Medical researchers often seek to identify biomarkers that are associated with certain clinical outcome. The challenge is that biomarkers tend to be numerous, often inter-correlated, and with varying degrees of association with the clinical outcome. We develop a straightforward yet effective visualization approach that allows researcher to see the inter-correlations of markers and their association with clinical outcomes rendered in a 2-D plot.
We first extract differences between outcome groups and apply t-Distributed Stochastic Neighbor Embedding method to reduce dimensionality of the data into 2-D data, which are then rendered as 2-D plots. Both original marker values and their ranking in the sample can be used for visualization.
|