Supplemental Figure S1 from Deep Transcriptome Profiling of Multiple Myeloma Using Quantitative Phenotypes
Year: 2023
Type: supplementary-materials
Abstract: A color analogy to illustrate the advantages of spectra variables for modeling. a) Individual observations of color. b) Dimension Reduction (additive color theory), all colors can be represented using 3 quantitative RGB variables. c) Standard-use, hierarchical clustering identifies 3 groups. Each sample is categorized into a group (A, B or C). Subsequent modeling uses one categorical variable. d) Multivariable modeling with quantitative spectra variables, multiple spectra integrated directly into a multivariable analysis. Each uncorrelated variable can be assessed separately for its predictive value for an outcome. This implementation retains higher resolution because the variables are quantitative and retain integrity to the initial data. Note, lower-resolution versions of xB and xG can be achieved using hierarchical groups, but the loss of quantification may lose power. xR cannot be captured by any group ordering and associations for this spectrum would be lost using hierarchical groups. Three dimensions shown for visualization purposes; transcriptome data is likely to harbor a higher number of dimensions.
Language: English
Subfield: Hematology
Field: Medicine
Domain: Health Sciences
Open Access status: gold
OpenAlex ID: W4362533481