search

sjspielman / dragon

Deep time Redox Analysis of the Geobiology Ontology Network
https://sjspielman.shinyapps.io/dragon/
Other
2 stars 1 forks source link

Color Keys of Network Clusters and Analysis Tukey Plot Clusters are sometimes Different #20

Closed mooreek closed 5 years ago

mooreek commented 5 years ago

Hi,

It appears that the color of the network clusters are different on the network figure vs. the analysis plot when network clusters are chosen as the predictor variable. For example, when I make an Arsenic (As) network - cluster 1 is red for the network and plot (same), but cluster 7 appears to be teal in the network and purple in the plot (different). Do you know why this could be happening, or am I just reading the keys wrong? I am using Leading Eigen as the clustering method in this case. When I use Louvain clustering the colors are the same for the network and the plot.

When I use Leading Eigen for the As network there are 14 network clusters produced, could this be complicating the color key for the plot?

Thanks, Eli

sjspielman commented 5 years ago

Currently there is no link between colors in the network and in the plot. I should be able to link them up, though. I'll add this to my to-do new features list.

sjspielman commented 5 years ago

Colors should now be consistent.

mooreek commented 5 years ago

Yes, the colors look consistent now. Thanks!