COVID-19 classifiers identified through Ultra-high-throughput clinical proteomics

Image

The current COVID-19 pandemic presents a challenge for clinical care to establish quick, cheap and reliable indicators of disease and disease severity.  A paper published in Cell Systems led by Prof. Markus Ralser (Charité Universitaetsmedizin Berlin and The Francis Crick Institute) involving CGEM research groups led by Prof David Porteous and Dr Riccardo Marioni, and utilising the Generation Scotland cohort, presents a technological breakthrough to this issue. 

Using an ultra-high-throughput Mass-spectrometry (MS)-based approach, the group were able to assess serum and plasma proteins that are potential biomarkers of health and disease. The low-cost method, combined with an advanced data analysis approach, called Machine Learning, is capable of quantifying 180 proteomes per day per spectrometer. The MS-based system does not depend on the use of affinity reagents, or prior knowledge of the disease, as required by antibody tests.

Using the platform the authors were able to identify 27 biomarkers of COVID-19 that classify mild and severe forms of the disease, some of which have potential as therapeutic targets. The proteins highlight the role of complement factors, the coagulation system, inflammation modulators as well as pro-inflammatory signalling upstream and downstream of Interleukin 6.  To enable the greatest impact and utility of this development the authors have made freely available all protocols and software for implementing the approach.

Large, inexpensive, and high quality proteomic studies will help to drive biomarker development and our understanding of disease. It’s terrific that pilot data from 200 Generation Scotland participants contributed to this study. We are now working with colleagues at the Crick to scale this up to the full cohort of 20,000 individuals.

Dr Riccardo Marioni

We are delighted that Generation Scotland were able to support this important study of COVID-19. All the world is looking for ways to predict how patients with the signs and symptoms of COVID-19 are likely to progress and how they react to experimental treatments. The added benefit of this approach is that it can be applied to any new virus outbreak in the future.

Professor David Porteous

Links

Read the paper in Cell Sytems (doi.org/10.1016/j.cels.2020.05.012)

Generation Scotland Website

Dr Riccardo Marioni's Research Group website