Related Experiment Video
Updated: Jul 16, 2026
08:20
Using Microarrays to Interrogate Microenvironmental Impact on Cellular Phenotypes in Cancer
Published on: May 21, 2019
Transcript and protein expression profiles of the NCI-60 cancer cell panel: an integromic microarray study
Uma T Shankavaram1, William C Reinhold, Satoshi Nishizuka
1Genomics and Bioinformatics Group, Laboratory of Molecular Pharmacology, Center for Cancer Research, National Cancer Institute/NIH, Bethesda, MD 20892, USA.
Molecular Cancer Therapeutics
|March 7, 2007
Summary
Transcript profiling accurately predicts protein expression in cancer cells. Combining data from multiple microarray platforms enhances prediction accuracy, showing transcript levels are useful for understanding protein information.
Area of Science:
- Genomics
- Proteomics
- Bioinformatics
Background:
- The NCI-60 cancer cell panel provides a valuable resource for studying cancer biology across nine different tissues of origin.
- Understanding the relationship between transcript and protein expression is crucial for cancer research and therapeutic development.
Purpose of the Study:
- To evaluate the utility of transcript profiling for predicting protein expression levels within the NCI-60 cancer cell line panel.
- To compare the predictive power of different microarray platforms and assess the performance of transcript versus protein data for functional prediction.
Main Methods:
- Generation of new NCI-60 transcript profile data sets using Affymetrix HG-U95 and HG-U133A chips.
- Creation of a new NCI-60 protein profile data set using reverse-phase protein lysate arrays.
- Development of a consensus transcript profile set from four microarray platforms and application of the nearest shrunken centroid algorithm for functional prediction.
Main Results:
- A consensus set of transcript profiles showed statistically significant transcript-protein correlation for 65% of genes, with generally higher correlations than previously reported.
- The consensus mRNA set outperformed individual mRNA platforms in predicting tissue of origin.
- Protein data showed a slight advantage over consensus mRNA data (P = 0.027) in this study, though both performed well.
- Gene Ontology analysis indicated that mRNA levels effectively predicted protein levels for structure-related genes (mean r = 0.71).
Conclusions:
- Transcript profiling, especially when integrating data from multiple platforms, is a valuable tool for predicting protein expression levels in cancer cells.
- Despite the utility of protein data, transcript-based technologies remain essential due to their maturity and scalability for assessing large numbers of genes.
- These findings support the continued use of transcript profiling in cancer research, even when the ultimate goal is to understand protein-level information.
