Hierarchical Clustering Explorer 

for Interactive Exploration of Multidimensional Data

About This Project | HCE 3.5 (Power Analysis) | HCE 3.0 (HCE2W) User Manual | HCE 2.0 | HCE 1.0 | Download

Multidimensional data sets are common in many research areas, including microarray experiment data sets. Genome researchers are using cluster analysis to find meaningful groups in microarray data. Some clustering algorithms, such as k-means, require users to specify the number of clusters as an input, but users rarely know the right number beforehand. Other clustering algorithms automatically determine the right number of clusters, but users may not be convinced of the result since they had little or no control over the clustering process. To avoid this dilemma, the Hierarchical Clustering Explorer (HCE) applies the hierarchical clustering algorithm without a predetermined number of clusters, and then enables users to determine the natural grouping with interactive visual feedback (dendrogram and color mosaic) and dynamic query controls. HCE 1.0  implemented four general techniques that could be used in interactive explorations of clustering results.

• overview of the entire dataset, coupled with a detail view so that high-level patterns and hot spots can be easily found and examined
• dynamic query controls so that users can restrict the number of clusters they view at a time and show those clusters more clearly
• coordinated displays: the overview mosaic has a bi-directional link to 2-dimensional scatterplots
• cluster comparisons to allow researchers to see how different clustering algorithms group the genes.

However, the high dimensionality of the data sets still hinders users from finding interesting patterns, clusters, and outliers. Determining the biological significance of such features remains problematic due to the difficulties of integrating biological knowledge. In addition, it is not efficient to perform a cluster analysis over the whole data set in cases where researchers know the approximate temporal pattern of the gene expression that they are seeking. To address these problems, we developed the Hierarchical Clustering Explorer 2.0 by adding three new features to HCE:

• scatterplot ordering methods so that all 2D projections of a high dimensional data set can be ordered according to relevant criteria
• a gene ontology browser, coupled with clustering results so that known gene functions within a cluster can be easily studied
• a profile search so that genes with a certain temporal pattern can be easily identified.

As an important part of our continuing effort to give users more controls over multidimensional data analysis processes and to enable more interactions with analysis results through interactive visualization techniques, we present a set of principles, GRID principles, that could enable users to better understand distributions in one (1D) or two dimensions (2D), and then discover relationships, clusters, gaps, outliers, and other features in multidimensional data sets. By combining information visualization techniques (overview, coordination, and dynamic query) with summaries and statistical methods, users can systematically examine the most important 1D and 2D axis-parallel projections. Detecting interesting features in low dimensions (1D or 2D) by utilizing powerful human perceptual abilities is crucial to understand the original multidimensional data set. Familiar graphical displays such as histograms, scatterplots, and other well-known 2D plots are effective to reveal features including basic summary statistics, and even unexpected features in the data set. There are also many algorithmic or statistical techniques that are especially effective in low dimensional spaces. While there have been many approaches utilizing such visual displays and low dimensional techniques, most of them lack a systematic framework that organizes such functionalities to help analysts in their feature detection tasks.

We summarize the GRID principles as:

• study 1D, study 2D, then find features
• ranking guides insight, statistics confirm.

Abiding by the GRID principles, we implement a systematic framework, rank-by-feature framework as two separate tabs in Hierarchical Clustering Explorer 3.0:

• Histogram Ordering for 1D projections
• Scatterplot Ordering for 2D projections

If you have any comment or question, send an email to Jinwook Seo (jinwook@cs.umd.edu).

• Ben Shneiderman, Professor, Computer Science
• Jinwook Seo, Graduate Research Assistant, Computer Science

• Jinwook Seo, Ben Shneiderman, "Interactively Exploring Hierarchical Clustering Results," IEEE Computer, Volume 35, Number 7, pp. 80-86, July 2002. [initial draft (pdf)]
• Jinwook Seo, Marina Bakay, Po Zhao, Yi-Wen Chen, Priscilla Clarkson, Ben Shneiderman, Eric P Hoffman, "Interactive Color Mosaic and Dendrogram Displays for Signal/Noise Optimization in Microarray Data Analysis,"  IEEE International Conference on Multimedia and Expo 2003.
• Po Zhao, Jinwook Seo, Zuyi Wang, Yue Wang, Ben Shneiderman, and Eric P Hoffman, "In vivo filtering of in vitro MyoD target data: An approach for identification of biologically relevant novel downstream targets of transcription fctors," Comptes Rendus Biologies, Vol. 326, Issues 10-11, October-November 2003, pp 1049-1065.
• Jinwook Seo, Ben Shneiderman, "Interactive Exploration of Multidimensional Microarray Data: Scatterplot Ordering, Gene Ontology Browser, and Profile Search," HCIL-2003-25, CS-TR-4486, UMIACS-TR-2003-55.
• Jinwook Seo, Marina Bakay, Yi-Wen Chen, Sara Hilmer, Ben Shneiderman, Eric P Hoffman, " Interactively optimizing signal-to-noise ratios in expression profiling: project-specific algorithm selection and detection p-value weighting in Affymetrix microarrays," Bioinformatics, Vol. 20, pp. 2534-2544, 2004.
• Jinwook Seo and Ben Shneiderman, "A Rank-by-Feature Framework for Unsupervised Multidimensional Data Exploration Using Low Dimensional Projections," Proc. IEEE InfoVis2004 , pp. 65-72.
• Jinwook Seo and Ben Shneiderman, "A Rank-by-Feature Framework for Interactive Exploration of Multidimensional Data," Information Visualization, Vol. 4, No. 2, pp. 99-113, 2005.
• Jinwook Seo and Ben Shneiderman, "Understanding Clusters in Multidimensional Spaces: Making Meaning by Combining Insights from Coordinated Views of Domain Knowledge," Technical Report, HCIL-2004-03, 2004.
• Jinwook Seo and Ben Shneiderman, "Knowledge Integration Framework for Information Visualization," will be published in LNCS, Vol. 3379, pp. 207-220, Springer-Verlag, Berlin Heidelberg New York, 2005.
• Jinwook Seo, "Information Visualization Design for Multidimensional Data: Integrating the Rank-By-Feature Framework with Hierarchical Clustering," Ph.D. Dissertation from the Dept. of Computer Science, HCIL-2005-20.
• Jinwook Seo and Ben Shneiderman, "Knowledge Discovery in High Dimensional Data: Case Studies and a User Survey for the Rank-by-Feature Framework," IEEE Transactions on Visualization and Computer Graphics, Vol. 12, No. 3, pp. 311-322, 2006.
• Marina Bakay, Zuyi Wang, Gisela Melcon, Louis Schiltz, Jianhua Xuan, Po Zhao, Vittorio Sartorelli, Jinwook Seo, Elena Pegoraro, Corrado Angelini, Ben Shneiderman, Diana Escolar, Yi-Wen Chen, Sara T. Winokur, Chenguang Fan, Raul Mandler, Yoram Nevo, Erynn Gordon, Yitan Zhu, Yibin Dong, Yue Wang, and Eric P. Hoffman. "Nuclear envelope dystrophies show a transcriptional fingerprint suggesting disruption of Rb-MyoD pathways in muscle regeneration," Brain, Vol. 129 No. 4, pp 996-1013, 2006.
• Jinwook Seo, Heather Gordish-Dressman, Eric P. Hoffman, " An Interactive Power Analysis Tool for Microarray Hypothesis Testing and Generation," Bioinformatics, Vol. 22, No. 7, pp. 808-814, 2006.
• Shneiderman, B., A telescope for high-dimensional data: User controlled data exploration with the rank-by-feature framework, Computing in Science and Engineering , Vol. 8, No. 2, pp. 48-53, 2006.
• Jinwook Seo and Heather Gordish-Dressman, "Exploratory Data Analysis with Categorical Variables: An Improved Rank-by-Feature Framework and a Case Study," International Journal of Human-Computer Interaction, Vol 23, No. 3, pp. 287-314, 2007.

• "Information Visualization Design for Multidimensional Data: Integrating the Rank-by-Feature Framework wit Hierarchical Clustering" for the dissertation defense talk
• "Gene Chasing with the Hierarchical Clustering Explorer : Finding Meaningful Clusters in High Dimensional Data" for The HCIL's 21st Annual Symposium and Open House
• "A Rank-by-Feature Framework for Interactive Multi-dimensional Data Exploration" for a talk at InfoVis 2004, at Austin Texas.

• General Application Examples
• Online Behavioral Data from the business school
• Several more examples in the Application Reports from Information Visualization Class

• HCE used for researches published in various journals:

  • P. L. Bollyky and S. B. Wilson, "CD1d-restricted T-cell subsets and dendritic cell function in autoimmunity," Immunology and Cell Biology, vol. 82, pp. 307-314, 2004.

  • W. H. Boylston, A. Gerstner, J. H. DeFord, M. Madsen, K. Flurkey, D. E. Harrison, and J. Papaconstantinou, "Altered cholesterologenic and lipogenic transcriptional profile in livers of aging Snell dwarf (Pit1dw/dwJ) mice," Aging Cell, vol. 3, pp. 283-296, 2004.

  • S. Cluzet, C. Torregrosa, C. Jacquet, C. Lafitte, J. Fournier, L. Mercier, S. Salamagne, X. Briand, M. T. Esquerre-Tugaye, and B. Dumas, "Gene expression profiling and protection of Medicago truncatula against a fungal infection in response to an elicitor from green algae Ulva spp," Plant, Cell and Environment, vol. 27, pp. 917-928, 2004.

  • E. Paux, V. Carocha, C. Marques, A. Mendes de Sousa, N. Borralho, P. Sivadon, and J. Grima-Pettenati, "Transcript profiling of Eucalyptus xylem genes during tension wood formation," New Phytologist, vol. 167, pp. 89-100, 2005.

  • J. Scheidtmann, A. Frantzen, G. Frenzer, and W. F. Maier, "A combinatorial technique for the search of solid state gas sensor materials," Measurement Science and Technology, vol. 16, pp. 119, 2005.

  • J.-M. Tsai, H.-C. Wang, J.-H. Leu, H.-H. Hsiao, A. H. J. Wang, G.-H. Kou, and C.-F. Lo, "Genomic and Proteomic Analysis of Thirty-Nine Structural Proteins of Shrimp White Spot Syndrome Virus," Journal of Virology, vol. 78, pp. 11360-11370, 2004.

HCE is a standalone Windows� application running on a general PC environment. It is freely downloadable for academic and/or research purposes. Commercial licenses can be negotiated with the UM Office of Technology Commercialization (Gayatri Varma, gayatri@umd.edu).
 

Download HCE 3.5 test version(released on Nov. 11, 2005)

User's Guide for HCE version 3

Register and Download HCE 3.0 (released on Dec. 29, 2004)

Register and Download HCE version 2.0 beta now! (released on May 5, 2003)

Register and Download HCE 1.0

Intel Pentium processor
Microsoft Windows 2000, Windows XP

This research has been partially supported by the grant N01 NS-1-2339 from the National Institutes of Health.

[Bioinformatics Visualization] [Bioinformatics Resources in GenMed]

Last updated

Web Accessibility