Team Leaders: Ryan McMinds, Jesper Madsen
Team Members: Ann Mathew, Sofia Bhatia, Samuel Coleman, Omkar Dokur, Janelle Donglasan, Jan Dahrendorff
GVN/USF mentors: Dr. Alash'le G. Abimiku
Linking geospatial datasets with viral evolution rates
1st pass:
-
download NextStrain US subset phylogenies, including time-scaled and divergence-scaled branch lengths, and geographic locations of tips.
-
impute mean geographic location of edges.
-
first impute ancestral nodes as a point on the shortest great-circle arc between the direct descendant nodes/tips. start at tips and iterate toward root of tree. The point for each ancestor corresponds to the location proportional to time-scaled branch lengths (e.g. if one descendant edge has length 2, and the other has length 1, the point is 1/3 the distance from descendant 1 to descendant 2.
-
then impute mean location for edges as midpoint between the ancestor and descendant
-
input imputed locations/distances into gaussian process or inla model
-
inla/gp model conditioned on geographic dataset, with imputed values at locations corresponding to the estimated location of phylogenetic edges
The following packages are required to run Viral SpaceTime:
RStudio
R-packages:
- library(shiny)
- library(leaflet)
- library(leaflet.extras)
- library(dplyr)
- library(DT)
Download and open the "app.R" file in RStudio. In the upper-right corner of RStudio, press the "Run App" button.
