Open AccessMultidimensionality of microarrays: Statistical challenges and (im)possible solutions
Author(s) -
Michiels Stefan,
Kramar Andrew,
Koscielny Serge
Publication year - 2011
Publication title -
molecular oncology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.332
H-Index - 88
eISSN - 1878-0261
pISSN - 1574-7891
DOI - 10.1016/j.molonc.2011.01.002
Subject(s) - curse of dimensionality , data science , dna microarray , computer science , curse , interpretation (philosophy) , statistical analysis , focus (optics) , computational biology , data mining , econometrics , bioinformatics , biology , machine learning , statistics , mathematics , genetics , gene expression , physics , sociology , anthropology , optics , gene , programming language
A typical array experiment yields at least tens of thousands of measurements on often not more than a hundred patients, a situation often denoted as the curse of dimensionality. With a focus on prognostic multi‐biomarker scores derived from microarrays, we highlight the multidimensionality of the problem and the issues in the multidimensionality of the data. We go over several statistical challenges raised by this curse occurring in each step of microarray analysis on patient data, from the hypothesis and the experimental design to the analysis methods, interpretation of results and clinical utility. Different analytical tools and solutions to answer these challenges are provided and discussed.