Covid-19 automated diagnosis and risk assessment through Metabolomics and Machine-Learning

Delafiori, Jeany; Navarro, Luiz Cláudio; Siciliano, Rinaldo Focaccia; Melo, Gisely Cardoso de; Busanello, Estela Natacha Brandt; Nicolau, José Carlos; Sales, Geovana Manzan; Oliveira, Arthur Noin de; Val, Fernando; Oliveira, Diogo Noin de; Eguti, Adriana; Santos, Luiz Augusto dos; Dalçóquio, Talia Falcão; Bertolin, Adriádne Justi; Alonso, Jonas Bodini; Netto, Rebeca Linhares Abreu; Salsoso, Rocío; Baía-da-Silva, Djane Clarys; Sampaio, Vanderson Souza; Judice, Carla C.; Costa, Fábio Trindade Maranhão; Perroud, Maurício W.; Sabino, Éster Cerdeira; Lacerda, Marcus Vinícius Guimarães; Reis, Leonardo Oliveira; Fávaro, Wagner José; Rocha, Anderson; Catharino, Rodrigo Ramos

doi:10.1101/2020.07.24.20161828

Cited by 12 publications

(17 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our study, we compared the lung and systemic responses in association with the lung transcriptome and demonstrated a robust correlation between lipid mediators, neurotransmitters, and their receptors in SARS-CoV-2 infection. However, contrary to what has been suggested by previous studies 37, 38 , only small amounts of eicosanoids were found in the plasma of patients with severe/critical disease, with significant differences observed only in AA, 5-HETE, and 11-HETE. Interestingly, the levels of 5-HETE and 11-HETE were found to be remarkably higher in BAL, as well as AA in plasma, suggesting that AA and its metabolites mediate responses to Covid-19.…”

Section: Discussioncontrasting

confidence: 99%

Cholinergic and lipid mediators crosstalk in Covid-19 and the impact of glucocorticoid therapy

Pérez

Pimentel

Fuzo

et al. 2021

Preprint

View full text Add to dashboard Cite

Cytokine storms and hyperinflammation, potentially controlled by glucocorticoids, occur in COVID-19; the roles of lipid mediators and acetylcholine (ACh) and how glucocorticoid therapy affects their release in Covid-19 remain unclear. Blood and bronchoalveolar lavage (BAL) samples from SARS-CoV-2- and non-SARS-CoV-2-infected subjects were collected for metabolomic/lipidomic, cytokines, soluble CD14 (sCD14), and ACh, and CD14 and CD36-expressing monocyte/macrophage subpopulation analyses. Transcriptome reanalysis of pulmonary biopsies was performed by assessing coexpression, differential expression, and biological networks. Correlations of lipid mediators, sCD14, and ACh with glucocorticoid treatment were evaluated. This study enrolled 190 participants with Covid-19 at different disease stages, 13 hospitalized non-Covid-19 patients, and 39 healthy-participants. SARS-CoV-2 infection increased blood levels of arachidonic acid (AA), 5-HETE, 11-HETE, sCD14, and ACh but decreased monocyte CD14 and CD36 expression. 5-HETE, 11-HETE, cytokines, ACh, and neutrophils were higher in BAL than in circulation (fold-change for 5-HETE 389.0; 11-HETE 13.6; ACh 18.7, neutrophil 177.5, respectively). Only AA was higher in circulation than in BAL samples (fold-change 7.7). Results were considered significant at P<0.05, 95%CI. Transcriptome data revealed a unique gene expression profile associated with AA, 5-HETE, 11-HETE, ACh, and their receptors in Covid-19. Glucocorticoid treatment in severe/critical cases lowered ACh without impacting disease outcome. We first report that pulmonary inflammation and the worst outcomes in Covid-19 are associated with high levels of ACh and lipid mediators. Glucocorticoid therapy only reduced ACh, and we suggest that treatment may be started early, in combination with AA metabolism inhibitors, to better benefit severe/critical patients.

show abstract

Section: Discussioncontrasting

confidence: 99%

Cholinergic and lipid mediators crosstalk in Covid-19 and the impact of glucocorticoid therapy

Pérez

Pimentel

Fuzo

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…First, mass spectrometric data is preprocessed for ion annotation (intensity, width, resolution, and m / z ), alignment, normalization, and denoising. 17 Three partitions of data are segregated according to the best practices of ML, consisting of fitting (training and validation), test, and blind test partitions. The final classification results are reported using the blind test (see Figure 2 a).…”

Section: Experimental Sectionmentioning

confidence: 99%

“…The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.analchem.0c04497 . Detailed data preprocessing and ML modeling method; Tables with model validation with several algorithms, multiclass confusion matrix, and biomarkers elucidation data through mass spectrometry; Preprocessed mass spectrometry 17 data available at: ; Deidentified patient information will be made available from corresponding author upon request ( PDF ) …”

mentioning

confidence: 99%

Covid-19 Automated Diagnosis and Risk Assessment through Metabolomics and Machine Learning

et al. 2021

Self Cite

View full text Add to dashboard Cite

COVID-19 is still placing a heavy health and financial burden worldwide. Impairment in patient screening and risk management plays a fundamental role on how governments and authorities are directing resources, planning reopening, as well as sanitary countermeasures, especially in regions where poverty is a major component in the equation. An efficient diagnostic method must be highly accurate, while having a cost-effective profile. We combined a machine learning-based algorithm with mass spectrometry to create an expeditious platform that discriminate COVID-19 in plasma samples within minutes, while also providing tools for risk assessment, to assist healthcare professionals in patient management and decision-making. A cross-sectional study enrolled 815 patients (442 COVID-19, 350 controls and 23 COVID-19 suspicious) from three Brazilian epicenters from April to July 2020. We were able to elect and identify 19 molecules related to the disease’s pathophysiology and several discriminating features to patient’s health-related outcomes. The method applied for COVID-19 diagnosis showed specificity >96% and sensitivity >83%, and specificity >80% and sensitivity >85% during risk assessment, both from blinded data. Our method introduced a new approach for COVID-19 screening, providing the indirect detection of infection through metabolites and contextualizing the findings with the disease’s pathophysiology. The pairwise analysis of biomarkers brought robustness to the model developed using machine learning algorithms, transforming this screening approach in a tool with great potential for real-world application.

show abstract

“…Powered by the growing applications of high-resolution mass spectrometry (MS), metabolomics has become a key member of the omics toolkit in biomedical research. Multiple metabolomics studies have been recently conducted across the world to study COVID-19, revealing key metabolic dysregulations during the disease’s progression [ 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 ]. For instance, several amino acids have been observed to be positively correlated with the severity of COVID-19 as key indicators of clinical prognosis of the disease [ 8 , 9 , 12 , 13 , 14 ].…”

Section: Introductionmentioning

confidence: 99%

Comprehensive Meta-Analysis of COVID-19 Global Metabolomics Datasets

et al. 2021

View full text Add to dashboard Cite

The novel coronavirus SARS-CoV-2 has spread across the world since 2019, causing a global pandemic. The pathogenesis of the viral infection and the associated clinical presentations depend primarily on host factors such as age and immunity, rather than the viral load or its genetic variations. A growing number of omics studies have been conducted to characterize the host immune and metabolic responses underlying the disease progression. Meta-analyses of these datasets have great potential to identify robust molecular signatures to inform clinical care and to facilitate therapeutics development. In this study, we performed a comprehensive meta-analysis of publicly available global metabolomics datasets obtained from three countries (United States, China and Brazil). To overcome high heterogeneity inherent in these datasets, we have (a) implemented a computational pipeline to perform consistent raw spectra processing; (b) conducted meta-analyses at pathway levels instead of individual feature levels; and (c) performed visual data mining on consistent patterns of change between disease severities for individual studies. Our analyses have yielded several key metabolic signatures characterizing disease progression and clinical outcomes. Their biological interpretations were discussed within the context of the current literature. To the best of our knowledge, this is the first comprehensive meta-analysis of global metabolomics datasets of COVID-19.

show abstract

Covid-19 automated diagnosis and risk assessment through Metabolomics and Machine-Learning

Cited by 12 publications

References 24 publications

Cholinergic and lipid mediators crosstalk in Covid-19 and the impact of glucocorticoid therapy

Cholinergic and lipid mediators crosstalk in Covid-19 and the impact of glucocorticoid therapy

Covid-19 Automated Diagnosis and Risk Assessment through Metabolomics and Machine Learning

Comprehensive Meta-Analysis of COVID-19 Global Metabolomics Datasets

Contact Info

Product

Resources

About