Biomedical Engineering Reference
In-Depth Information
Figure 16.5
(
a
) Basic statistics can be generated for two cohorts when two concepts
are selected from the study ontology using drag-and-drop paradigm. (
b
) These selected
concepts will defi ne the subjects, which then form the basis of a simple statistical cal-
culation, which then is presented as a set of plots showing distribution of age, gender,
and race for the two cohorts.
Figure 16.6
Phenotype distribution for two cohorts can be calculated once the two
cohorts are defi ned. Through a simple drag-and-drop action, the scientist can identify
the concept for which the distribution is calculated and plotted across the two cohorts
as well as calculate and plot the appropriate statistics. For continuous variables
t
-statistics and for categorical variables
χ
2
- statistics are calculated.
Figure 16.7
If the concept of interest is a biomarker such as gene expression or
protein expression, heatmaps can be generated to show the distribution of the levels
across the samples of the two cohorts.
information about genes, pathways, compounds, diseases, drug trials, or com-
binations of therein. The underlying data consist of analyzed proprietary non-
clinical and clinical biomarker studies, including gene expression and
proteomics data. Additionally, data from public sources were added to the
warehouse, including data sets from the National Center for Biotechnology
Information (NCBI) GEO [19], EBI Array Express [23], Dana Farber Cancer
Search WWH ::
Custom Search