A Comprehensive Machine Learning Approach for COVID-19 Target Discovery in the Small-Molecule Metabolome
<b>Background/Objectives:</b> Respiratory viruses, including Influenza, RSV, and COVID-19, cause various respiratory infections. Distinguishing these viruses relies on diagnostic methods such as PCR testing. Challenges stem from overlapping symptoms and the emergence of new strains. Adva...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Metabolites |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2218-1989/15/1/44 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587978232299520 |
|---|---|
| author | Md. Shaheenur Islam Sumon Md Sakib Abrar Hossain Haya Al-Sulaiti Hadi M. Yassine Muhammad E. H. Chowdhury |
| author_facet | Md. Shaheenur Islam Sumon Md Sakib Abrar Hossain Haya Al-Sulaiti Hadi M. Yassine Muhammad E. H. Chowdhury |
| author_sort | Md. Shaheenur Islam Sumon |
| collection | DOAJ |
| description | <b>Background/Objectives:</b> Respiratory viruses, including Influenza, RSV, and COVID-19, cause various respiratory infections. Distinguishing these viruses relies on diagnostic methods such as PCR testing. Challenges stem from overlapping symptoms and the emergence of new strains. Advanced diagnostics are crucial for accurate detection and effective management. This study leveraged nasopharyngeal metabolome data to predict respiratory virus scenarios including control vs. RSV, control vs. Influenza A, control vs. COVID-19, control vs. all respiratory viruses, and COVID-19 vs. Influenza A/RSV. <b>Method:</b> We proposed a stacking-based ensemble technique, integrating the top three best-performing ML models from the initial results to enhance prediction accuracy by leveraging the strengths of multiple base learners. Key techniques such as feature ranking, standard scaling, and SMOTE were used to address class imbalances, thus enhancing model robustness. SHAP analysis identified crucial metabolites influencing positive predictions, thereby providing valuable insights into diagnostic markers. <b>Results:</b> Our approach not only outperformed existing methods but also revealed top dominant features for predicting COVID-19, including Lysophosphatidylcholine acyl C18:2, Kynurenine, Phenylalanine, Valine, Tyrosine, and Aspartic Acid (Asp). <b>Conclusions:</b> This study demonstrates the effectiveness of leveraging nasopharyngeal metabolome data and stacking-based ensemble techniques for predicting respiratory virus scenarios. The proposed approach enhances prediction accuracy, provides insights into key diagnostic markers, and offers a robust framework for managing respiratory infections. |
| format | Article |
| id | doaj-art-2582023809744a0e967375bc2abd27df |
| institution | Kabale University |
| issn | 2218-1989 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Metabolites |
| spelling | doaj-art-2582023809744a0e967375bc2abd27df2025-01-24T13:41:16ZengMDPI AGMetabolites2218-19892025-01-011514410.3390/metabo15010044A Comprehensive Machine Learning Approach for COVID-19 Target Discovery in the Small-Molecule MetabolomeMd. Shaheenur Islam Sumon0Md Sakib Abrar Hossain1Haya Al-Sulaiti2Hadi M. Yassine3Muhammad E. H. Chowdhury4Department of Electrical Engineering, Qatar University, Doha P.O. Box 2713, QatarDepartment of Biochemistry, University of Regina, Regina, SK S4S 0A2, CanadaDepartment of Biomedical Sciences, College of Health Sciences, Qatar University, Doha P.O. Box 2713, QatarDepartment of Biomedical Sciences, College of Health Sciences, Qatar University, Doha P.O. Box 2713, QatarDepartment of Electrical Engineering, Qatar University, Doha P.O. Box 2713, Qatar<b>Background/Objectives:</b> Respiratory viruses, including Influenza, RSV, and COVID-19, cause various respiratory infections. Distinguishing these viruses relies on diagnostic methods such as PCR testing. Challenges stem from overlapping symptoms and the emergence of new strains. Advanced diagnostics are crucial for accurate detection and effective management. This study leveraged nasopharyngeal metabolome data to predict respiratory virus scenarios including control vs. RSV, control vs. Influenza A, control vs. COVID-19, control vs. all respiratory viruses, and COVID-19 vs. Influenza A/RSV. <b>Method:</b> We proposed a stacking-based ensemble technique, integrating the top three best-performing ML models from the initial results to enhance prediction accuracy by leveraging the strengths of multiple base learners. Key techniques such as feature ranking, standard scaling, and SMOTE were used to address class imbalances, thus enhancing model robustness. SHAP analysis identified crucial metabolites influencing positive predictions, thereby providing valuable insights into diagnostic markers. <b>Results:</b> Our approach not only outperformed existing methods but also revealed top dominant features for predicting COVID-19, including Lysophosphatidylcholine acyl C18:2, Kynurenine, Phenylalanine, Valine, Tyrosine, and Aspartic Acid (Asp). <b>Conclusions:</b> This study demonstrates the effectiveness of leveraging nasopharyngeal metabolome data and stacking-based ensemble techniques for predicting respiratory virus scenarios. The proposed approach enhances prediction accuracy, provides insights into key diagnostic markers, and offers a robust framework for managing respiratory infections.https://www.mdpi.com/2218-1989/15/1/44metabolomicsrespiratory virusesmachine learningdiagnostic markersCOVID-19 |
| spellingShingle | Md. Shaheenur Islam Sumon Md Sakib Abrar Hossain Haya Al-Sulaiti Hadi M. Yassine Muhammad E. H. Chowdhury A Comprehensive Machine Learning Approach for COVID-19 Target Discovery in the Small-Molecule Metabolome Metabolites metabolomics respiratory viruses machine learning diagnostic markers COVID-19 |
| title | A Comprehensive Machine Learning Approach for COVID-19 Target Discovery in the Small-Molecule Metabolome |
| title_full | A Comprehensive Machine Learning Approach for COVID-19 Target Discovery in the Small-Molecule Metabolome |
| title_fullStr | A Comprehensive Machine Learning Approach for COVID-19 Target Discovery in the Small-Molecule Metabolome |
| title_full_unstemmed | A Comprehensive Machine Learning Approach for COVID-19 Target Discovery in the Small-Molecule Metabolome |
| title_short | A Comprehensive Machine Learning Approach for COVID-19 Target Discovery in the Small-Molecule Metabolome |
| title_sort | comprehensive machine learning approach for covid 19 target discovery in the small molecule metabolome |
| topic | metabolomics respiratory viruses machine learning diagnostic markers COVID-19 |
| url | https://www.mdpi.com/2218-1989/15/1/44 |
| work_keys_str_mv | AT mdshaheenurislamsumon acomprehensivemachinelearningapproachforcovid19targetdiscoveryinthesmallmoleculemetabolome AT mdsakibabrarhossain acomprehensivemachinelearningapproachforcovid19targetdiscoveryinthesmallmoleculemetabolome AT hayaalsulaiti acomprehensivemachinelearningapproachforcovid19targetdiscoveryinthesmallmoleculemetabolome AT hadimyassine acomprehensivemachinelearningapproachforcovid19targetdiscoveryinthesmallmoleculemetabolome AT muhammadehchowdhury acomprehensivemachinelearningapproachforcovid19targetdiscoveryinthesmallmoleculemetabolome AT mdshaheenurislamsumon comprehensivemachinelearningapproachforcovid19targetdiscoveryinthesmallmoleculemetabolome AT mdsakibabrarhossain comprehensivemachinelearningapproachforcovid19targetdiscoveryinthesmallmoleculemetabolome AT hayaalsulaiti comprehensivemachinelearningapproachforcovid19targetdiscoveryinthesmallmoleculemetabolome AT hadimyassine comprehensivemachinelearningapproachforcovid19targetdiscoveryinthesmallmoleculemetabolome AT muhammadehchowdhury comprehensivemachinelearningapproachforcovid19targetdiscoveryinthesmallmoleculemetabolome |