Title :
Detection of heterogeneously expressed genes in cancer
Author :
Haney, Scott ; Kam, Moshe ; Hrebien, Leonid
Author_Institution :
Data Fusion Lab., Drexel Univ., Philadelphia, PA
Abstract :
Several mutations have been found to be heterogeneous throughout cancer samples. As an example, approximately 20% of lung cancer cases have a mutation in the proto-oncogene RAS. In order to specifically test for heterogeneous mutations we introduce a new test that measures differences between quantiles. It will be shown that this test can achieve larger power values than a previously published test by Coombes et al. when sample sizes are small (10-20 samples) and the probability of mutation is small (10%-40%). When applied to two publicly available microarray data sets our method also produces lists of results that are 12% and 27% the size of those produced by Coombes et al.
Keywords :
cancer; genetics; lung; medical computing; gene mutations; heterogeneous mutations; heterogeneously expressed gene detection; lung cancer; microarray data sets; mutation probability; proto-oncogene RAS mutation; Bioinformatics; Cancer detection; Diseases; Genetic mutations; Genomics; Humans; Laboratories; Lungs; Probability density function; Testing;
Conference_Titel :
Genomic Signal Processing and Statistics, 2008. GENSiPS 2008. IEEE International Workshop on
Conference_Location :
Phoenix, AZ
Print_ISBN :
978-1-4244-2371-2
Electronic_ISBN :
978-1-4244-2372-9
DOI :
10.1109/GENSIPS.2008.4555657