Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Apr 15;12(4):798.
doi: 10.3390/microorganisms12040798.

Machine Learning to Identify Critical Biomarker Profiles in New SARS-CoV-2 Variants

Christoph Schatz  1   2 , Ludwig Knabl  1 , Hye Kyung Lee  3 , Rita Seeboeck  4   5 , Dorothee von Laer  6 , Eliott Lafon  7 , Wegene Borena  6 , Harald Mangge  8 , Florian Prüller  8 , Adelina Qerimi  2 , Doris Wilflingseder  7   9 , Wilfried Posch  7 , Johannes Haybaeck  10   11   12   13
Affiliations

Machine Learning to Identify Critical Biomarker Profiles in New SARS-CoV-2 Variants

Christoph Schatz et al. Microorganisms. .

Abstract

The global dissemination of SARS-CoV-2 resulted in the emergence of several variants, including Alpha, Alpha + E484K, Beta, and Omicron. Our research integrated the study of eukaryotic translation factors and fundamental components in general protein synthesis with the analysis of SARS-CoV-2 variants and vaccination status. Utilizing statistical methods, we successfully differentiated between variants in infected individuals and, to a lesser extent, between vaccinated and non-vaccinated infected individuals, relying on the expression profiles of translation factors. Additionally, our investigation identified common causal relationships among the translation factors, shedding light on the interplay between SARS-CoV-2 variants and the host's translation machinery.

Keywords: Alpha; Alpha + E484K; Beta; F1 score; Omicron; PC algorithm; Restricted Boltzmann Machine neural network; SARS-CoV-2; machine learning; precision; recall; vaccination state; variants; z-scores.

PubMed Disclaimer

Conflict of interest statement

Authors Christoph Schatz and Ludwig Knabl were employed by the company Tyrolpath Obrist Brunhuber GmbH. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
Graph: directed acyclic graph (DAG) shows a defined directed flow (lines and arrows) between the investigated genes. Variables were tested for conditional independence based on log2fold changes using pooled SARS-CoV-2 variants that led to directions between variables. Each gene expression (circle with gene symbol) was tested against the others for conditional independence. Found relations indicate a flow from one gene to another, as indicated by an arrowhead towards the target variable. No connected line represents no found connection between the variables, and arrowheads both from the next variable to the previous variable and from the previous variable to the next variable indicate an unmeasured confound.
Figure 2
Figure 2
Z-scores of log2fold changes in translation factors and mTOR between samples of unvaccinated versus vaccinated samples (median datapoint), with all variants pooled together. The bars show how much the expression of the vaccinated group differed from the expression of the unvaccinated group for each investigated gene. Light blue bars indicate that the z-score was between −1 and lower than −2 for the gene, indicating a standardized lower expression of the vaccinated group for the gene. The dark blue bar indicates a z-score higher than −1, showing that the expression of the vaccinated group was lower than −2 compared with the unvaccinated group for the gene. The light red bar indicates that the expression (z-score higher than 1 and lower than 2) for the gene was higher in the vaccinated group.
Figure 3
Figure 3
Precision against recall for the translation factors and mTOR, grouped by comparisons of Beta versus ALPHA, Beta versus ALPHA and E484K, Beta versus OMICRON, and unvaccinated versus vaccinated. For each sample of the compared groups, higher values indicate that the expression of the unvaccinated group was more often higher than the vaccinated group, respectively, and that the expression of the Beta group was more often higher than the expression of each other group (ALPHA, E48K, or OMICRON) for the investigated genes. The figure shows the precision and recall per gene for (A) Beta–OMICRON, (B) Beta–ALPHAE484K, (C) Beta–ALPHA and (D) unvaccinated–vaccinated.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Weiss S.R., Leibowitz J.L. Coronavirus Pathogenesis. Adv. Virus Res. 2011;81:85–164. doi: 10.1016/B978-0-12-385885-6.00009-2. - DOI - PMC - PubMed
    1. Tang Q., Song Y., Shi M., Cheng Y., Zhang W., Xia X.-Q. Inferring the Hosts of Coronavirus Using Dual Statistical Models Based on Nucleotide Composition. Sci. Rep. 2015;5:17155. doi: 10.1038/srep17155. - DOI - PMC - PubMed
    1. Weiss S.R., Navas-Martin S. Coronavirus Pathogenesis and the Emerging Pathogen Severe Acute Respiratory Syndrome Coronavirus. Microbiol. Mol. Biol. Rev. 2005;69:635–664. doi: 10.1128/MMBR.69.4.635-664.2005. - DOI - PMC - PubMed
    1. Su S., Wong G., Shi W., Liu J., Lai A.C.K., Zhou J., Liu W., Bi Y., Gao G.F. Epidemiology, Genetic Recombination, and Pathogenesis of Coronaviruses. Trends Microbiol. 2016;24:490–502. doi: 10.1016/j.tim.2016.03.003. - DOI - PMC - PubMed
    1. Mechanisms of SARS-CoV-2 Entry into Cells|Nature Reviews Molecular Cell Biology. [(accessed on 27 August 2023)]. Available online: https://www.nature.com/articles/s41580-021-00418-x. - PMC - PubMed

Grants and funding

This work was supported in part (H.K.L. and L.K.) by the Intramural Research Programs (IRPs) of the National Institute of Diabetes and Digestive and Kidney Diseases, USA.

LinkOut - more resources

-