It is imperative to understand that the standard curve run at different times will not yield the exact OD values for every run. The reason for this variation includes inter-operator and assay such as the incubation conditions and pipettors used. Thus, the OD@450 nm value for a point on the standard curve on one given day may not be the same on the next day. To obtain accurate results in calculating concentrations of unknown samples, the interpolation of data must not be done using a standard curve from a previous run. A separate standard curve should be generated for every run along with unknown samples to yield reliable data.
These findings demonstrate inter-operator variability by comparing separate assays performed by two operators using the same lot of an ELISA kit. Additionally, to assess inter-assay variability, we compare the results of separate runs of ELISA by the same operator using the same lot from the ELISA kit.
Methods
Inter-Operator variability: Standard curves were prepared in duplicates by three different operators on three different dates following the kit protocol for product 42400-1 Verikine Mouse Interferon Beta ELISA Kit, lot 3700.
Inter-assay variability: To assess inter-assay variability three different standard curves were prepared in duplicates by the same operator on two different dates using Verikine Mouse Interferon Beta ELISA Kit lot 3765.
After the completion of the assay, the plates were read at OD 450 nm using Vmax Kinetic Plate Reader (Molecular Devices Corporation, CA, USA). The data points comprising each standard curve were fitted by 4-parameter fit using SoftMax Pro ver 5.0 (Molecular Devices Corporation, CA, USA).
Results
Inter-Operator Variability
Figure 1: Standard curves were prepared by three different operators on three different days. The curves were fitted by 4-parameter fit of the data points
Green Curve: Run on 6.14.07 by Operator 1
Blue Curve: Run on 7.6.07 by Operator 2
Red Curve: Run on 7.18.07 by Operator 3

Table 1: Mean OD values of the data points on the three standard curves prepared by three different operators
Curve Point
Mean OD@450 nm
S7
S6
S5
S4
S3
S2
S1
Concentration (pg/ml)
Operator 1 (green curve)
Operator 2 (blue curve)
Operator 3 (red curve)
1000
500
250
125
62.5
31.25
15.625
3.239
1.874
1.127
0.55
0.394
0.26
0.192
2.32
1.293
0.863
0.523
0.341
0.222
0.161
3.009
1.74
0.926
0.637
0.327
0.227
0.146
Figure 2: Inter-Assay Variability, Standard curves prepared by same operator on two different days.
Blue Curve: Run on 11.30.07 by Operator 2
Red Curve: Run on 11.30.07 by Operator 2 in parallel with run1 (blue curve) on a different plate
Green Curve: Run on 12.7.07 by Operator 2

Table 2: Mean OD values of the data points on the three standard curves prepared by the same operator.
Curve Point
Mean OD@450 nm
S7
S6
S5
S4
S3
S2
S1
Concentration (pg/ml)
Run 1 (Blue curve)
Run 2 (Red curve)
Run 3 (Green curve)
1000
500
250
125
62.5
31.25
15.625
3.498
2.035
1.353
0.746
0.512
0.341
0.278
3.74
2.331
1.494
0.915
0.609
0.455
0.335
3.561
1.955
1.378
0.836
0.567
0.403
0.331
Conclusion
From Table 1, it is evident that the mean OD values for the data points in the standard curve are different for curves generated by different operators. For example, the Mean OD values for point S7 are 3.239, 2.32 and 3.009 when the curves are generated by Operator 1, Operator 2 and Operator 3, respectively. As a result, the OD value for an unknown sample may also vary when tested by different operators. However, since the concentration of the sample is calculated using the standard curve, any operator-induced variation can be negated. This occurs if the concentration of the sample is interpolated from the standard curve prepared by the same operator on the same plate on which the sample is being tested.
From Table 2, it is seen that the mean OD values for points on the standard curves vary even if the curves have been generated by the same operator. This is due to the variations in assay conditions during different runs. Though the variation may appear within a standard deviation of 0.1-0.2, it can amount to a 10 % error in accurate estimation of the concentration of an unknown sample, if the sample concentration is not interpolated from the correct curve, i.e. the standard curve generated on the same plate on which the sample is being tested. For example, if the mean OD@450 nm for point S6 reads 2.025 on day 1 and the OD@450 nm of an unknown sample reads 2.076 on day 1 and 1.761 on day 2, the two concentrations of the sample interpolated from the single curve will be different.
In conclusion, due to inter-assay and inter-operator variability, we strongly recommend that every time the concentration of an unknown sample(s) is to be estimated, a standard curve must be run on the same plate along with the sample(s). Otherwise, any results obtained could be inaccurate.
It is imperative to understand that the standard curve run at different times will not yield the exact OD values for every run. The reason for this variation includes inter-operator and assay such as the incubation conditions and pipettors used. Thus, the OD@450 nm value for a point on the standard curve on one given day may not be the same on the next day. To obtain accurate results in calculating concentrations of unknown samples, the interpolation of data must not be done using a standard curve from a previous run. A separate standard curve should be generated for every run along with unknown samples to yield reliable data.
These findings demonstrate inter-operator variability by comparing separate assays performed by two operators using the same lot of an ELISA kit. Additionally, to assess inter-assay variability, we compare the results of separate runs of ELISA by the same operator using the same lot from the ELISA kit.
Methods
Inter-Operator variability: Standard curves were prepared in duplicates by three different operators on three different dates following the kit protocol for product 42400-1 Verikine Mouse Interferon Beta ELISA Kit, lot 3700.
Inter-assay variability: To assess inter-assay variability three different standard curves were prepared in duplicates by the same operator on two different dates using Verikine Mouse Interferon Beta ELISA Kit lot 3765.
After the completion of the assay, the plates were read at OD 450 nm using Vmax Kinetic Plate Reader (Molecular Devices Corporation, CA, USA). The data points comprising each standard curve were fitted by 4-parameter fit using SoftMax Pro ver 5.0 (Molecular Devices Corporation, CA, USA).
Results
Inter-Operator Variability
Figure 1: Standard curves were prepared by three different operators on three different days. The curves were fitted by 4-parameter fit of the data points
Green Curve: Run on 6.14.07 by Operator 1
Blue Curve: Run on 7.6.07 by Operator 2
Red Curve: Run on 7.18.07 by Operator 3
Table 1: Mean OD values of the data points on the three standard curves prepared by three different operators
Curve Point |
Mean OD@450 nm |
|||
S7 S6 S5 S4 S3 S2 S1 |
Concentration (pg/ml)
|
Operator 1 (green curve) |
Operator 2 (blue curve) |
Operator 3 (red curve) |
1000 500 250 125 62.5 31.25 15.625 |
3.239 1.874 1.127 0.55 0.394 0.26 0.192 |
2.32 1.293 0.863 0.523 0.341 0.222 0.161 |
3.009 1.74 0.926 0.637 0.327 0.227 0.146 |
Figure 2: Inter-Assay Variability, Standard curves prepared by same operator on two different days.
Blue Curve: Run on 11.30.07 by Operator 2
Red Curve: Run on 11.30.07 by Operator 2 in parallel with run1 (blue curve) on a different plate
Green Curve: Run on 12.7.07 by Operator 2
Table 2: Mean OD values of the data points on the three standard curves prepared by the same operator.
Curve Point |
Mean OD@450 nm |
|||
S7 S6 S5 S4 S3 S2 S1 |
Concentration (pg/ml)
|
Run 1 (Blue curve) |
Run 2 (Red curve) |
Run 3 (Green curve) |
1000 500 250 125 62.5 31.25 15.625 |
3.498 2.035 1.353 0.746 0.512 0.341 0.278 |
3.74 2.331 1.494 0.915 0.609 0.455 0.335 |
3.561 1.955 1.378 0.836 0.567 0.403 0.331 |
Conclusion
From Table 1, it is evident that the mean OD values for the data points in the standard curve are different for curves generated by different operators. For example, the Mean OD values for point S7 are 3.239, 2.32 and 3.009 when the curves are generated by Operator 1, Operator 2 and Operator 3, respectively. As a result, the OD value for an unknown sample may also vary when tested by different operators. However, since the concentration of the sample is calculated using the standard curve, any operator-induced variation can be negated. This occurs if the concentration of the sample is interpolated from the standard curve prepared by the same operator on the same plate on which the sample is being tested.
From Table 2, it is seen that the mean OD values for points on the standard curves vary even if the curves have been generated by the same operator. This is due to the variations in assay conditions during different runs. Though the variation may appear within a standard deviation of 0.1-0.2, it can amount to a 10 % error in accurate estimation of the concentration of an unknown sample, if the sample concentration is not interpolated from the correct curve, i.e. the standard curve generated on the same plate on which the sample is being tested. For example, if the mean OD@450 nm for point S6 reads 2.025 on day 1 and the OD@450 nm of an unknown sample reads 2.076 on day 1 and 1.761 on day 2, the two concentrations of the sample interpolated from the single curve will be different.
In conclusion, due to inter-assay and inter-operator variability, we strongly recommend that every time the concentration of an unknown sample(s) is to be estimated, a standard curve must be run on the same plate along with the sample(s). Otherwise, any results obtained could be inaccurate.
