I'm trying to cluster this genetic data but even with multiple different methods k-means/medoids spectral, etc... I can't seem to resolve the two big clusters. Any suggestions? I would really like to be able to identify that central cluster around y=1. The goal is to run this on multiple datasets.