Poster: Verification of Liquid Handling Parameters
Verification of Liquid Handling Parameters is Critical to Biological Assay Performance
Poster Presentation at Miptec 2012
Stefan Schork1, Tanya Knaide1 and Nathaniel Hentz2
1 Artel, 25 Bradley Drive, Westbrook, ME, USA 2Golden LEAF Biomanufacturing Training and Education Center, NC State University, Raleigh, NC, USA
Selection of biologically active compounds by means of high throughput screening (HTS) is a well-established but demanding process through which compounds are often tested just once before being selected for further analysis. Scientists in HTS labs are also frequently challenged with improving efficiency by implementing new technologies and complex biology. Therefore, in order to identify promising drug candidates, a high degree of accuracy and quality in the assays is critical.
Assay variability can dramatically affect the selection of drug candidates and is influenced by the nature of the assay methodology, instrumental performance variation and random errors. Liquid handling variability is often underestimated, yet can have a serious impact on the result of an assay. Laboratories often rely on only precision to evaluate the quality of their liquid handling equipment and methods. Optimization of liquid handling parameters is so crucial to improve the performance of an assay that both accuracy and precision must be assessed.
The present study was conducted to investigate tools for liquid handling quality control (QC) and to determine the effect of liquid handling variability on a model biological assay. The Artel Multichannel Verification System (MVS) was used as the tool for evaluating quality control because of its capability to assess the performance of automated liquid handlers and because it provides traceable data for both accuracy and precision. This study evaluated various types of microplates compatible with the MVS. To minimize perturbation introduced by the liquid handler, a handheld automated analytical syringe (eVol, SGE Analytical Science) was used to assess the influence of the plates. Testing at different volumes (10 μL to 200 μL) revealed that the Artel Verification plates performed best in terms of both accuracy (-0.1 to 0.74 % inaccuracy) and precision (0.15 to 0.3 %CV). In contrast, alternate off-the-shelf plates yielded drastically different accuracy and precision profiles, which could lead to improper calibration of a liquid handler.
Similar results were obtained when the experiment was conducted with an automated system (Precision XS, BioTek) that better corresponds to the reality in a screening laboratory.
Finally, this study illustrates how an incorrect calibration of a liquid handler can influence the assay quality. Data presented shows that only small changes in delivered volumes can affect IC50 values of a model biological assay.