Precise liquid handling is not enough; the instrumentation must also dispense accurate volumes.
Imagine that you run a primary screening laboratory at a biopharmaceutical company. You use sophisticated automated liquid delivery instrumentation to run anywhere from 100,000 to one million or more compounds per year in assays that can cost up to $100,000 per plate. Your facility is regulated by stringent SOPs, your data could be audited by the FDA, and you’re in a fiercely competitive race to bring new drugs to market. There is clearly no room for error.
Because liquid handling is a core component of your drug discovery processes, qualification of automated liquid delivery devices, validation of assays that depend on liquid delivery steps, and a standardized measurement method are all necessary.
Information about precision of the instrument is important, but it only provides details about the closeness of the dispensed volumes to each other. Information on the accuracy of the instrument, on the other hand, determines the deviation of the dispensed volumes from the target volume. For example, if an instrument programmed to dispense 100 microliters consistently dispenses 115 microliters, the instrument may be operating with perfect precision, even though it is inaccurate and is over-dispensing.
To satisfy customer demands for accuracy and precision data, and to provide stronger assurance of the integrity of results produced by its liquid handling equipment, Beckman Coulter upgraded its Field Service Operational Qualification Program for its Biomek automated liquid handlers with the ARTEL MVS (Multichannel Verification System), which is based on ratiometric photometry, a technology measuring the absorbance of light by two proprietary dye solutions.
Operational qualification
Operational qualification proves that equipment performs within specifications and complies with regulations. Third-party validation of equipment performance strengthens data integrity and facilitates compliance.
Beckman Coulter offers three levels of instrument operational qualification; OQ3, the highest level provides for certification and system performance tests using the ARTEL MVS to determine whether the liquid handlers are performing within manufacturer’s specifications, which are typically tighter than regulations require.
First, pre-qualification is conducted to determine the as-found performance based on both accuracy and precision dispensing. Tests are run at both high and low volumes, with target volumes determined by customer applications and type of head employed (Figures 1 and 2). The instrument or method is adjusted so it operates within the desired tolerances. Preventive maintenance is conducted as needed. The final step is an as-left qualification, where service representatives ensure instrument performance within set tolerances and provide documentation proving that the instrument is operational.
Figure 1. Biomek NX Multichannel P200 head qualification
Volumes for Plate 1 (µL): (Source: Beckman Coulter) |
||||||||||||
|
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
11 |
12 |
A |
98.97 |
98.79 |
99.25 |
99.16 |
98.88 |
99.25 |
98.79 |
99.15 |
98.14 |
99.25 |
99.44 |
99.81 |
B |
98.69 |
99.06 |
98.69 |
98.70 |
98.59 |
98.24 |
98.69 |
98.51 |
99.06 |
99.90 |
98.97 |
99.25 |
C |
98.33 |
98.41 |
98.50 |
99.94 |
98.68 |
97.68 |
98.89 |
98.41 |
98.41 |
99.16 |
99.07 |
99.34 |
D |
98.69 |
98.79 |
99.24 |
99.15 |
99.34 |
98.07 |
97.85 |
98.23 |
98.42 |
98.98 |
99.43 |
98.79 |
E |
97.96 |
98.23 |
99.34 |
99.05 |
98.78 |
98.05 |
98.78 |
98.05 |
98.06 |
99.16 |
99.43 |
97.96 |
F |
98.14 |
98.64 |
98.78 |
98.61 |
98.79 |
97.77 |
98.78 |
97.50 |
97.96 |
98.97 |
99.35 |
98.61 |
G |
98.15 |
98.05 |
98.32 |
98.70 |
98.33 |
97.95 |
98.33 |
98.05 |
98.24 |
98.79 |
98.61 |
98.89 |
H |
98.13 |
98.42 |
98.51 |
98.06 |
98.14 |
98.24 |
98.89 |
98.33 |
98.43 |
98.24 |
98.62 |
98.89 |
Plate 1 Statistics
Mean Volume (µL): 98.66
Relative Inaccuracy: -1.34%
Standard Deviation (µL): 0.51
Coefficient of Variation: 0.52%
Previously, Beckman Coulter employed a photometric method using a bead-based solution to verify that equipment was performing within precision specifications, as determined per instrument for both ends of the volume spectrum. However, there was no proof to confirm that the instrument was dispensing accurately. This technology was problematic at low temperatures and service technicians had to manually record the results.
Traditionally, laboratories have used gravimetry to verify the accuracy and precision of liquid delivery devices. However, this method, which verifies volume by weighing liquid quantities on a balance, is time consuming and prone to environmental and manual error, especially at small volumes. In addition, gravimetric calibration is not ideal for multichannel devices because it only provides an aggregate assessment of all channels as opposed to information regarding how each individual channel is performing.
Figure 2. Span 8 with 1 mL tip qualification (inaccurate well is highlighted in orange)
Group 2 Well Volumes (µL): (Source: Beckman Coulter) |
||||||
|
7 |
8 |
9 |
10 |
11 |
12 |
A |
100.0 |
100.6 |
100.4 |
100.5 |
100.9 |
100.5 |
B |
100.6 |
101.4 |
100.6 |
100.2 |
100.1 |
99.9 |
C |
100.9 |
100.9 |
100.3 |
100.7 |
100.3 |
99.3 |
D |
95.8 |
101.6 |
102.2 |
101.7 |
101.0 |
100.6 |
E |
101.2 |
101.2 |
101.1 |
101.9 |
100.9 |
100.3 |
F |
101.3 |
100.9 |
100.7 |
100.6 |
100.5 |
99.7 |
G |
101.3 |
101.5 |
100.9 |
100.9 |
101.0 |
99.8 |
H |
100.5 |
99.7 |
99.9 |
99.9 |
98.9 |
98.8 |
Group 2 Channel Statistics: |
|||||
Channel |
Mean Volume |
Inaccuracy |
Standard Deviation |
CV |
Status |
1 |
100.5 |
0.5% |
0.3 |
0.3% |
Passed |
2 |
100.5 |
0.5% |
0.5 |
0.5% |
Passed |
3 |
100.4 |
0.4% |
0.6 |
0.6% |
Passed |
4 |
100.5 |
0.5% |
2.4 |
2.4% |
Passed |
5 |
101.1 |
1.1% |
0.5 |
0.5% |
Passed |
6 |
100.6 |
0.6% |
0.5 |
0.5% |
Passed |
7 |
100.9 |
0.9% |
0.6 |
0.6% |
Passed |
8 |
99.6 |
-0.4% |
0.7 |
0.7% |
Passed |
The ARTEL MVS is based on proprietary ratiometric photometry, which employs a dual-dye, dual-wavelength absorbance method for accurate and precise measurement of small target volumes. It relies on two colorimetric dyes with distinct absorbance maxima at 520 nm (red) and 730 nm (blue). Using calculations derived from the Beer-Lambert Law, the MVS is used to quantify both the precision and accuracy of each dispensing channel in one experiment by measuring the absorbance of these dyes.
The Beer-Lambert law states that when light passes through a solution containing some concentration of dye, the amount of light absorbed by the dye solution is proportional to both the concentration of dye and the interaction path length of the light with the solution. If both the molar absorptivity and concentration of the dye solution are known and closely controlled, which is the case with the MVS, the law can be used to determine an unknown path length traversed by a light beam. By measuring the path length through the solution and knowing the dye concentration, the unknown volume can be calculated through a series of equations.
This process is accomplished by the determination of the absorbance per unit path length of the red and blue dyes. This NIST-traceable information is recorded on a barcode on each reagent bottle, further reducing measurement variability and uncertainty. Barcodes on the system’s microtiter plates, sample solutions, and calibrator plate contain performance information that is passed to the system’s software through a barcode reader. The plate reader collects photometric measurements of the dye solutions dispensed into the microtiter wells by the liquid delivery device. These measurements, together with the barcode information, are used to rapidly determine both the precision and accuracy of the volume delivered from each tip of the device being tested.
The measurement results are also traceable to international standards, allowing comparability of all Beckman Coulter devices regardless of model, location, or number of dispensing channels.
The MVS allows Beckman Coulter to provide more information about liquid delivery device performance, facilitating assay validation and method transfer. The automated documentation feature eliminates the time-consuming and error-prone steps of manually recording results. Malfunctioning channels to be immediately identified and corrected.
The Importance of Accuracy
A Beckman Coulter customer using a Biomek liquid handling system found inconsistencies in assay results when automating a manually developed method. The application required the addition of 15 microliters of the assay’s rate-limiting reagent, a critical determinant of the assay results. The Beckman Coulter service representative ran the critical assay volume through the MVS and determined that the instrument was operating well within specifications, dispensing a volume of 14.84 microliters, producing accuracy results within 1 percent and precision results within 1.3 percent at 10 microliters.
However, the automated assay was returning very different results than was the manually developed method. After evaluating the manually pipetted results using the MVS, it was determined that the manual pipetting process was dispensing 15.75 microliters, five percent over the target volume. The automated liquid handler was adjusted to over-pipette by the same amount and it produced the same assay results as the manually pipetted method.
About the Authors
Keith Albert is responsible for automated liquid delivery system performance research, customer education and assay/method validation services associated with ARTEL’s LHQA (Liquid Handling Quality Assurance) services.
Julie Stanis Farias is the technical support engineer for the Biomek liquid handler product line and created the OQ3 service program.
Angelica Olcott is the product manager for laboratory research automation. She manages the Biomek liquid handler product line and Biomek consumables.
Group 2 Well Volumes (µL): (Source: Beckman Coulter) | ||||||
|
7 |
8 |
9 |
10 |
11 |
12 |
A |
100.0 |
100.6 |
100.4 |
100.5 |
100.9 |
100.5 |
B |
100.6 |
101.4 |
100.6 |
100.2 |
100.1 |
99.9 |
C |
100.9 |
100.9 |
100.3 |
100.7 |
100.3 |
99.3 |
D |
95.8 |
101.6 |
102.2 |
101.7 |
101.0 |
100.6 |
E |
101.2 |
101.2 |
101.1 |
101.9 |
100.9 |
100.3 |
F |
101.3 |
100.9 |
100.7 |
100.6 |
100.5 |
99.7 |
G |
101.3 |
101.5 |
100.9 |
100.9 |
101.0 |
99.8 |
Filed Under: Drug Discovery