DifFUZZY: a fuzzy clustering algorithm for complex datasets

Cominetti, Ornella; Matzavinos, Anastasios; Samarasinghe, Sandhya; Kulasiri, Don; Liu, Sijia; Maini, Philip K.; Erban, Radek

doi:10.1504/ijcibsb.2010.038222

Cited by 23 publications

(14 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(We used a value of 1.4 for the "fuzzifier parameter" (m), because the commonly used value of 2 produced a clustering that was uninformative-all points were assigned equal membership to all clusters.) We should also note that linking fuzzy clustering algorithms with the geometry that diffusion induces on data has also been successfully implemented in the DifFUZZY algorithm of Cominetti et al 57 The fuzzy clustering algorithm allows us to identify those points in a cluster that unambiguously belong to it. For every point i, the algorithm returns a grade u ij ∈ [0, 1] indicating the point's degree of membership in cluster j.…”

Section: F a Fuzzy Markov Modelmentioning

confidence: 99%

Diffusion maps, clustering and fuzzy Markov modeling in peptide folding transitions

Nedialkova

Amat

Kevrekidis

et al. 2014

The Journal of Chemical Physics

View full text Add to dashboard Cite

Using the helix-coil transitions of alanine pentapeptide as an illustrative example, we demonstrate the use of diffusion maps in the analysis of molecular dynamics simulation trajectories. Diffusion maps and other nonlinear data-mining techniques provide powerful tools to visualize the distribution of structures in conformation space. The resulting low-dimensional representations help in partitioning conformation space, and in constructing Markov state models that capture the conformational dynamics. In an initial step, we use diffusion maps to reduce the dimensionality of the conformational dynamics of Ala5. The resulting pretreated data are then used in a clustering step. The identified clusters show excellent overlap with clusters obtained previously by using the backbone dihedral angles as input, with small-but nontrivial-differences reflecting torsional degrees of freedom ignored in the earlier approach. We then construct a Markov state model describing the conformational dynamics in terms of a discrete-time random walk between the clusters. We show that by combining fuzzy C-means clustering with a transition-based assignment of states, we can construct robust Markov state models. This state-assignment procedure suppresses short-time memory effects that result from the non-Markovianity of the dynamics projected onto the space of clusters. In a comparison with previous work, we demonstrate how manifold learning techniques may complement and enhance informed intuition commonly used to construct reduced descriptions of the dynamics in molecular conformation space.

show abstract

Section: F a Fuzzy Markov Modelmentioning

confidence: 99%

Diffusion maps, clustering and fuzzy Markov modeling in peptide folding transitions

Nedialkova

Amat

Kevrekidis

et al. 2014

The Journal of Chemical Physics

View full text Add to dashboard Cite

show abstract

“…For this study, the single-link method best matches the biology. If we set the clustering distance to ε and connect all of the data points in each cluster, we obtain the ε-neighborhood graph (Cominetti et al 2010; Schaeffer 2007). For more information about alternative clustering approaches, see Tan et al (2007), Fan and Pardalos (2010).…”

Section: Introductionmentioning

confidence: 99%

“…The intrinsic clustering distance d I is then defined to be the smallest distance for which there is a maximum number of clusters. A similar concept was introduced inCominetti et al (2010). Clustering for simulated random data is studied and used to normalize the clustering distance for the biological data.…”

Section: Introductionmentioning

confidence: 99%

Using Hierarchical Clustering and Dendrograms to Quantify the Clustering of Membrane Proteins

et al. 2011

View full text Add to dashboard Cite

Cell biologists have developed methods to label membrane proteins with gold nanoparticles and then extract spatial point patterns of the gold particles from transmission electron microscopy images using image processing software. Previously, the resulting patterns were analyzed using the Hopkins statistic, which distinguishes nonclustered from modestly and highly clustered distributions, but is not designed to quantify the number or sizes of the clusters. Clusters were defined by the partitional clustering approach which required the choice of a distance. Two points from a pattern were put in the same cluster if they were closer than this distance. In this study, we present a new methodology based on hierarchical clustering to quantify clustering. An intrinsic distance is computed, which is the distance that produces the maximum number of clusters in the biological data, eliminating the need to choose a distance. To quantify the extent of clustering, we compare the clustering distance between the experimental data being analyzed with that from simulated random data. Results are then expressed as a dimensionless number, the clustering ratio that facilitates the comparison of clustering between experiments. Replacing the chosen cluster distance by the intrinsic clustering distance emphasizes densely packed clusters that are likely more important to downstream signaling events. We test our new clustering analysis approach against electron microscopy images from an experiment in which mast cells were exposed for 1 or 2 minutes to increasing concentrations of antigen that crosslink IgE bound to its high affinity receptor, FcεRI, then fixed and the FcεRI β subunit labeled with 5 nm gold particles. The clustering ratio analysis confirms the increase in clustering with increasing antigen dose predicted from visual analysis and from the Hopkins statistic. Access to a robust and sensitive tool to both observe and quantify clustering is a key step toward understanding the detailed fine scale structure of the membrane, and ultimately to determining the role of spatial organization in the regulation of transmembrane signaling.

show abstract

“…As the clustering problem requires an unsupervised approach, the definition of reliable parameters is a key ingredient in the segmentation process. There are many standard methods in the literature for clustering, among them, we cite hierarchical [56], partitioning [57][58][59][60][61], hybrid method [62], density-based [63] and fuzzy clustering [64][65][66]. Thanks to its manifold applications, recent efforts in the processing of MRI data are documented [67][68][69][70], including unsupervised assessment of fat distribution [71][72][73][74][75], segmentation [74,[76][77][78][79][80][81] and whole-image optimization [82].…”

Section: Clusteringmentioning

confidence: 99%

A data-oriented self-calibration and robust chemical-shift encoding by using clusterization (OSCAR): Theory, optimization and clinical validation in neuromuscular disorders

Siracusano

Corte

Gaeta

et al. 2018

Magnetic Resonance Imaging

View full text Add to dashboard Cite

DifFUZZY: a fuzzy clustering algorithm for complex datasets

Cited by 23 publications

References 24 publications

Diffusion maps, clustering and fuzzy Markov modeling in peptide folding transitions

Diffusion maps, clustering and fuzzy Markov modeling in peptide folding transitions

Using Hierarchical Clustering and Dendrograms to Quantify the Clustering of Membrane Proteins

A data-oriented self-calibration and robust chemical-shift encoding by using clusterization (OSCAR): Theory, optimization and clinical validation in neuromuscular disorders

Contact Info

Product

Resources

About