Would be nice to have a few functions for filtering count tables

[sterrettJD]

For example,

• removal of low-variance features
• removal of low-count features
• grouping of perfectly co-linear features/removal of identical features

Tagging in @jboconnor13 in case you'd like to contribute any of these!

[jboconnor13]

Maybe it might be worth it to remove constant features as well?

[sterrettJD]

• removal of features that correlate with % host reads removed from a sample (https://forum.biobakery.org/t/need-for-human-sequences-in-chocophlan-database/5749) or maybe these should just be included in the model?

[sterrettJD]

Implemented the following in the HoMiCorr branch for testing on real data:
removal of low-variance features
removal of low-count features