A method for visual identification of small sample subgroups and potential biomarkers

TL;DR

This paper introduces a novel dissimilarity measure integrated with Multidimensional Scaling to visually identify small sample subgroups and potential biomarkers in biomedical data, enhancing exploratory analysis beyond traditional biplots.

Contribution

It presents a new dissimilarity measure and visualization method that jointly represents samples and variables, improving the detection of small, distinct subgroups in complex data.

Findings

Effective visualization of small subgroups and biomarkers.

Better identification of variable patterns in small sample groups.

Enhanced exploratory analysis capabilities.

Abstract

In order to find previously unknown subgroups in biomedical data and generate testable hypotheses, visually guided exploratory analysis can be of tremendous importance. In this paper we propose a new dissimilarity measure that can be used within the Multidimensional Scaling framework to obtain a joint low-dimensional representation of both the samples and variables of a multivariate data set, thereby providing an alternative to conventional biplots. In comparison with biplots, the representations obtained by our approach are particularly useful for exploratory analysis of data sets where there are small groups of variables sharing unusually high or low values for a small group of samples.