Clustermatch: discovering hidden relations in highly diverse kinds of qualitative and quantitative data without standardization

Abstract

Motivation: Heterogeneous and voluminous data sources are common in modern datasets, particularlyin systems biology studies. For instance, in multi-holistic approaches in the fruit biology field, data sourcescan include a mix of measurements such as morpho-agronomic traits, different kinds of molecules (nucleicacids and metabolites) and consumer preferences. These sources not only have different types of data(quantitative and qualitative), but also large amounts of variables with possibly non-linear relationshipsamong them. An integrative analysis is usually hard to conduct, since it requires several manualstandardization steps, with a direct and critical impact on the results obtained. These are important issuesin clustering applications, which highlight the need of new methods for uncovering complex relationshipsin such diverse repositories.Results: We designed a new method named Clustermatch to easily and efficiently perform data-miningtasks on large and highly heterogeneous datasets. Our approach can derive a similarity measure betweenany quantitative or qualitative variables by looking on how they influence on the clustering of the biologicalmaterials under study. Comparisons with other methods in both simulated and real datasets show thatClustermatch is better suited for finding meaningful relationships in complex datasets.Availability: Files can be downloaded from https://sourceforge.net/projects/sourcesinc/files/clustermatch/and https://bitbucket.org/sinc-lab/clustermatch/.In addition,a web-demo is available athttp://sinc.unl.edu.ar/web-demo/clustermatch/