Hello everyone, I want to try Dunn Index to validate my clustering results from scRNA-Seq data. I know the Dunn index starts from 0 and goes to infinity and higher results mean better clustering. I tried this method from scRNA-Seq which clustered with buildSNNGraph() function of Scater/Scran package and the graph is clustered with Louvain algorithm of igraph package. I tried range of k values and want to score them. Most of the Dunn indexes are between 0.08 and 0.1. Can I use these values to compare my clustering results or Dunn index is working for methods of distance based clustering rather than graph based clustering methods?

Thank you in advance

What about comparing the Louvain modularity scores across your different k values? In principle, Dunn index can be applied to assess clustering results for any method that groups items, presuming you can compute distances between the items. Calculating the Dunn index on your clusters would tell you something about the inter-vs-intra cluster distances in your clustering. However, your clusters were created based on Louvain optimization of modularity in the weighted nearest neighbor graph. So your clustering method was optimizing a different quantity (density of linkages) than what the Dunn index evaluates (ratio of distances). I think you'd have to decide which measure you consider to be the best way to assess of the 'quality' of your clustering.

Thanks for your explanation and comment.

Cross-posted on Bioconductor: https://support.bioconductor.org/p/118905/

I did not know it is a problem. I will try to delete my post here.

Thank you.

You do not have to delete it. Users will now go to the other site and get the answer there. Thanks