Ray Fu

hi, 2 things i can think of: 1. what does m_list look like? 2. directly pass `metadata = as.character(metadata$seurat_clusters)` and leave out `cluster_col =`. We have not tested what happens...

Hi, you can pass reference markers as a list instead of same-length dataframe: Example below: # pbmc_markers as FindAllMarkers output gene list pbmc_input

dropbox link here https://www.dropbox.com/sh/p41l7d7pmmds38o/AAAtDjiQIw090Co_Mnw43Ylqa?dl=0 the clustering step is very very slow?

currently, it's easy to coerce GeomPointTrace/GeomLineTrace to simple dot/line plot in `ggplotly()` ``` geom2trace.GeomPointTrace

Extremely hacky way to get here: Several issues, mainly ggplotly() converts 1 ggplot object layer to 1 set of scatter to draw

by group plotting would be possible if we change data to 1 layer

agreed