Elizabeth Purdom

use fastcluster for hierarchical clustering

5

http://danifold.net/fastcluster.html Should probably be as simple as calling the hclust from this package rather than stats.

Use pheatmap instead of aheatmap?

4

Look into using pheatmap which is like aheatmap but puts names on side (like my hack does, but built into the function, and also for genes).

Currently, the code for plotTracking treats "-1","-2" as any other cluster for the purposes of aligning the clusters and giving them colors, and then does a post-processing change to the...

Warnings in mergeClusters

2

I get this warning in the vignette: ce

To use `distFun` with clusterMany, user gives character names of functions and they are looked up in global.env by the internal function .makeDiss. This is problematic and not good use...

If user chooses findBestK=c(TRUE,FALSE) with a range of values, e.g. `ks=4:15` we are extremely inefficient, since we run each for 4-15 with findBestK=FALSE, and then for findBestK=TRUE, we RERUN all...

Fix getClusterManyParams so can subset to certain parameter values at the same time. Tricky part is what the input would look like, since don't want to have to list all...

dimReduce and scaling

2

If do not do PCA for dimReduce, should we have ability to scale the data before clustering? An option with dimReduce? Except you might want to both filter genes and...

It would be nice if could get plotHeatmap to only show legend for specific factors, not all. This is probably issue with aheatmap.