Application of Fuzzy c-Means Clustering in Data Analysis of Metabolomics

Abstract

Fuzzy c-means (FCM) clustering is an unsupervised method derived from fuzzy logic that is suitable for solving multiclass and ambiguous clustering problems. In this study, FCM clustering is applied to cluster metabolomics data. FCM is performed directly on the data matrix to generate a membership matrix which represents the degree of association the samples have with each cluster. The method is parametrized with the number of clusters (C) and the fuzziness coefficient (m), which denotes the degree of fuzziness in the algorithm. Both have been optimized by combining FCM with partial least-squares (PLS) using the membership matrix as the Y matrix in the PLS model. The quality parameters R(2)Y and Q(2) of the PLS model have been used to monitor and optimize C and m. Data of metabolic profiles from three gene types of Escherichia coli were used to demonstrate the method above. Different multivariable analysis methods have been compared. Principal component analysis failed to model the metabolite data, while partial least-squares discriminant analysis yielded results with overfitting. On the basis of the optimized parameters, the FCM was able to reveal main phenotype changes and individual characters of three gene types of E. coli. Coupled with PLS, FCM provides a powerful research tool for metabolomics with improved visualization, accurate classification, and outlier estimation.

Independent component analysis in non-hypothesis driven metabolomics: Improvement of pattern discovery and simplification of biological data interpretation demonstrated with plasma samples of exercising humans

“…A(77) was similar to A(72), but S(77) was not similar to S(72). The situation above suggested the independent components in IC 3 3 were all confused. A possible explanation of this situation could be that the maximization process of the ICA could not find three extremes simultaneously, resulting in a mixing between two extremes (i.e.…”

“…2C shows S(72) was similar to S(17), although there are obvious differences in the absolute values (S(72) was larger than S (17)). Moreover, the kurtosis values of the ICs of IC 3 3 showed a pronounced difference, e.g. the kurtosis values of S(17) and S(77) were 120% different.…”

“…If the knowledge about the biological processes or the analytical information (e.g. noise) is ambiguous or lacking, supervised methods may fail to separate biological information from miscellaneous data [2], or they bear a high risk of over-fitting the data and misinterpretation of the observations may happen [2,3].…”

Quantitative Assessment and Classification of Tissue-Based Biomarker Expression With Color Content Analysis

“…In this study, we used the fuzzy c-means (FCM) clustering method. 63,64 Fuzzy c-means is an unsupervised method derived from fuzzy logic that is suitable for solving multiclass and ambiguous clustering problems. Fuzzy c-means was implemented using the MAT-LAB fuzzy toolbox 65 by using the following procedure.…”

Screening of Metabolite Regulating Lactic Acid Production ofExiguobacterium aurantiacumATCC49676 by Metabolic Fingerprint Analysis

“…Monitoring the fluctuations of intercellular metabolites usually provides valuable clues to the corresponding phenotypic traits. [9] In microbiological field, the application of metabolic fingerprint analysis involves in elucidating the roles of genes with unknown functions, [10,11] differentiating phenotypes [9,12] and investigating the responses to external stimuli. [13,14] Currently, effective analytical techniques for metabolic fingerprint analysis include nuclear magnetic resonance (NMR), gas chromatography-mass spectrometry (GC-MS), liquid chromatography-mass spectrometry (LC-MS), capillary electrophoresis-mass spectrometry (CE-MS) and so on.…”