Experimental eye research · 2026
Integrated Bioinformatics Identification and Independent Validation of FLNA, CCL2, and VMO1 as Candidate Biomarkers for Dexamethasone-Induced Glaucoma.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Zhao Dingding, Wang Mengna, Yin Biao, Zhang Lei, Shi Zhenyu, Li Yongqiang, Ji Xinying, Dang Yalong
Аннотация
BACKGROUND: Dexamethasone-induced glaucoma (DIG) lacks early diagnostic biomarkers, and exosome-associated signatures remain underexplored, hindering timely clinical intervention.
METHODS: We integrated three GEO trabecular meshwork transcriptomic datasets. GSE16643 and GSE37474 (11 control, 11 treated) formed the training set; GSE124114 (9 control, 9 treated) served as independent validation. Differentially expressed genes (DEGs) were intersected with an exosome-associated gene set from GeneCards (relevance score >2). Candidate biomarkers were prioritized via LASSO, SVM-RFE, and Random Forest, followed by cross-validation. A nomogram was constructed and internally validated using 200 bootstrap resamples. Functional enrichment and molecular docking (CB-Dock2) were also performed.
RESULTS: We identified 399 DEGs, of which 16 overlapped with exosome-related genes. Six candidates emerged from multi-algorithm screening, but only three (FLNA, CCL2, and VMO1) passed independent validation; ITGB4 was excluded due to inconsistent expression trends between cohorts. These three genes demonstrated robust discriminative ability in both training and validation sets. Enrichment analyses implicated MAPK signaling, extracellular matrix remodeling, oxidative stress, and lipid metabolic reprogramming, including adipogenesis and cholesterol homeostasis. The nomogram integrating these genes exhibited favorable calibration. Molecular docking revealed strong binding affinity of ellagic acid to FLNA and moderate affinity to CCL2.
CONCLUSION: Our findings suggest that FLNA, CCL2, and VMO1 may serve as promising exosome-associated biomarkers for DIG, potentially involved in immune-metabolic regulation and extracellular matrix homeostasis. However, given the limited cohort size and the computational nature of the docking predictions, these results should be interpreted as hypothesis-generating and require further validation in independent cohorts and experimental systems.
METHODS: We integrated three GEO trabecular meshwork transcriptomic datasets. GSE16643 and GSE37474 (11 control, 11 treated) formed the training set; GSE124114 (9 control, 9 treated) served as independent validation. Differentially expressed genes (DEGs) were intersected with an exosome-associated gene set from GeneCards (relevance score >2). Candidate biomarkers were prioritized via LASSO, SVM-RFE, and Random Forest, followed by cross-validation. A nomogram was constructed and internally validated using 200 bootstrap resamples. Functional enrichment and molecular docking (CB-Dock2) were also performed.
RESULTS: We identified 399 DEGs, of which 16 overlapped with exosome-related genes. Six candidates emerged from multi-algorithm screening, but only three (FLNA, CCL2, and VMO1) passed independent validation; ITGB4 was excluded due to inconsistent expression trends between cohorts. These three genes demonstrated robust discriminative ability in both training and validation sets. Enrichment analyses implicated MAPK signaling, extracellular matrix remodeling, oxidative stress, and lipid metabolic reprogramming, including adipogenesis and cholesterol homeostasis. The nomogram integrating these genes exhibited favorable calibration. Molecular docking revealed strong binding affinity of ellagic acid to FLNA and moderate affinity to CCL2.
CONCLUSION: Our findings suggest that FLNA, CCL2, and VMO1 may serve as promising exosome-associated biomarkers for DIG, potentially involved in immune-metabolic regulation and extracellular matrix homeostasis. However, given the limited cohort size and the computational nature of the docking predictions, these results should be interpreted as hypothesis-generating and require further validation in independent cohorts and experimental systems.