ABSTRACT
Domain recognition software was employed to assess recrystallization-inhibition (RI) activity as an index of antifreeze potential. This represents a key step in the development of a high-throughput analysis for RI activity. Analysis of measurement error indicates an average coefficient of variation for individual crystals of about 8%, which is very small in relation to other sources of variation. Our analysis demonstrates an inverse correlation between AFGP 8 concentration and average crystal size with consistently small, but detectable differences in average crystal size at the edge and centre of the ice wafer. Sensitivity analysis using Monte Carlo re-sampling methods indicate that measuring of 12-15 crystals per field of view are sufficient to obtain accurate estimates of the first two moments (mean and variance) of the crystal size distribution, thereby greatly reducing the time required to assess recrystallization activity. These results suggest that this method has considerable potential for high-throughput analysis of RI activity.
