extract_entropy function

extract_entropy

The extract_entropy() is used to compute spatial entropy at each distance interval for all cell types of interest. The goal is to capture the diversity in cellular composition, such as similar proportions across cell types or dominance of a single type, at a specific distance range. Additionally, spatial patterns, including clustered, independent, or regular, among cell types can also be acquired. In this example, we will look at the spatial heterogeneity across T cells, macrophages, and others. To focus on the local cell-to-cell interactions, we set the default maximum of the distance range (i.e., rmax) to be 400 microns. The default number of distance breaks/intervals is set to 50. Then, a sequence of distance breaks is generated by linearly decreasing from rmax to 0 on a log scale. At each distance range, partial spatial entropy and residual entropy are calculated as in Vu et al. (2023), Altieri et al. (2018). These spatial entropy functions can then be used as input functions for FPCA.

extract_entropy(mxFDAobject, markvar, marks, n_break = 50, rmax = 400)

Arguments

• mxFDAobject: object of class mxFDA
• markvar: The name of the variable that denotes cell type(s) of interest. Character.
• marks: Character vector that denotes cell types of interest.
• n_break: Total number of distance ranges/intervals of interest made from 0 to rmax for calculating entropy
• rmax: Max distance between pairs of cells

Returns

object of class mxFDA with a dataframe in the multivariate_summaries slot