Internet-based Profiler system as integrative framework to support translational research.

TitleInternet-based Profiler system as integrative framework to support translational research.
Publication TypeJournal Article
Year of Publication2005
AuthorsKim R, Demichelis F, Tang J, Riva A, Shen R, Gibbs DF, Mahavishno V, Chinnaiyan AM, Rubin MA
JournalBMC Bioinformatics
Volume6
Pagination304
Date Published2005
ISSN1471-2105
KeywordsAutomation, Biological Markers, Computational Biology, Database Management Systems, Databases, Genetic, Gene Expression Profiling, Humans, Image Processing, Computer-Assisted, Internet, Neoplasms, Oligonucleotide Array Sequence Analysis, Programming Languages, Software, Software Design, Tumor Markers, Biological
Abstract

BACKGROUND: Translational research requires taking basic science observations and developing them into clinically useful tests and therapeutics. We have developed a process to develop molecular biomarkers for diagnosis and prognosis by integrating tissue microarray (TMA) technology and an internet-database tool, Profiler. TMA technology allows investigators to study hundreds of patient samples on a single glass slide resulting in the conservation of tissue and the reduction in inter-experimental variability. The Profiler system allows investigator to reliably track, store, and evaluate TMA experiments. Here within we describe the process that has evolved through an empirical basis over the past 5 years at two academic institutions.

RESULTS: The generic design of this system makes it compatible with multiple organ system (e.g., prostate, breast, lung, renal, and hematopoietic system,). Studies and folders are restricted to authorized users as required. Over the past 5 years, investigators at 2 academic institutions have scanned 656 TMA experiments and collected 63,311 digital images of these tissue samples. 68 pathologists from 12 major user groups have accessed the system. Two groups directly link clinical data from over 500 patients for immediate access and the remaining groups choose to maintain clinical and pathology data on separate systems. Profiler currently has 170 K data points such as staining intensity, tumor grade, and nuclear size. Due to the relational database structure, analysis can be easily performed on single or multiple TMA experimental results. The TMA module of Profiler can maintain images acquired from multiple systems.

CONCLUSION: We have developed a robust process to develop molecular biomarkers using TMA technology and an internet-based database system to track all steps of this process. This system is extendable to other types of molecular data as separate modules and is freely available to academic institutions for licensing.

DOI10.1186/1471-2105-6-304
Alternate JournalBMC Bioinformatics
PubMed ID16364175
PubMed Central IDPMC1343596
Grant ListCA 97063 / CA / NCI NIH HHS / United States
P50CA69568 / CA / NCI NIH HHS / United States
R01AG21404 / AG / NIA NIH HHS / United States