Lysosome-Related Diagnostic Biomarkers for Pediatric Sepsis Integrated by Machine Learning
Yang Yang,1 Genhao Zhang2 1Department of Nuclear Medicine, Henan Provincial People’s Hospital, People’s Hospital of Zhengzhou University, People’s Hospital of Henan University, Zhengzhou, Henan, 450003, People’s Republic of China; 2Department of Blood Transfusion, Zhengzhou University First Affiliat...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Dove Medical Press
2023-11-01
|
| Series: | Journal of Inflammation Research |
| Subjects: | |
| Online Access: | https://www.dovepress.com/lysosome-related-diagnostic-biomarkers-for-pediatric-sepsis-integrated-peer-reviewed-fulltext-article-JIR |
| _version_ | 1827659933031596032 |
|---|---|
| author | Yang Y Zhang G |
| author_facet | Yang Y Zhang G |
| author_sort | Yang Y |
| collection | DOAJ |
| description | Yang Yang,1 Genhao Zhang2 1Department of Nuclear Medicine, Henan Provincial People’s Hospital, People’s Hospital of Zhengzhou University, People’s Hospital of Henan University, Zhengzhou, Henan, 450003, People’s Republic of China; 2Department of Blood Transfusion, Zhengzhou University First Affiliated Hospital, Zhengzhou, Henan, 450052, People’s Republic of ChinaCorrespondence: Genhao Zhang, Email smilegenhao@163.comBackground: There is currently no biomarker that can reliably identify sepsis, despite recent scientific advancements. We systematically evaluated the value of lysosomal genes for the diagnosis of pediatric sepsis.Methods: Three datasets (GSE13904, GSE26378, and GSE26440) were obtained from the gene expression omnibus (GEO) database. LASSO regression analysis and random forest analysis were employed for screening pivotal genes to construct a diagnostic model between the differentially expressed genes (DEGs) and lysosomal genes. The efficacy of the diagnostic model for pediatric sepsis identification in the three datasets was validated through receiver operating characteristic curve (ROC) analysis. Furthermore, a total of 30 normal samples and 35 pediatric sepsis samples were gathered to detect the expression levels of crucial genes and assess the diagnostic model’s efficacy in diagnosing pediatric sepsis in real clinical samples through real-time quantitative PCR (qRT-PCR).Results: Among the 83 differentially expressed genes (DEGs) related to lysosomes, four key genes (STOM, VNN1, SORT1, and RETN) were identified to develop a diagnostic model for pediatric sepsis. The expression levels of these four key genes were consistently higher in the sepsis group compared to the normal group across all three cohorts. The diagnostic model exhibited excellent diagnostic performance, as evidenced by area under the curve (AUC) values of 1, 0.971, and 0.989. Notably, the diagnostic model also demonstrated strong diagnostic ability with an AUC of 0.917 when applied to the 65 clinical samples, surpassing the efficacy of conventional inflammatory indicators such as procalcitonin (PCT), white blood cell (WBC) count, C-reactive protein (CRP), and neutrophil percentage (NEU%).Conclusion: A four-gene diagnostic model of lysosomal function was devised and validated, aiming to accurately detect pediatric sepsis cases and propose potential target genes for lysosomal intervention in affected children.Keywords: pediatric sepsis, diagnostic marker, ROC curves, lysosomal, GSEA |
| first_indexed | 2024-03-09T23:32:07Z |
| format | Article |
| id | doaj.art-623ea1cd782f416ea463efade65e5839 |
| institution | Directory Open Access Journal |
| issn | 1178-7031 |
| language | English |
| last_indexed | 2024-03-09T23:32:07Z |
| publishDate | 2023-11-01 |
| publisher | Dove Medical Press |
| record_format | Article |
| series | Journal of Inflammation Research |
| spelling | doaj.art-623ea1cd782f416ea463efade65e58392023-11-23T17:07:55ZengDove Medical PressJournal of Inflammation Research1178-70312023-11-01Volume 165575558388467Lysosome-Related Diagnostic Biomarkers for Pediatric Sepsis Integrated by Machine LearningYang YZhang GYang Yang,1 Genhao Zhang2 1Department of Nuclear Medicine, Henan Provincial People’s Hospital, People’s Hospital of Zhengzhou University, People’s Hospital of Henan University, Zhengzhou, Henan, 450003, People’s Republic of China; 2Department of Blood Transfusion, Zhengzhou University First Affiliated Hospital, Zhengzhou, Henan, 450052, People’s Republic of ChinaCorrespondence: Genhao Zhang, Email smilegenhao@163.comBackground: There is currently no biomarker that can reliably identify sepsis, despite recent scientific advancements. We systematically evaluated the value of lysosomal genes for the diagnosis of pediatric sepsis.Methods: Three datasets (GSE13904, GSE26378, and GSE26440) were obtained from the gene expression omnibus (GEO) database. LASSO regression analysis and random forest analysis were employed for screening pivotal genes to construct a diagnostic model between the differentially expressed genes (DEGs) and lysosomal genes. The efficacy of the diagnostic model for pediatric sepsis identification in the three datasets was validated through receiver operating characteristic curve (ROC) analysis. Furthermore, a total of 30 normal samples and 35 pediatric sepsis samples were gathered to detect the expression levels of crucial genes and assess the diagnostic model’s efficacy in diagnosing pediatric sepsis in real clinical samples through real-time quantitative PCR (qRT-PCR).Results: Among the 83 differentially expressed genes (DEGs) related to lysosomes, four key genes (STOM, VNN1, SORT1, and RETN) were identified to develop a diagnostic model for pediatric sepsis. The expression levels of these four key genes were consistently higher in the sepsis group compared to the normal group across all three cohorts. The diagnostic model exhibited excellent diagnostic performance, as evidenced by area under the curve (AUC) values of 1, 0.971, and 0.989. Notably, the diagnostic model also demonstrated strong diagnostic ability with an AUC of 0.917 when applied to the 65 clinical samples, surpassing the efficacy of conventional inflammatory indicators such as procalcitonin (PCT), white blood cell (WBC) count, C-reactive protein (CRP), and neutrophil percentage (NEU%).Conclusion: A four-gene diagnostic model of lysosomal function was devised and validated, aiming to accurately detect pediatric sepsis cases and propose potential target genes for lysosomal intervention in affected children.Keywords: pediatric sepsis, diagnostic marker, ROC curves, lysosomal, GSEAhttps://www.dovepress.com/lysosome-related-diagnostic-biomarkers-for-pediatric-sepsis-integrated-peer-reviewed-fulltext-article-JIRpediatric sepsisdiagnostic markerroc curveslysosomalgsea |
| spellingShingle | Yang Y Zhang G Lysosome-Related Diagnostic Biomarkers for Pediatric Sepsis Integrated by Machine Learning Journal of Inflammation Research pediatric sepsis diagnostic marker roc curves lysosomal gsea |
| title | Lysosome-Related Diagnostic Biomarkers for Pediatric Sepsis Integrated by Machine Learning |
| title_full | Lysosome-Related Diagnostic Biomarkers for Pediatric Sepsis Integrated by Machine Learning |
| title_fullStr | Lysosome-Related Diagnostic Biomarkers for Pediatric Sepsis Integrated by Machine Learning |
| title_full_unstemmed | Lysosome-Related Diagnostic Biomarkers for Pediatric Sepsis Integrated by Machine Learning |
| title_short | Lysosome-Related Diagnostic Biomarkers for Pediatric Sepsis Integrated by Machine Learning |
| title_sort | lysosome related diagnostic biomarkers for pediatric sepsis integrated by machine learning |
| topic | pediatric sepsis diagnostic marker roc curves lysosomal gsea |
| url | https://www.dovepress.com/lysosome-related-diagnostic-biomarkers-for-pediatric-sepsis-integrated-peer-reviewed-fulltext-article-JIR |
| work_keys_str_mv | AT yangy lysosomerelateddiagnosticbiomarkersforpediatricsepsisintegratedbymachinelearning AT zhangg lysosomerelateddiagnosticbiomarkersforpediatricsepsisintegratedbymachinelearning |