In the first part of this article, the difference between precision and accuracy and its significance was discussed. The second part takes a closer look at these two performance indicators.

Expert mode: Evaluating precision and accuracy in intra-run, inter-run and tip-to-tip dimensions

In order to enable fast and effective identification of sources of error, manufacturers and operators of liquid handlers require an even more systematic and differentiated assessment of liquid handling performance.

This can be achieved by breaking down liquid handling, for example on microplates, to the following very basic operations:

Three possible subsets of a data set including L × M × N measurements:

1) One dispensing channel dispenses one run of N aliquots for the intra-run experiments.

2) One dispensing channel dispenses M runs of N individual liquid volumes per run for the evaluation of inter-run performance.

3) L dispensing channels dispense one run each of N individual liquid volumes for tip-to-tip experiments.

The characterization of dispensing performance within the four levels of (1) intra-run, (2) inter-run, (3) tip-to-tip and (4) load-to-load performance enables a systematic, detailed and above all differentiated analysis of precision, accuracy and possible sources of error. (1) Intra-run measurements primarily provide information about the fundamental precision and accuracy of a single dispensing channel of a liquid handler dispensing aliquots in a continuous run. Errors incorporated by pause times or the usage of different tips are excluded from this analysis. (2) Inter-run measurements reveal errors due to pause time effects and evaluate the reproducibility or stability of a system. Delay time between any two runs for the purpose of characterization can be adjusted to the anticipated pause times of the respective application, ranging from a few seconds to hours or days. (3) Tip-to-tip measurements evaluate sources of errors due to dispensing channel variances (e.g. different cartridges, nozzle variances, tubing variances, differing pressure levels for different dispensing channels among others). (4) Load-to-load measurements assess sources of error due to loading operations using, for example, a cartridge or a disposable tip.