Linhui Zhang

Clear-cell renal cell carcinoma (ccRCC) is characterized by profound metabolic reprogramming and substantial inherited susceptibility. The molecular mechanisms linking germline risk variation to tumor metabolic reprogramming could potentially be exploited for precision interventions. In this study, we aimed to delineate the germline regulatory architecture underlying ccRCC susceptibility and to determine how inherited noncoding variation influences mitochondrial metabolism and tumor progression. Whole-genome sequencing and transcriptomic profiling of matched ccRCC tumor and adjacent normal kidney tissues were integrated with cross-ancestry genome-wide association, cis-expression quantitative trait locus, transcriptome-wide association, and fine-mapping analyses. LINC02068 was identified as a genetically regulated tumor suppressor associated with inherited ccRCC susceptibility, and cross-ancestry fine-mapping prioritized rs234043 as a candidate causal variant. The rs234043 risk allele preferentially recruited HOXC8 and repressed LINC02068 transcription. Reduced LINC02068 expression promoted malignant cellular phenotypes and was associated with adverse clinical outcomes. Mechanistically, LINC02068 localized to mitochondria, interacted with the cochaperone DNAJA3, and promoted DNAJA3 destabilization, thereby attenuating mitochondrial oxidative phosphorylation. Allele-specific editing and genetic rescue experiments further supported a functional rs234043–HOXC8–LINC02068–DNAJA3 regulatory axis driving ccRCC progression. Pharmacologic impairment of mitochondrial complex I suppressed ccRCC cell migration and proliferation. These findings establish a mechanistic link between a germline noncoding variation and mitochondrial metabolic reprogramming in ccRCC and identify targeting oxidative phosphorylation as a potential avenue for precision therapy. Significance: A clear-cell renal cell carcinoma risk variant acts as an allele-specific regulatory element that preferentially recruits HOXC8 to repress LINC02068 and promote oxidative phosphorylation, revealing how inherited variation shapes tumor metabolism.

Interest in RNA posttranscriptional modifications, particularly 5‐methylcytosine (m5C), has surged in recent years. Studies have shown that m5C plays a key role in cellular processes and is closely linked to tumourigenesis. This growing focus emphasises the importance of understanding the diverse impacts of m5C modifications in both normal cellular functions and cancer development. Moreover, strides in methodologies for discerning m5C have facilitated intricate transcriptome cartography of RNA methylation at the solitary nucleotide echelon. This technical progress has fueled a surge in m5C‐centric investigations, facilitating further exploration of this RNA modification. This review provides a comprehensive analysis of the oncogenic potential of m5C RNA modification, elucidating the precise molecular mechanisms driving its role in cancer development. It consolidates current knowledge regarding the biological consequences of m5C RNA modification in tumour cells. Understanding the role of methylation‐related processes in tumourigenesis shows promise for advancing cancer diagnosis and therapeutic strategies.Highlights: m5C modifications are dynamically regulated by writers, readers, and erasers, influencing cancer progression, metastasis, and immune evasion. Distinct m5C regulatory networks exist across cancers, modulating oncogenic pathways and therapy responses. m5C signatures serve as biomarkers for cancer prognosis and treatment stratification, highlighting their role in precision oncology.

Background: Gestational diabetes mellitus (GDM) is a common endocrine metabolic disease, involving a carbohydrate intolerance of variable severity during pregnancy. The incidence of GDM-related complications and adverse pregnancy outcomes has declined, in part, due to early screening. Machine learning (ML) models are increasingly used to identify risk factors and enable the early prediction of GDM. Objective: The aim of this study was to perform a meta-analysis and comparison of published prognostic models for predicting the risk of GDM and identify predictors applicable to the models. Methods: Four reliable electronic databases were searched for studies that developed ML prediction models for GDM in the general population instead of among high-risk groups only. The novel Prediction Model Risk of Bias Assessment Tool (PROBAST) was used to assess the risk of bias of the ML models. The Meta-DiSc software program (version 1.4) was used to perform the meta-analysis and determination of heterogeneity. To limit the influence of heterogeneity, we also performed sensitivity analyses, a meta-regression, and subgroup analysis. Results: A total of 25 studies that included women older than 18 years without a history of vital disease were analyzed. The pooled area under the receiver operating characteristic curve (AUROC) for ML models predicting GDM was 0.8492; the pooled sensitivity was 0.69 (95% CI 0.68-0.69; P<.001; I2=99.6%) and the pooled specificity was 0.75 (95% CI 0.75-0.75; P<.001; I2=100%). As one of the most commonly employed ML methods, logistic regression achieved an overall pooled AUROC of 0.8151, while non–logistic regression models performed better, with an overall pooled AUROC of 0.8891. Additionally, maternal age, family history of diabetes, BMI, and fasting blood glucose were the four most commonly used features of models established by the various feature selection methods. Conclusions: Compared to current screening strategies, ML methods are attractive for predicting GDM. To expand their use, the importance of quality assessments and unified diagnostic criteria should be further emphasized.

BACKGROUND Gestational diabetes mellitus (GDM) is a common endocrine metabolic disease, involving a carbohydrate intolerance of variable severity during pregnancy. The incidence of GDM-related complications and adverse pregnancy outcomes has declined, in part, due to early screening. Machine learning (ML) models are increasingly used to identify risk factors and enable the early prediction of GDM. OBJECTIVE The aim of this study was to perform a meta-analysis and comparison of published prognostic models for predicting the risk of GDM and identify predictors applicable to the models. METHODS Four reliable electronic databases were searched for studies that developed ML prediction models for GDM in the general population instead of among high-risk groups only. The novel Prediction Model Risk of Bias Assessment Tool (PROBAST) was used to assess the risk of bias of the ML models. The Meta-DiSc software program (version 1.4) was used to perform the meta-analysis and determination of heterogeneity. To limit the influence of heterogeneity, we also performed sensitivity analyses, a meta-regression, and subgroup analysis. RESULTS A total of 25 studies that included women older than 18 years without a history of vital disease were analyzed. The pooled area under the receiver operating characteristic curve (AUROC) for ML models predicting GDM was 0.8492; the pooled sensitivity was 0.69 (95% CI 0.68-0.69; P<.001; I2=99.6%) and the pooled specificity was 0.75 (95% CI 0.75-0.75; P<.001; I2=100%). As one of the most commonly employed ML methods, logistic regression achieved an overall pooled AUROC of 0.8151, while non–logistic regression models performed better, with an overall pooled AUROC of 0.8891. Additionally, maternal age, family history of diabetes, BMI, and fasting blood glucose were the four most commonly used features of models established by the various feature selection methods. CONCLUSIONS Compared to current screening strategies, ML methods are attractive for predicting GDM. To expand their use, the importance of quality assessments and unified diagnostic criteria should be further emphasized.