dylkot / cNMF

Code and example data for running Consensus Non-negative Matrix Factorization on single-cell RNA-Seq data
MIT License
243 stars 57 forks source link

stability/error diagnostic plot interpretation #50

Closed ejc043 closed 1 year ago

ejc043 commented 1 year ago

Hello! Thank you so much for making such a wonderful package. I was hoping to receive some support for the stability/error diagnostic plot interpretation to select K. For example, given the following plot:

image

Why was K=24 selected instead of K=22? I know the paper affirms that programs are pretty stable across a range of K's but I was interested in the team's explanation. Thank you!!