Hierarchical classification-based pan-cancer methylation analysis to classify primary cancer.

Hierarchical classification offers a more specific categorization of data and breaks down large classification problems into subproblems, providing improved prediction accuracy and predictive power for undefined categories, while also mitigating the impact of poor-quality data. Despite these advantages, its application in predicting primary cancer is rare. To leverage the similarity of cancers and the specificity of methylation patterns among them, we developed the Cancer Hierarchy Classification Tool (CHCT) using the idea of hierarchical classification, with methylation data from 30 cancer types and 8239 methylome samples downloaded from publicly available databases (The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO)). We used unsupervised clustering to divide the classification subproblems and screened differentially methylated sites using Analysis of variance (ANOVA) test, Tukey-kramer test, and Boruta algorithms to construct models for each classifier module. After validation, CHCT accurately classified 1568 out of 1660 cases in the test set, with an average accuracy of 94.46%. We further curated an independent validation cohort of 677 cancer samples from GEO and assigned a diagnosis using CHCT, which showed high diagnostic potential with generally high accuracies (an average accuracy of 91.40%). Moreover, CHCT demonstrates predictive capability for additional cancer types beyond its original classifier scope as demonstrated in the medulloblastoma and pituitary tumor datasets. In summary, CHCT can hierarchically classify primary cancer by methylation profile, by splitting a large-scale classification of 30 cancer types into ten smaller classification problems. These results indicate that cancer hierarchical classification has the potential to be an accurate and robust cancer classification method.

Proteomic profiling of saliva reveals association of complement system with primary Sjögren's syndrome.

INTRODUCTION: To compare the saliva proteomes of experimental Sjögren's syndrome (ESS) model mice and healthy controls to identify potential diagnostic biomarkers for primary Sjögren's syndrome (pSS). METHODS: Proteins were extracted from the saliva of three ESS and three normal control mice using the data-independent acquisition technique. R language was used to identify the differentially expressed proteins (DEPs). Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway analyses were performed to functionally annotate the DEPs. The protein-protein interaction (PPI) network was constructed and the core proteins were identified with the STRING website and Cytoscape software. The concentrations of Serpin family G member 1 (SERPING1), C3, complement factor H (CFH), fibrinogen alpha (FGA), and fibrinogen gamma (FGG) in saliva were determined by ELISA. RESULTS: A total of 1722 DEPs were identified in the saliva of the ESS mice relative to the controls, of which 50 showed significantly different expression levels between the two groups. SERPING1, C3, CFH, FGA, and FGG were significantly downregulated, and keratin 4 (Krt4) and transglutaminase 3 (TGM3) were upregulated in the saliva of ESS mice. The PPI network showed that SERPING1, C3, FGG, FGA, TGM3, and hemopexin (HPX) were the core proteins. ELISA results showed that the expression of C3, CFH, FGA, and SERPING1 were significantly downregulated in the saliva of ESS mice. However, the expression of FGG was a little downregulated but with no significant difference. SERPING1, FGG, and FGA may downregulate the complement C3 by inhibiting immune complement system, thereby promoting pSS progression. CONCLUSIONS: The salivary proteome of ESS mice was markedly different from that of healthy controls, suggesting that salivary proteomics is a promising noninvasive diagnostic tool for pSS. SERPING1, C3, CFH, FGA, and FGG are potential biomarkers of pSS.

An integrated methodology for real-time driving risk status prediction using naturalistic driving data.

Real-time driving risk status prediction is critical for developing proactive traffic intervention strategies and enhance driving safety. However, the optimal observation time window length and prediction time window length, which should be the prerequisite for the timeliness and accuracy of real-time driving risk status prediction model, have been rarely explored in previous studies. In this study, a methodology which integrates driving risk status identification, rolling time window-based feature extraction, real-time driving risk status prediction and driving risk influencing factors analysis was proposed to accurately evaluate and predict real-time driving risk status. The methodology was tested based on 1,440 car-following events from Shanghai Naturalistic Driving Study. Results show that four driving risk statuses (safe, low-risk, median-risk and high-risk) are most appropriate to establish risk labelling criteria. In addition, results from driving risk status prediction show that when the observation time window length is 0.5 s, the accuracy rate of predicting medium-risk or high-risk status occurring in the next 0.7 s is higher than 85 % using multi-layer perceptron model. Meanwhile, the results from the analysis of influencing factors show that the input variables related to the risk status score higher in the ranking of feature importance. A part from that, speed difference, headway distance, speed and acceleration are still important in predicting driving risk status. The proposed methods in this paper can be applied in connected and autonomous vehicle (CAV) to reduce driver cognitive workload and hence improve driving safety fed with naturalistic driving data collected using in-vehicle systems.

Crash prediction based on traffic platoon characteristics using floating car trajectory data and the machine learning approach.

Predicting crash propensity helps study safety on urban expressways in order to implement countermeasures and make improvements. It also helps identify and prevent crashes before they happen. However, collecting real-time wide-coverage traffic information for crash prediction has been challenging. More importantly, previous studies have failed to consider the characteristics of the traffic platoon (vehicle group) that the crash vehicle belongs to before the crash occurs. This paper aims to model crash propensity based on traffic platoon characteristics collected by the floating car method, which provides a time-efficient and reliable solution to collecting traffic information. Crash and floating car data are collected from the Middle Ring Expressway in Shanghai, China. Both the binary logistic model and the support vector machine are applied. A data preparation method, involving crash data filtering, floating car data filtering and data matching on the road network, is introduced for the safety analysis purpose. Results suggest that the traffic platoon information collected from floating cars accompanied works reasonably in predicting crashes on expressways. The support vector machine, with an overall accuracy of 85%, outperformed the binary logistic model which had an overall accuracy of 60%. Results further suggest the application of floating car technologies and the support vector machine in real-time crash prediction. Despite this, the study also concludes the merits of the binary logistic model over the support vector machine model in explaining the impact of different factors that contribute to crash occurrences.