Identifying dynamical persistent biomarker structures for rare events using modern integrative machine learning approach

Abstract

The evolution of omics and computational competency has accelerated discoveries of the underlying biological processes in an unprecedented way. High throughput methodologies, such as flow cytometry, can reveal deeper insights into cell processes, thereby allowing opportunities for scientific discoveries related to health and diseases. However, working with cytometry data often imposes complex computational challenges due to high‐dimensionality, large size, and nonlinearity of the data structure. In addition, cytometry data frequently exhibit diverse patterns across biomarkers and suffer from substantial class imbalances which can further complicate the problem. The existing methods of cytometry data analysis either predict cell population or perform feature selection. Through this study, we propose a “wisdom of the crowd” approach to simultaneously predict rare cell populations and perform feature selection by integrating a pool of modern machine learning (ML) algorithms. Given that our approach integrates superior performing ML models across different normalization techniques based on entropy and rank, our method can detect diverse patterns existing across the model features. Furthermore, the method identifies a dynamic biomarker structure that divides the features into persistently selected, unselected, and fluctuating assemblies indicating the role of each biomarker in rare cell prediction, which can subsequently aid in studies of disease progression.

PerSEveML: A Web-Based Tool to Identify Persistent Biomarker Structure for Rare Events Using Integrative Machine Learning Approach

“…For instance, in the case of the Nilsson rare data set, where we applied the TopS method with an 80:20 percentage split, 10-fold cross-validation, and a cut-off of 40%, we identified XgBoost, naïve Bayes, LDA, and decision tree as models meeting the criterion of sensitivity and specificity above 0.70. This aligns with the outcomes reported by our original paper 1 , except for the decision tree showing superior performance.…”

“…For instance, our method can predict stem cells that belong to multiple clustering groups instead of just one. This is evident from our unsupervised clustering analysis in our previous work 1 . PerSEveML can also analyze protein complexes consisting of modules with shared subunits and mutually exclusive pairs with complex topological structures.…”

“…PerSEveML calculated the entropies and ranks of individual features based on their importance. Entropy is often associated as a measurement of uncertainty in a system 1 . To calculate the entropy score across the different ML methods, the adopted mathematical formula is expressed using equation (5).…”

“…The PerSEveML interface allows users to visualize the correlation between input features as well as the persistent feature structure created using the integrative ML approach. Our earlier research 1 , introduced the use of cutpoint analysis to amalgamate feature importance derived from diverse ML methods utilizing entropy and rank score. Thus, the persistent feature structure calculation heavily depends on the cut-off point analysis.…”

“…Machine learning (ML) methods have gained traction for analyzing rare events, yet there remains a limited exploration of bioinformatics tools that integrate ML techniques to comprehend the underlying biology. Expanding upon our previously developed computational framework of an integrative machine learning approach 1 , we introduce PerSEveML, an interactive web-based that uses crowd-sourced intelligence to predict rare events and determine feature selection structures. PerSEveML provides a comprehensive overview of the integrative approach through evaluation metrics that help users understand the contribution of individual ML methods to the prediction process.…”