Clarification on Reference Samples for WisecondorX

[ilya90906767]

I’m planning to use the WisecondorX tool for analyzing 30,000 genomes. However, I don’t initially know whether there are samples with aneuploidy in this dataset.

Should I create the reference from all 30,000 samples and then make predictions based on that reference? Or would it be better to manually select 20-100 normal samples first and create the reference from them?