Identification and study of mood-related biomarkers and potential molecular mechanisms in type 2 diabetes mellitus

• Published in: Journal of molecular histology Vol. 56; no. 2; p. 82
• Main Authors: Wang, Menglong, Wang, Tongrui, Liu, Yang, Zhou, Lurong, Yin, Yuanping, Gu, Feng
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.04.2025; Springer Nature B.V
• Subjects: Affect; Animals; Biomarkers; Biomarkers - metabolism; Biomedical and Life Sciences; Biomedicine; Cell Biology; Computational Biology - methods; Developmental Biology; Diabetes; Diabetes mellitus (non-insulin dependent); Diabetes Mellitus, Type 2 - diagnosis; Diabetes Mellitus, Type 2 - genetics; Diabetes Mellitus, Type 2 - metabolism; Diabetes Mellitus, Type 2 - psychology; DNA microarrays; Gene Expression Profiling; Gene Regulatory Networks; Humans; Life Sciences; Machine Learning; Male; Mice; Molecular modelling; Mood; Original Paper; Biomarker; Machine learning algorithms; Mood; Type 2 diabetes mellitus; Regulatory mechanism
• ISSN: 1567-2379; 1567-2387; 1567-2387
• DOI: 10.1007/s10735-025-10353-2

A significant correlation between type 2 diabetes mellitus (T2DM) and mood has been reported. However, the specific mechanism of mood’s role in T2DM is unclear. This study aims to discover mood-related biomarkers in T2DM and further elucidate their underlying molecular mechanisms. The GSE81965 and GSE55650 datasets were sourced from public databases, and mood-related genes (MRGs) were retrieved from previous literature. Initially, differentially expressed MRGs (DE-MRGs) were obtained by combining differential expression analysis and weighted gene co-expression network analysis (WGCNA). Subsequently, the DE-MRGs were incorporated into the LASSO and SVM to identify diagnostic biomarkers for T2DM. Four machine learning methods were utilized to construct the diagnostic models in T2DM, and the model with the optimal algorithm was screened. Further, based on biomarkers, functional enrichment, immune infiltration, and regulatory network analyses were conducted to excavate deeper into the pathogenesis of T2DM. In vivo experiments were used to validate the expression of the biomarkers. A total of 23 DE-MRGs were identified by overlapping 723 DEGs and 64 key modules, and there were strong positive correlations between these DE-MRGs. Afterward, KCTD16, SLC8A1, RAB11FIP1, and RASGEF1B were identified as biomarkers associated with mood in T2DM, and they had favorable diagnostic performance. Meanwhile, the RF diagnostic model constructed based on biomarkers was performed optimally and had high diagnostic accuracy for T2DM patients. Animal experiments indicated that expression levels of SLC8A1, RAB11FIP1, and RASGEF1B in T2DM were consistent with the microarray results. In conclusion, KCTD16, SLC8A1, RAB11FIP1, and RASGEF1B were identified as biomarkers related to mood in T2DM.