Neuro-Oncology Branch, National Cancer Institute, National Institutes of Neurological Disorder and Stroke, National Institutes of Health, Bethesda, MD 20892, USA.
BMC Med Inform Decis Mak. 2010 Jul 15;10:38. doi: 10.1186/1472-6947-10-38.
Advances in generating genome-wide gene expression data have accelerated the development of molecular-based tumor classification systems. Tools that allow the translation of such molecular classification schemas from research into clinical applications are still missing in the emerging era of personalized medicine.
We developed GliomaPredict as a computational tool that allows the fast and reliable classification of glioma patients into one of six previously published stratified subtypes based on sets of extensively validated classifiers derived from hundreds of glioma transcriptomic profiles. Our tool utilizes a principle component analysis (PCA)-based approach to generate a visual representation of the analyses, quantifies the confidence of the underlying subtype assessment and presents results as a printable PDF file. GliomaPredict tool is implemented as a plugin application for the widely-used GenePattern framework.
GliomaPredict provides a user-friendly, clinically applicable novel platform for instantly assigning gene expression-based subtype in patients with gliomas thereby aiding in clinical trial design and therapeutic decision-making. Implemented as a user-friendly diagnostic tool, we expect that in time GliomaPredict, and tools like it, will become routinely used in translational/clinical research and in the clinical care of patients with gliomas.
全基因组基因表达数据的生成技术的进步加速了基于分子的肿瘤分类系统的发展。在个性化医疗的新兴时代，仍然缺乏将此类分子分类方案从研究转化为临床应用的工具。
我们开发了GliomaPredict 作为一种计算工具，它可以根据从数百个胶质瘤转录组谱中得出的经过广泛验证的分类器集，快速可靠地将胶质瘤患者分为六个已发表的分层亚型之一。我们的工具利用基于主成分分析（PCA）的方法生成分析的可视化表示，量化潜在亚型评估的置信度，并以可打印的 PDF 文件形式呈现结果。GliomaPredict 工具作为广泛使用的 GenePattern 框架的插件应用程序实现。
GliomaPredict 为胶质瘤患者提供了一个用户友好的、临床适用的新型平台，可立即基于基因表达进行亚型分配，从而有助于临床试验设计和治疗决策。作为一个用户友好的诊断工具实施，我们预计随着时间的推移，GliomaPredict 和类似的工具将成为胶质瘤患者的转化/临床研究和临床护理的常规手段。
