How might you convert the relative abundance measures to pseudo-RPKs that are sensitive to each sample's read depth?

[nousiaso]

This interesting question about using graphlan with different samples…

Thanks a lot in advance for the help!