Optics: measuring and testing – By dispersed light spectroscopy – With sample excitation
Reexamination Certificate
2002-02-27
2004-01-06
Smith, Zandra V. (Department: 2877)
Optics: measuring and testing
By dispersed light spectroscopy
With sample excitation
C356S307000, C250S459100, C250S461200
Reexamination Certificate
active
06674527
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This application relates generally to fluorescence polarization measurements and, in particular, to a method of correcting fluorescence polarization measurements for the effects of background noise.
2. Description of the Related Art
Fluorescence polarization (FP) measurements are used to detect molecular interaction, since molecular interaction typically leads to an increase in the polarization signal. See John C. Owicki, “Fluorescence Polarization and Anisotrophy in High Throughput Screening: Perspectives and Primer”, JOURNAL OF BIOMOLECULAR SCREENING, Vol. 5, No. 5, 2000 (Hereinafter referred to as “
FP/FA Primer
”, and incorporated by reference) for various examples of assays using fluorescence polarization.
FIG. 1
shows a conventional FP Plate Reader which is used for taking FP measurements from samples in a microtitre plate. Plate
101
contains Samples
105
in Wells
110
. Samples
105
are typically comprised of a cell preparation, buffers, a drug or other target substance being tested, and fluorophores that will fluoresce when excited by light. This excitation light is emitted by Lamp
120
, filtered by Filter
125
, and polarized by Excitation Polarizer
130
, before reflecting off of Dichroic Beamsplitter
140
through Objective
145
onto Samples
105
. The light emitted by the fluorophores in Samples
105
will proceed through Objective
145
and Dichroic Beamsplitter
140
, before being filtered by Filter
150
and analyzed by Emission Polarizer
160
. The analyzed emitted light is then detected by Detector
170
.
The conventional equation for calculating FP takes the difference between the measured signals when the polarizers are parallel (i.e., when Excitation Polarizer
130
lets light through with the same polarization as light let through Emission Polarizer
160
) and when the polarizers are perpendicular (i.e., when Excitation Polarizer
130
lets light through with a polarization orthogonal to the light let through Emission Polarizer
160
), and divides this C difference by the sum of the two measurements, as shown by:
F
P
=
1000
*
I
∥
-
I
⊥
I
∥
+
I
⊥
(
Equation
1
)
where I
∥
is the intensity measurement with the polarizers parallel, and I
⊥
is the intensity measurement with the polarizers perpendicular. The FP signal is a unitless number that indicates degree of fluorescence polarization (DOP), and is typically called millipolarization, “milliP” or “mP”.
It is the difference (I
∥
-I
⊥
) that carries the assay information, the remaining terms in the equation are for normalization. The object of detection of FP measurements, successful molecular binding events, are indicated by typically small variations in the difference (I
∥
-I
⊥
). Because of the need for detecting these small variations, the utility of instruments designed for FP measurement is primarily dependent on the precision of the instrument, and secondarily on the accuracy. In other words, it is more important that the intensity measurements I
∥
and I
⊥
are precise relative to each other rather than they be individually absolutely accurate.
When making conventional FP measurements using conventional FP measurement devices, one must often choose between accuracy and precision. As an example, consider the problem of background fluorescence in a conventional FP Plate Reader. Background fluorescence is caused by non-probe or non-subject (the subject being the molecule under study) elements, such as Wells
110
or the cell preparation and buffers in Samples
105
, which fluoresce and add to the emitted light which reaches Detector
170
. This can affect the accuracy of the FP measurements. The accuracy problem can be illustrated by the following hypothetical situation, where a “pure” FP measurement (meaning “purely” from the fluorophores or probes) is sought:
F
P
probe
=
1000
*
I
∥
probe
-
I
⊥
probe
I
∥
probe
+
I
⊥
probe
(
Equation
2
)
where
I
∥
probe
=the intensity signal of parallel polarized light from the probes; and
I
⊥
probe
=the intensity signal of perpendicularly polarized light from the probes.
In our hypothetical, we will assume that I
∥
probe
=60,000 and I
⊥
probe
=40,000, so the actual FP of the probes, FP
probe
,=200 mP. Furthermore, we will assume that the intensity signal of the background is equal to some proportion of the total intensity signal of the probes:
I
bkgrnd
=I
bkgrnd::probe
*I
probe
where
I
bkgrnd
=total intensity signal of the background;
I
bkgrnd::probe
=the ratio of background signal to probe signal; and
I
probe
=total intensity signal of the probes.
A reasonable assumed value for I
bkgrnd::probe
is ⅓ or 0.33, which will result in I
bkgrnd
=33,000. Now, we will determine the parallel and perpendicularly polarized components of the background signal:
I
∥
bkgrnd
=
I
bkgrnd
*
(
1
+
FP
bkgrnd
)
2000
I
⊥
bkgrnd
=I
bkgrnd
−I
∥
bkgrnd
where
I
∥
bkgrnd
=the intensity of the parallel polarized component of the background;
FP
bkgrnd
=the fluorescence polarization of the background; and
I
⊥
bkgrnd
=the intensity of the perpendicularly polarized component of the background.
Assuming a value of 450 mP for the background FP (FP
bkgrnd
), the parallel polarized component of the background, I
∥
bkgrnd
, equals 23,930 and the perpendicularly polarized component, I
⊥
bkgrnd
, equals 9075. The total intensity of the polarized signal received by Detector
170
equals the combination of the probe signal and the background signal:
I
⊥
meas
=I
⊥
bkgrnd
+I
⊥
probe
I
⊥
meas
=83,930
I
∥
meas
=I
∥
bkgrnd
+I
∥
probe
I
∥
meas
=9075
In the end, the final measured FP as calculated by the intensity measurements at the detector is:
F
P
meas
=
1000
*
I
∥
meas
-
I
⊥
meas
I
∥
meas
+
I
⊥
meas
(
Equation
3
)
FP
meas
=262
Thus, the measured FP of 262 is more than 25% in error (from the real FP of 200). If this type of error margin is unacceptable to the experimenter, she may use a method to compensate for the signal noise generated by the background. Such methods have been developed over time, as discussed in FP/FA Primer. These conventional methods to compensate for background fluorescence use a subtractive approach to decrease the background signal source. For example, Equation 4 below directly subtracts parallel and perpendicular measurements of background wells (containing one or more assay components, but not the fluorophore) from the parallel and perpendicular measurements of the sample wells under study (all assay materials including fluorophore):
F
P
meas
=
1000
*
(
I
∥
meas
-
I
∥
bkgrnd
)
-
(
I
⊥
meas
-
I
⊥
bkgrnd
)
(
I
∥
meas
-
I
∥
bkgrnd
)
+
(
I
⊥
meas
-
I
⊥
bkgrnd
)
(
Equation
4
)
The disadvantage of this approach is that the accuracy of background correction is subject to changes in excitation lamp intensity, well-to-well variations in sampled volume and meniscus, and the degree of photobleaching between the background measurement and the sample measurement. In effect, the background correction step injects into the calculation additional noise, which reduces measurement precision. Thus, in order to achieve greater accuracy in one's FP measurements, one may end up sacrificing the precision of such measurements. But, as is discussed above, this is not desirable, since the precision of these measurements relative to each other is more important than the accuracy of the individual measurements.
Some experimenters may prefer losing the accuracy of th
Cambridge Research & Instrumentation Inc.
Cohen & Pontani, Lieberman & Pavane
Geisel Kara
Smith Zandra V.
LandOfFree
Ratiometric background correction for fluorescence... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Ratiometric background correction for fluorescence..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Ratiometric background correction for fluorescence... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3252713