This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Molecular Classification of Cancer: Class Discovery and Class Prediction by Gene Expression Monitoring T. R. Golub, 1,2 * ² D. K. Slonim, 1 ² P. Tamayo, 1 C. Huard, 1 M. Gaasenbeek, 1 J. P. Mesirov, 1 H. Coller, 1 M. L. Loh, 2 J. R. Downing, 3 M. A. Caligiuri, 4 C. D. Bloomfield, 4 E. S. Lander 1,5 * Although cancer classification has improved over the past 30 years, there has been no general approach for identifying new cancer classes (class discovery) or for assigning tumors to known classes (class prediction). Here, a generic approach to cancer classification based on gene expression monitoring by DNA microarrays is described and applied to human acute leukemias as a test case. A class discovery procedure automatically discovered the distinction between acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL) without previous knowledge of these classes. An automatically derived class predictor was able to determine the class of new leukemia cases. The results demonstrate the feasibility of cancer classification based solely on gene expression moni- toring and suggest a general strategy for discovering and predicting cancer classes for other types of cancer, independent of previous biological knowledge. The challenge of cancer treatment has been to target specific therapies to pathogenetically distinct tumor types, to maximize efficacy and minimize toxicity. Improvements in can- cer classification have thus been central to advances in cancer treatment. Cancer classi- fication has been based primarily on morpho- logical appearance of the tumor, but this has serious limitations. Tumors with similar his- topathological appearance can follow signif- icantly different clinical courses and show different responses to therapy. In a few cases, such clinical heterogeneity has been ex- plained by dividing morphologically similar tumors into subtypes with distinct pathogen- eses. Key examples include the subdivision of acute leukemias, non-Hodgkin’s lympho- mas, and childhood “small round blue cell tumors” [tumors with variable response to chemotherapy ( 1 ) that are now molecularly subclassified into neuroblastomas, rhabdo- myosarcoma, Ewing’s sarcoma, and other types ( 2 )]. For many more tumors, important subclasses are likely to exist but have yet to be defined by molecular markers. For exam- ple, prostate cancers of identical grade can have widely variable clinical courses, from indolence over decades to explosive growth causing rapid patient death. Cancer classifi- cation has been difficult in part because it has historically relied on specific biological in- sights, rather than systematic and unbiased approaches for recognizing tumor subtypes....
View Full Document
This note was uploaded on 07/08/2011 for the course CS 101 taught by Professor Khliu during the Spring '11 term at Xiamen University.
- Spring '11
- Machine Learning