I'm out of practice with this and would appreciate even quick and dirty advice! I have two compound libraries for screening, A and B. Library A has already been screened. Scientists are now trying to decide whether to also screen B, but as part of that decision they'd like some metric describing how much more chemical space coverage they'd get by screening library B. I assume we're happy using some fingerprint-based metric. I see straightforward ways to calculate internal diversity within a library, but I'm not sure OTTOMH how that translates to comparing multiple libraries. How about counting unique fingerprint bits in B that are not also in A, then dividing by total count of unique fingerprint bits in A+B to get an estimate of the extra percentage of stuff in B?

Thanks!

Chris