COVID19 Severity

Ádám Nagy, Department of Bioinformatics, Semmelweis University, Budapest, Hungary
Balázs Ligeti, Faculty of Information Technology and Bionics, Pázmány Péter Catholic University, Budapest, Hungary
János Szebeni, Semmelweis University Dept. of Nanomedicine, Budapest
Sándor Pongor, Faculty of Information Technology and Bionics, Pázmány Péter Catholic University, Budapest, Hungary
Balázs Győrffy, Department of Bioinformatics, Semmelweis University, Budapest, Hungary
doi: https://doi.org/10.1101/2021.04.01.438063

Abstract

Numerous studies demonstrate frequent mutations in the genome of SARS-CoV-2. Our goal was to statistically link mutations to severe disease outcome. We used an automated machine learning approach where 1594 viral genomes with available clinical follow-up data were used as the training set (797 ‘severe’ and 797 ‘mild’). The best algorithm, based on random forest classification combined with the LASSO feature selection algorithm, was employed to the training set to link mutation signatures and outcome. The performance of the final model was estimated by repeated, stratified, 10-fold cross validation (CV) and then adjusted for multiple testing with Bootstrap Bias Corrected CV. We identified 26 protein and Untranslated Region (UTR) mutations significantly linked to severe outcome. The best classification algorithm uses a mutation signature of 22 mutations as well as the patient’s age as the input and shows high classification efficiency with an area under the curve (AUC) of 0.94 [confidence interval (CI): [0.912, 0.962]] and a prediction accuracy of 87% (CI: [0.830, 0.903]). Finally, we established an online platform (https://covidoutcome.com/) that is capable to use a viral sequence and the patient’s age as the input and provides a percentage estimation of disease severity. We demonstrate a statistical association between mutation signatures of SARS-CoV-2 and severe outcome of COVID-19. The established analysis platform enables a real-time analysis of new viral genomes.
Read more: COVIDOUTCOME – Estimating COVID Severity Based on Mutation Signatures in the SARS-CoV-2 Genome

How was AutoML used?

A statistical link between SARS-Cov-2 mutation status and severe COVID outcome was established using automated machine learning techniques(more specifically JADBio) based on random forest and logistic regression combined with feature selection algorithms.

A mutation signature based on 3,779 protein coding and 36 UTR mutations capable to identify severe outcome cases was established.

The trained model showed high classification performance (AUC=0.94 (CI: [0.912, 0.962]), accuracy=0.87 (CI: [0.830, 0.903])).

A registration-free web-server for automated classification of new samples was set up and is accessible at www.covidoutcome.com.

The established pipeline provides a quick assessment of future patients warranting a prospective clinical validation.