Reproducibility of crystal formation.

“-”, not applicable. Mean and standard deviation were calculated using Microsoft Excel. Twelve replicates were set up for each protein/nucleant combination tested, therefore the maximum possible number for the mean per protein/nucleant combination is 12. The no-nucleant control experiment to test reproducibility produced no crystals in any of the drops (therefore the mean±standard deviation per protein/nucleant combination for the no-nucleant control is 0. 0±0.0 in all cases).