Towards a Holistic, Yet Gene-Centered Analysis of Gene Expression Profiles: A Case Study of Human Lung Cancers

Abstract

Genome-wide gene expression profile studies encompass increasingly large number of samples, posing a challenge to their presentation and interpretation without losing the notion that each transcriptome constitutes a complex biological entity. Much like pathologists who visually analyze information-rich histological sections as a whole, we propose here an integrative approach. We use a self-organizing maps -based software, the gene expression dynamics inspector (GEDI) to analyze gene expression profiles of various lung tumors. GEDI allows the comparison of tumor profiles based on direct visual detection of transcriptome patterns. Such intuitive “gestalt” perception promotes the discovery of interesting relationships in the absence of an existing hypothesis. We uncovered qualitative relationships between squamous cell tumors, small-cell tumors, and carcinoid tumor that would have escaped existing algorithmic classifications. These results suggest that GEDI may be a valuable explorative tool that combines global and gene-centered analyses of molecular profiles from large-scale microarray experiments.

The Influence of Developmental Age on the Early Transcriptomic Response of Children with Septic Shock

“…Gene expression mosaics representing the expression patterns of differentially regulated genes were generated using the Gene Expression Dynamics Inspector (GEDI) (15)(16)(17). The signature graphic outputs of GEDI are expression mosaics that give microarray data a "face" that is intuitively recognizable via human pattern recognition.…”

Embryonic Morphogenetic Field Induces Phenotypic Reversion in Cancer Cells. Review Article

“…On the other hand, if we move from genomic alterations to cancer transcriptome, considered as "a whole", as suggested by Huang and Kaufmann [59], a more interesting picture emerges and a stunning degree of organization and order of global expression patterns is observed. Indeed, cluster analysis of tumor transcriptomes readily classifies tumors into a small number of discretely distinct groups [60] -reminiscent of the organization of transcriptomes of normal cell types into groups of related tissues [61]. Noteworthy, such a clustering is well correlated with the traditional classification of tumor types derived from morphology.…”

Portraying the expression landscapes of cancer subtypes

“…It uses meta-genes instead of single genes as the basal data, which has the advantage of improving the representativeness and resolution of the results. 6,31 We applied second-level SOM analysis as proposed by Guo et al 9 to visualize the similarity between the individual SOM meta-gene expression patterns.…”

“…Importantly, all these visualizations of similarities are based on the meta-genes, providing a better resolution comparison than single gene-based similarity analyses due the lower noise after SOM dimensionality reduction. 6,9 The MST plot shows a chain-like structure which connects the samples with the strongest mutual correlations. This has the key advantage that it converts multi-dimensional clusters into a relatively simple graph.…”