Since we are proposing a full evaluation of sensitivity of algorithms to the distribution of covariates (e.g., cell type, developmental stages, etc) this quantification should feed into this proposal.
Going to merge this from @ejfertig. Note that this probably puts us over the word limit, so I'll have a few more light revisions to bring it back down.
Since we are proposing a full evaluation of sensitivity of algorithms to the distribution of covariates (e.g., cell type, developmental stages, etc) this quantification should feed into this proposal.