{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T05:02:24Z","timestamp":1789621344377,"version":"build-2803163510"},"reference-count":53,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2026,9,13]],"date-time":"2026-09-13T00:00:00Z","timestamp":1789257600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Viruses"],"abstract":"<jats:p>Background: Whole-blood transcriptomic profiling can capture systemic immune-response alterations associated with COVID-19 and may support host-response-based classification. However, evidence regarding the discriminatory value of targeted immune-gene panels remains limited, and in small, high-dimensional datasets, the timing of feature selection may substantially affect model performance and interpretation. Aim: This study aimed to evaluate whether NanoString Human Immunology Panel profiles could distinguish COVID-19 from healthy-control measurements and to compare full-dataset feature selection (FDFS) with leave-one-out cross-validation (LOOCV)-embedded feature selection (LEFS). Methods: Publicly available E-MTAB-8871 data comprising 579 genes and 32 whole-blood transcriptomic profiles were analyzed. The dataset included 22 longitudinal COVID-19 measurements obtained from three participants and 10 measurements obtained from 10 healthy controls. Elastic Net regularization was used for feature selection. Random Forest, XGBoost, and LightGBM classifiers were evaluated using sample-level LOOCV. Model performance was assessed using threshold-dependent, discrimination, and probability-based metrics. A separate exploratory LightGBM model was analyzed using SHapley Additive exPlanations (SHAP) to characterize feature contributions. Results: FDFS identified a fixed 40-gene set, whereas LEFS selected a mean of 42 genes per fold (range: 40\u201347). LightGBM correctly classified all 32 measurement-level profiles (derived from 13 unique participants: 10 healthy controls and three longitudinally sampled COVID-19 participants) in both frameworks, achieving area under the receiver operating characteristic curve (ROC-AUC) and area under the precision\u2013recall curve (PR-AUC) values of 1.000 and Brier scores of 0.005 and 0.006 in the FDFS and LEFS frameworks, respectively. Random Forest achieved accuracies of 0.969 and 1.000, whereas XGBoost achieved an accuracy of 0.969 in both frameworks. SHAP analyses consistently identified AICDA as the dominant contributor to model predictions, followed by ARHGDIB. Conclusions: This exploratory analysis showed that targeted NanoString immune-response profiles contained a compact transcriptomic signal capable of distinguishing COVID-19 from healthy-control measurements within the analyzed dataset. These findings provide proof-of-concept evidence of internal measurement-level discrimination. However, because the COVID-19 profiles consisted of repeated measurements from only three participants, sample-level LOOCV did not constitute independent participant-level validation. External validation in larger cohorts comprising independently sampled participants is required. Given that the COVID-19 arm comprised only three independent participants, these biological findings should be regarded as hypothesis-generating and require validation in substantially larger independent cohorts.<\/jats:p>","DOI":"10.3390\/v18091009","type":"journal-article","created":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T08:10:23Z","timestamp":1789373423000},"page":"1009","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Tree-Based Classification of COVID-19 Using NanoString Whole-Blood Immune-Response Profiles: Comparison of Full-Dataset and LOOCV-Embedded Feature Selection"],"prefix":"10.3390","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6950-6013","authenticated-orcid":false,"given":"Zeynep Burcin","family":"Yilmaz","sequence":"first","affiliation":[{"name":"Department of Infectious Diseases and Clinical Microbiology, Inonu University Faculty of Medicine, Malatya 44280, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7956-9272","authenticated-orcid":false,"given":"Zeynep","family":"Kucukakcali","sequence":"additional","affiliation":[{"name":"Department of Biostatistics and Medical Informatics, Inonu University Faculty of Medicine, Malatya 44280, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6864-7711","authenticated-orcid":false,"given":"Sami","family":"Akbulut","sequence":"additional","affiliation":[{"name":"Department of Biostatistics and Medical Informatics, Inonu University Faculty of Medicine, Malatya 44280, T\u00fcrkiye"},{"name":"Department of Surgery and Liver Transplant Institute, Inonu University Faculty of Medicine, Malatya 44280, T\u00fcrkiye"},{"name":"Department of Public Health, Inonu University Faculty of Medicine, Malatya 44280, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,9,13]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"165878","DOI":"10.1016\/j.bbadis.2020.165878","article-title":"Insights into SARS-CoV-2 genome, structure, evolution, pathogenesis and therapies: Structural genomics approach","volume":"1866","author":"Naqvi","year":"2020","journal-title":"Biochim. Biophys. Acta Mol. Basis Dis."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"43","DOI":"10.1186\/s12982-026-01372-9","article-title":"Understanding the global public health impact of long COVID and strategies for sustainable care","volume":"23","author":"Elmi","year":"2026","journal-title":"Discov. Public Health"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"141","DOI":"10.1038\/s41579-020-00459-7","article-title":"Characteristics of SARS-CoV-2 and COVID-19","volume":"19","author":"Hu","year":"2021","journal-title":"Nat. Rev. Microbiol."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"130","DOI":"10.1038\/s41586-022-05522-2","article-title":"The WHO estimates of excess mortality associated with the COVID-19 pandemic","volume":"613","author":"Msemburi","year":"2023","journal-title":"Nature"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"103315","DOI":"10.1016\/j.eclinm.2025.103315","article-title":"Impact of COVID-19 on disease-specific mortality, healthcare resource utilization, and disease burden across a population over 1 billion in 31 countries: An interrupted time-series analysis","volume":"85","author":"Choi","year":"2025","journal-title":"EClinicalMedicine"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"28","DOI":"10.1038\/s41591-020-01202-8","article-title":"Immune determinants of COVID-19 disease presentation and severity","volume":"27","author":"Brodin","year":"2021","journal-title":"Nat. Med."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"75","DOI":"10.1038\/s41579-023-01001-1","article-title":"Clinical course and management of COVID-19 in the era of widespread population immunity","volume":"22","author":"Meyerowitz","year":"2024","journal-title":"Nat. Rev. Microbiol."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"515","DOI":"10.1056\/NEJMoa2403211","article-title":"Postacute Sequelae of SARS-CoV-2 Infection in the Pre-Delta, Delta, and Omicron Eras","volume":"391","author":"Xie","year":"2024","journal-title":"N. Engl. J. Med."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Daamen, A.R., Bachali, P., Grammer, A.C., and Lipsky, P.E. (2023). Classification of COVID-19 Patients into Clinically Relevant Subsets by a Novel Machine Learning Pipeline Using Transcriptomic Features. Int. J. Mol. Sci., 24.","DOI":"10.3390\/ijms24054905"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1036","DOI":"10.1016\/j.cell.2020.04.026","article-title":"Imbalanced Host Response to SARS-CoV-2 Drives Development of COVID-19","volume":"181","author":"Liu","year":"2020","journal-title":"Cell"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1623","DOI":"10.1038\/s41591-020-1038-6","article-title":"A dynamic COVID-19 immune signature includes associations with poor prognosis","volume":"26","author":"Laing","year":"2020","journal-title":"Nat. Med."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"103262","DOI":"10.1016\/j.ebiom.2021.103262","article-title":"Temporal dynamics of the host molecular responses underlying severe COVID-19 progression and disease resolution","volume":"65","author":"Ong","year":"2021","journal-title":"EBioMedicine"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1007\/s10142-024-01359-2","article-title":"Whole blood transcriptome signature predicts severe forms of COVID-19: Results from the COVIDeF cohort study","volume":"24","author":"Armignacco","year":"2024","journal-title":"Funct. Integr. Genom."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Zhang, C., Feng, Y.G., Tam, C., Wang, N., and Feng, Y. (2021). Transcriptional Profiling and Machine Learning Unveil a Concordant Biosignature of Type I Interferon-Inducible Host Response Across Nasal Swab and Pulmonary Tissue for COVID-19 Diagnosis. Front. Immunol., 12.","DOI":"10.3389\/fimmu.2021.733171"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"1079","DOI":"10.1038\/s41467-021-21289-y","article-title":"Dysregulated transcriptional responses to SARS-CoV-2 in the periphery","volume":"12","author":"McClain","year":"2021","journal-title":"Nat. Commun."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"879","DOI":"10.1016\/j.chom.2020.03.021","article-title":"A Dynamic Immune Response Shapes COVID-19 Progression","volume":"27","author":"Ong","year":"2020","journal-title":"Cell Host Microbe"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Gardinassi, L.G., Souza, C.O.S., Sales-Campos, H., and Fonseca, S.G. (2020). Immune and Metabolic Signatures of COVID-19 Revealed by Transcriptomics Data Reuse. Front. Immunol., 11.","DOI":"10.3389\/fimmu.2020.01636"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Daamen, A.R., Bachali, P., Bonham, C.A., Somerville, L., Sturek, J.M., Grammer, A.C., Kadl, A., and Lipsky, P.E. (2022). COVID-19 patients exhibit unique transcriptional signatures indicative of disease severity. Front. Immunol., 13.","DOI":"10.3389\/fimmu.2022.989556"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"587","DOI":"10.1186\/s12967-022-03751-7","article-title":"Longitudinal transcriptional analysis of peripheral blood leukocytes in COVID-19 convalescent donors","volume":"20","author":"Gedda","year":"2022","journal-title":"J. Transl. Med."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"291","DOI":"10.1186\/s12967-020-02456-z","article-title":"A modular framework for the development of targeted Covid-19 blood transcript profiling panels","volume":"18","author":"Rinchai","year":"2020","journal-title":"J. Transl. Med."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"317","DOI":"10.1038\/nbt1385","article-title":"Direct multiplexed measurement of gene expression with color-coded probe pairs","volume":"26","author":"Geiss","year":"2008","journal-title":"Nat. Biotechnol."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Li, X., Zhou, X., Ding, S., Chen, L., Feng, K., Li, H., Huang, T., and Cai, Y.D. (2022). Identification of Transcriptome Biomarkers for Severe COVID-19 with Machine Learning Methods. Biomolecules, 12.","DOI":"10.3390\/biom12121735"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"5599","DOI":"10.1038\/s41598-023-32268-2","article-title":"A comprehensive analysis of gene expression profiling data in COVID-19 patients for discovery of specific and differential blood biomarker signatures","volume":"13","author":"Momeni","year":"2023","journal-title":"Sci. Rep."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"6562","DOI":"10.1073\/pnas.102102699","article-title":"Selection bias in gene extraction on the basis of microarray gene-expression data","volume":"99","author":"Ambroise","year":"2002","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Varma, S., and Simon, R. (2006). Bias in error estimation when using cross-validation for model selection. BMC Bioinform., 7.","DOI":"10.1186\/1471-2105-7-91"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1186\/1758-2946-6-10","article-title":"Cross-validation pitfalls when selecting and assessing regression and classification models","volume":"6","author":"Krstajic","year":"2014","journal-title":"J. Cheminform."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"5981","DOI":"10.1007\/s00216-018-1217-1","article-title":"Overoptimism in cross-validation when using partial least squares-discriminant analysis for omics data: A systematic study","volume":"410","author":"Marco","year":"2018","journal-title":"Anal. Bioanal. Chem."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"e49023","DOI":"10.2196\/49023","article-title":"Practical Considerations and Applied Examples of Cross-Validation for Model Development and Evaluation in Health Care: Tutorial","volume":"2","author":"Wilimitis","year":"2023","journal-title":"JMIR AI"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1038\/s42256-019-0138-9","article-title":"From Local Explanations to Global Understanding with Explainable AI for Trees","volume":"2","author":"Lundberg","year":"2020","journal-title":"Nat. Mach. Intell."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"1454","DOI":"10.1038\/s41592-024-02359-7","article-title":"Applying interpretable machine learning in computational biology-pitfalls, recommendations and opportunities for new developments","volume":"21","author":"Chen","year":"2024","journal-title":"Nat. Methods"},{"key":"ref_31","unstructured":"Ong, E. (2026, September 07). Transcriptomic Analysis of Immune Response in Healthy Controls and COVID-19 Cases Using the NanoString Human Immunology Panel. BioStudies, E-MTAB-8871. Available online: https:\/\/www.ebi.ac.uk\/biostudies\/ArrayExpress\/studies\/E-MTAB-8871."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"30","DOI":"10.1001\/jama.2024.22530","article-title":"The 2024 Revision to the Declaration of Helsinki: Modern Ethics for Medical Research","volume":"333","author":"Brubaker","year":"2025","journal-title":"JAMA"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1500","DOI":"10.1016\/j.ijsu.2014.07.014","article-title":"Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and elaboration","volume":"12","author":"Vandenbroucke","year":"2014","journal-title":"Int. J. Surg."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"1444","DOI":"10.1038\/s41592-024-02362-y","article-title":"Guiding questions to avoid data leakage in biological machine learning applications","volume":"21","author":"Bernett","year":"2024","journal-title":"Nat. Methods"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"vbad048","DOI":"10.1093\/bioadv\/vbad048","article-title":"nestedcv: An R package for fast implementation of nested cross-validation with embedded feature selection designed for transcriptomics and high-dimensional data","volume":"3","author":"Lewis","year":"2023","journal-title":"Bioinform. Adv."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"2069","DOI":"10.1002\/eji.200324133","article-title":"Activation-induced cytidine deaminase: A dual role in class-switch recombination and somatic hypermutation","volume":"33","author":"Durandy","year":"2003","journal-title":"Eur. J. Immunol."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1016\/bs.ai.2016.11.002","article-title":"Molecular Mechanisms of Somatic Hypermutation and Class Switch Recombination","volume":"133","author":"Methot","year":"2017","journal-title":"Adv. Immunol."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"367","DOI":"10.1615\/CritRevImmunol.v27.i4.60","article-title":"Regulation of aicda expression and AID activity: Relevance to somatic hypermutation and class switch DNA recombination","volume":"27","author":"Xu","year":"2007","journal-title":"Crit. Rev. Immunol."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"1283","DOI":"10.1002\/eji.202048838","article-title":"Expansion of plasmablasts and loss of memory B cells in peripheral blood from COVID-19 patients with pneumonia","volume":"50","author":"Meschiari","year":"2020","journal-title":"Eur. J. Immunol."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"eabd7114","DOI":"10.1126\/sciimmunol.abd7114","article-title":"Comprehensive mapping of immune perturbations associated with severe COVID-19","volume":"5","author":"Pampena","year":"2020","journal-title":"Sci. Immunol."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"1296","DOI":"10.1016\/j.immuni.2020.11.017","article-title":"Longitudinal Multi-omics Analyses Identify Responses of Megakaryocytes, Erythroid Cells, and Plasmablasts as Hallmarks of Severe COVID-19","volume":"53","author":"Bernardes","year":"2020","journal-title":"Immunity"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1093\/cei\/uxad044","article-title":"Plasmablasts in previously immunologically na\u00efve COVID-19 patients express markers indicating mucosal homing and secrete antibodies cross-reacting with SARS-CoV-2 variants and other beta-coronaviruses","volume":"213","author":"Lundgren","year":"2023","journal-title":"Clin. Exp. Immunol."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"103325","DOI":"10.1016\/j.isci.2021.103325","article-title":"Maturation trajectories and transcriptional landscape of plasmablasts and autoreactive B cells in COVID-19","volume":"24","author":"Paschold","year":"2021","journal-title":"iScience"},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"493","DOI":"10.1038\/nrm3153","article-title":"The \u2018invisible hand\u2019: Regulation of RHO GTPases by RHOGDIs","volume":"12","author":"Boulter","year":"2011","journal-title":"Nat. Rev. Mol. Cell Biol."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"7568","DOI":"10.1073\/pnas.90.16.7568","article-title":"Ly-GDI, a GDP-dissociation inhibitor of the RhoA GTP-binding protein, is expressed preferentially in lymphocytes","volume":"90","author":"Scherle","year":"1993","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Tripathi, M., Colige, A., and Deroanne, C.F. (2023). The Dual Function of RhoGDI2 in Immunity and Cancer. Int. J. Mol. Sci., 24.","DOI":"10.3390\/ijms24044015"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"5491","DOI":"10.1111\/bph.12658","article-title":"Targeting Rho-GTPases in immune cell migration and inflammation","volume":"171","author":"Biro","year":"2014","journal-title":"Br. J. Pharmacol."},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"630","DOI":"10.1038\/nri2606","article-title":"Rho family GTPases and their regulators in lymphocytes","volume":"9","author":"Tybulewicz","year":"2009","journal-title":"Nat. Rev. Immunol."},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"496","DOI":"10.1182\/blood.2023023237","article-title":"Integrative single-cell chromatin and transcriptome analysis of human plasma cell differentiation","volume":"144","author":"Alaterre","year":"2024","journal-title":"Blood"},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"1451","DOI":"10.1038\/s41375-021-01234-0","article-title":"RNA-sequencing data-driven dissection of human plasma cell differentiation reveals new potential transcription regulators","volume":"35","author":"Kassambara","year":"2021","journal-title":"Leukemia"},{"key":"ref_51","doi-asserted-by":"crossref","unstructured":"Li, Y., Han, L., Li, P., Ge, J., Xue, Y., and Chen, L. (2023). Potential network markers and signaling pathways for B cells of COVID-19 based on single-cell condition-specific networks. BMC Genom., 24.","DOI":"10.1186\/s12864-023-09719-1"},{"key":"ref_52","doi-asserted-by":"crossref","first-page":"1829","DOI":"10.1038\/s41467-024-46150-w","article-title":"Data leakage inflates prediction performance in connectome-based machine learning models","volume":"15","author":"Rosenblatt","year":"2024","journal-title":"Nat. Commun."},{"key":"ref_53","doi-asserted-by":"crossref","first-page":"e87728","DOI":"10.2196\/87728","article-title":"Participant-Aware Model Validation for Repeated-Measures Data: Comparative Cross-Validation Study","volume":"5","author":"Karbalaie","year":"2026","journal-title":"JMIR AI"}],"container-title":["Viruses"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4915\/18\/9\/1009\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T04:14:32Z","timestamp":1789618472000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4915\/18\/9\/1009"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,13]]},"references-count":53,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["v18091009"],"URL":"https:\/\/doi.org\/10.3390\/v18091009","relation":{},"ISSN":["1999-4915"],"issn-type":[{"value":"1999-4915","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,13]]}}}