Visualizing and clustering high throughput sub-cellular localization imaging

Abstract

Background: The expansion of automatic imaging technologies has created a need to be able to efficiently compare and review large sets of image data. To enable comparisons of image data between samples we need to define the normal variation within distinct images of the same sample. Even with tightly controlled experimental conditions, protein expression can vary widely between cells, and because of the difficulty in viewing and comparing large image sets this might not be observed. Here we introduce a novel methodology, iCluster, for visualizing, clustering and comparing large sub-cellular localization image sets. For each member of an image set, iCluster generates statistics that have been found to be useful in distinguishing sub-cellular localization. The statistics are mapped into two or three dimensions such as to preserve distances between the statistics vectors. The complete image set is then visualized in two or three dimensions using the coordinates so determined. The result is images that are statistically similar are spatially close in the visualization allowing for easy comparison of images that are similar and distinguishment of dissimilar images into distinct clusters.

Quantification and cluster analysis of actin cytoskeletal structures in plant cells: role of actin bundling in stomatal movement during diurnal cycles in Arabidopsis guard cells

“…Its application to microscopic images is currently unfamiliar, but is progressing in the context of the large‐scale analysis of intracellular localizations (Chen et al. , 2006; Hamilton and Teasdale, 2008) and gene expression patterns (Peng et al. , 2007).…”

“…Clustering is an unsupervised classification, and a common and powerful method to order a large quantity of omics data. Its application to microscopic images is currently unfamiliar, but is progressing in the context of the large-scale analysis of intracellular localizations (Chen et al, 2006;Hamilton and Teasdale, 2008) and gene expression patterns (Peng et al, 2007). As shown in this study, the clustering approach also appears to be useful for investigating intracellular dynamics.…”

LIPS database with LIPService: a microscopic image database of intracellular structures in Arabidopsis guard cells

“…To demonstrate this utility, we performed image clustering analysis of images obtained from another database, the Plant Organelle Database (http://podb.nibb.ac.jp/Organellome/) [6], using guard cell images of nuclei labeled by GFP fusions with a nuclear localization sequence (NLS-GFP; Additional file 1: Figure S1A), mitochondria labeled by DsRed fusions with the pre-sequence of the delta-prime subunit of mitochondrial F1-ATPase (F1-ATPase-δ-DsRed; Additional file 1: Figure S1B), and chloroplasts labeled by GFP fusions with CAS, a thylakoid membrane-localized protein (CAS-GFP) [14] (Additional file 1: Figure S1C). After preprocessing (Additional file 1: Figure S1D-H), image clustering analysis was performed using the freely-available image clustering software iCluster (http://icluster.imb.uq.edu.au/) [15] (Figure 3, Additional file 2: Movie S1) and the 55 selected LIPS images (5 examples × 11 probes) that users can download as a ZIP file (860 KiB). iCluster gathered images of the same organelles (Figure 3), including nuclei (NLS-GFP vs. HistoneH2B-RFP; Additional file 1: Figure S2A), mitochondria (F1-ATPase-δ-DsRed vs. Mt-GFP; Additional file 1: Figure S2B), and chloroplasts (CAS-GFP vs. autofluorescence; Additional file 1: Figure S2C).…”

Overview

“…Along these same lines, the biophysical protein characterization (e.g. Structural Genomics Consortium (SGC); review [98]) and in vivo cell localization (review [99, 100]) are now being carried out in higher throughput mode although often in model systems.…”