Comprehensive study of semi-supervised learning for DNA methylation-based supervised classification of central nervous system tumors

Abstract

Background Precision medicine for cancer treatment relies on an accurate pathological diagnosis. The number of known tumor classes has increased rapidly, and reliance on traditional methods of histopathologic classification alone has become unfeasible. To help reduce variability, validation costs, and standardize the histopathological diagnostic process, supervised machine learning models using DNA-methylation data have been developed for tumor classification. These methods require large labeled training data sets to obtain clinically acceptable classification accuracy. While there is abundant unlabeled epigenetic data across multiple databases, labeling pathology data for machine learning models is time-consuming and resource-intensive, especially for rare tumor types. Semi-supervised learning (SSL) approaches have been used to maximize the utility of labeled and unlabeled data for classification tasks and are effectively applied in genomics. SSL methods have not yet been explored with epigenetic data nor demonstrated beneficial to central nervous system (CNS) tumor classification. Results This paper explores the application of semi-supervised machine learning on methylation data to improve the accuracy of supervised learning models in classifying CNS tumors. We comprehensively evaluated 11 SSL methods and developed a novel combination approach that included a self-training with editing using support vector machine (SETRED-SVM) model and an L2-penalized, multinomial logistic regression model to obtain high confidence labels from a few labeled instances. Results across eight random forest and neural net models show that the pseudo-labels derived from our SSL method can significantly increase prediction accuracy for 82 CNS tumors and 9 normal controls. Conclusions The proposed combination of semi-supervised technique and multinomial logistic regression holds the potential to leverage the abundant publicly available unlabeled methylation data effectively. Such an approach is highly beneficial in providing additional training examples, especially for scarce tumor types, to boost the prediction accuracy of supervised models.

Intersecting Pathways in Bioinformatics and Translational Informatics: A One Health Perspective on Key Contributions and Future Directions

“…Lastly, the scalability of computational models for use in clinical settings is a key theme in the papers [6,9,14], which illustrate the potential for implementing precision models at scale in existing academic medical centers and how machine learning models can efficiently leverage available data to produce accurate results, facilitating more personalized and effective patient care.…”

“…Some papers, including Wang et al [9] and Wang et al [11], evaluate the deployment of precision models at scale in existing academic medical centers. Others, such as Weitz et al [6] and Tran et al [14], develop models that can more efficiently leverage available data to produce accurate results.…”

“…Tran et al [14] apply semi-supervised machine learning approaches to classify central nervous system tumors into known subclasses based on DNA methylation patterns. This allows the authors to leverage a much larger collection of unlabeled methylation data to generate pseudo-labels that can improve classification performance without requiring the expensive clinical resources required to generate fully labeled data.…”

Methods in DNA methylation array dataset analysis: A review

“…This model used the β-values of CpG sites derived from EWAS, as the predictor variable for predicting the diagnosis of Gestational Diabetes Mellitus [134] . While SVMs offer advantages such as handling nonlinear relationships, robustness in high-dimensional spaces, and effectiveness with small sample sizes, they do have limitations [135] . These include the inability of the Minfi package to capture all relevant biological variability and the challenge of interpreting SVM models [136] .…”

DNA-methylation subgroups carry no prognostic significance in ATRT-SHH patients in clinical trial cohorts

“…S1). Next, we assigned consensus labels to SJCRH cases using a combination of unsupervised and semi-supervised methods [ 8 – 10 ] (Fig. S2).…”