Pawsey expertise accelerated genome assembly for early COVID-19 detection

Researchers using Pawsey supercomputing and expertise have recently demonstrated a fast, sensitive diagnostic test for SARS-CoV-2.  The new approach can reliably detect lower concentrations of the virus than most existing tests and provides more genetic information about the novel coronavirus for researchers trying to trace its development and spread and create a vaccine.

The automated analysis pipeline developed was beta tested at Pawsey towards providing a one-click diagnostic report containing the test result (positive or negative), the SARS-CoV-2 genome identified, as well as any other viral genome assemblies also detected, such as the local flu.

Supercomputing specialist Dr Basha Mehaboob at the Pawsey Centre worked with the team towards optimising the workflow of the algorithm to maximise its performance to align the genetic fragments in each sample and compare them against reference genomes created from the original Wuhan outbreak, reducing the computational time to under one minute.

Early detection of infection, specifically at or before symptom onset when carriers are most likely to be contagious, is key to managing the current global pandemic, so the development of a more sensitive test that can detect very low viral loads is crucial.

Scientists from DNA Zoo Australia and the University of Western Australia’s Faculty of Science are part of a global team that has developed the new test.  The new Pathogen-Oriented Low-cost Assembly and Resequencing (POLAR) test can reliably detect 84 genome-equivalents per millilitre, more sensitive than nearly all diagnostic tests for SARS-CoV-2 currently approved by the US Food and Drug Administration (FDA).

Associate Professor Parwinder Kaur, the Australian project lead for POLAR, said because the test was able to detect such low viral concentrations compared to other diagnostic tests, it could even be used to monitor COVID-19 through wastewater treatment plants to track its spread through communities.

POLAR is currently being used for COVID-19 research and is seeking EUA approvals from US FDA for diagnostic testing for SARS-CoV-2.  Once approved, it will provide a fast, sensitive test for the wider community.

This preliminary SARS-CoV-2 testing approach has been published on bioRxiv, and the open-source analysis pipeline including documentation and test set is publicly available at https://github.com/aidenlab/Polar

The analysis pipeline can be run efficiently on a wide range of high-performance computing platforms.