DE102004023519A1 - Equipment for automatic measurement of chemicals based on immunochemical methods, includes pipetting apparatus with reader and vibrator under randomizing software control - Google Patents

Abstract

The equipment includes automatic analysis equipment. It has pipetting functions, a washer, reader and vibrator. These are under dedicated software control. The equipment is combined with a powerful temperature control system for heating and cooling of the entire interior. Individual cavities of the microtitration plate are controlled randomly in time and space. The robot serves not only for parallel execution of all operating stages, but is also used in accordance with known principles of randomization from statistical investigation planning. The idea is that errors arising through averaging at least partially cancel, because the individual values are determined under different conditions. In contrast, with parallel measurement, there is a great risk that all measurement values, although very similar to each other, all have the same error. Randomization is realized spatially on the microtitration plate and also temporally through a stratified randomization sequence of the individual cavities, using specially-developed software control. The equipment is further detailed in accordance with the foregoing principles. Further known statistical techniques are invoked.

Description

On The market is a series of analysis robots for fully automated Use of ELISA tests available. In many cases, these robots contain at least one pipetting or Gripper arm, a washing station, a shaker, a reader, a control and evaluation and partially electronically tempered incubators (Heater). Suppliers of such systems are e.g. TECAN (e.g., Freedom EVO) and IBL (device: Triturus). The field of application of these analysis robots is mainly in the Field of human and veterinary medicine, the focus is usually on a high sample throughput, i.e. it is desirable, as possible many samples daily analyze. For These systems are valid as far as possible measuring errors or disturbances in the measurements should be avoided by making the measurements as possible be carried out simultaneously under identical conditions. This purpose is particularly the use of multiple pipettes, which allow parallel analysis of multiple samples. The disadvantage of this However, the procedure is that the measured values, although very similar to each other, all have the same error. This disadvantage leads in particular to that also by repeating measurements and averaging individual values only slight improvements in measurement accuracy can be achieved.

Farther use all automated measuring systems, as far as they have an evaluation unit usually, the so-called 4-parameter equation, in order the measuring signal of the reader (microtiter plate photometer), the so-called. Convert OD value (OD = optical density) into a concentration value. This generated 4-parameter curve considered neither the different position of the measured cavity on the Microtiter plate, still the time of processing or deviations the calibration curve from the 4-parameter curve. The aspects mentioned can However, especially at low concentrations a very strong influence have on the measurement result and should therefore be considered accordingly become. After all contain the measuring systems mentioned - if at all - no generally accepted System for determining the detection and determination limit. Manages Therefore, at the detection limit with the use of the 3 Sigma range. The measuring range therefore begins only above that by the threefold Standard deviation of the blank value signal calculated detection limit.

at all measuring instruments becomes parallelism aspired to measurement and calibration. It is therefore desired that the errors that occur are due to their parallelism Occurrence as possible cancel each other out.

detailed description

outgoing from this prior art, the present invention is the Task to eliminate measurement errors to a significant to obtain a more sensitive measuring method, which - depending on the specification the measurement parameter - one Improvement of sensitivity by a factor of 10 or more possible. A new analysis robot is used to implement this measuring method for ELISA tests, a series of innovations. This analysis robot is with a Pipetting function, furthermore with a washer for the pipetting needle, a reader with integrated shaker, a powerful temperature control unit together with associated Software control.

A technical Realization of the analysis robot

at the pipetting needle is a simple double needle, which serves both for pipetting the reagents and samples, as also for washing the individual cavities by pipetting the washing liquid and simultaneous suction. The pipetting needle is on a simple Robotic arm attached and can over Software commands are moved arbitrarily. The tempering unit exists from a Peltier element with fan, by air temperature control both heating and cooling the entire interior allows. Temperature sensors are used for monitoring the temperature. So far existing devices allow at best a warming of the entire interior, but not its cooling, but what a measurement is very important by ELISA near the detection limit.

The casing is on all sides with insulating material lined. The pipetting is carried out by means of a pipetting robot on the microtiter plate, which is on the transport carriage of the reader engaged rests. The reader should have a shaking function be equipped so waived a separate shaker and associated gripper arm can be.

B Application of the principle the temporal and spatial Randomization of the individual measurements

The Software control of the analysis robot is designed so that the individual cavities the microtiter plate randomly and spatially controlled become. The up to 15 steps (pipetting and aspiration various substances with integrated washing and shaking steps) be over controlled a single work cycle. In every work cycle becomes a single cavity the microtiter plate treated so that each step in total 96 bars (for the 96 cavities) required are. Every cavity So is individually by means of a double needle (pipetting, aspirating and washing). The successive steps be over a database controlled. about the software control is logged every step and the associated ones Environmental conditions (temperature etc.) registered. By the clock control is guaranteed that the work steps for every cavity be carried out in the same chronological order. This approach has opposite the usual parallel processing the big one Advantage that the always effective measuring deviations per measured value in other means are effective, so that by averaging a large part these deviations can be canceled. This also means that for needed the calibration Standard samples randomized on the microtiter plate can be arranged. This prevents the usual systematic error.

To to 25 repetitions of individual samples are provided. This allows at the optical density to improve the sensitivity the factor 5, and the concentration - depending on the variation of the individual values - an improvement by a factor of 20.

C Automatic correction the measurement error resulting from different conditions

Furthermore, the analysis robot includes an evaluation module which enables the measurement errors resulting from different conditions to be corrected. In addition to the optical density, the evaluation module takes into account the respective column and column and row of the cavity and their ordinal number (in chronological sequence) during calibration. Possibly. The temperature can also be taken into account. The underlying calibration function then takes the following form

In this case, OD denotes the optical density, A, B, C, D are the parameters of the 4-parameter equation, c denotes the concentration, column and line effect are the spatial effects assumed to be linear, trend denotes the temporal effect (which is assumed to be linear) the ordinal number can be described), and the temperature effect is also a linear term that is proportional to the particular temperature. The unknown parameters are determined by means of the known maximum likelihood method. The determination of the parameters of the disturbing effects (column effect, line effect, trend, temperature effect) can be carried out automatically and specifically for each microtiter plate. Alternatively, a cross-plate definition and then entry in a database is possible. This has the advantage that the number of measurements required for the calibration can be reduced. It should be noted that in previous systems an automatic error correction of the measurement deviations is not possible in principle, since the measurements are processed in parallel and the calibration function always has only the simple 4-parametric shape, ie

D calibration by means of a generalized 5-parameter function

In addition to the optical density, the evaluation module takes into account the respective column and column and row of the cavity and their ordinal number (in chronological sequence) during calibration. Possibly. The temperature can also be taken into account. The underlying calibration function then takes the following form

The Determination of the unknown parameters takes place by means of the known Maximum likelihood. The parameter F is usually fixed, e.g. F = exp (In (D) - 2). The 5-parameter function has opposite the otherwise used 4-parameter function the big one Advantage, that especially in the vicinity of the detection limit observed Deviations of the calibration curve can be better modeled after "down".

E determination the detection limit

The Evaluation module allows continue to record the detection limit based on the likelihood quotient test. For every the sample to be examined is the calculation of the likelihood function based on the specified calibration function including the standards and the sample to be examined, the concentration this sample is assumed to be unknown. Continue to take place the same calculation assuming that the concentration of said sample is at zero. The difference between the two likelihood values serves as the basis of the likelihood ratio test, in which normal distribution the optical densities is assumed. He allows a decision whether the concentration of the unknown sample is significantly above 0 If the measured optical density of the sample is replaced by a theoretical value, receives by equating the difference between the two likelihood values with the critical value and reshaping of the expression the detection limit, where this detection limit of the number of repeated measures dependent is.

F The evaluation module in a flexible data entry and evaluation device

The in C-E described functions of the evaluation module can be in one flexible data entry and evaluation device, which e.g. can be connected directly to the reader, summarize. Thus, these functions are also for manually pipetted microtiter plates realizable. However, an accurate temperature control is so not possible anymore, as well as the principle of temporal and spatial Randomization: To avoid pipetting errors and confusion, In practice, it is always ensured that all pipetting be processed in a systematic order.

G Application validation

Of the Analysis robot also includes a special module, which a automatic execution of a Application validation according to EC 2002/657. For the Application validation according to EC 2002/657 requires eight microtiter plates, those with unloaded samples of different origin after one factorial plan become. For each single plate is a random pipetting pattern is used, wherein different surrounding conditions are varied (in particular Temperature). The software control therefore includes the import option for a external XML file, as well as the ability to read the results in a given XML format. This XML file then becomes external by a specialist for Evaluated application validation, which includes the XML file with the randomized Specifies pipetting pattern. To evaluate the XML file with the results, a separate software is required.

H The ELISA test

• – Messung der optischen Dichte bei einer anderen Wellenlänge
• – Verwendung eines anderen chromogenen Substrates. Außer TMB können auch eine Vielzahl anderer Substrate eingesetzt werden, die teils durch Peroxidase, teils durch andere Enzyme (siehe Punkt 3), umgesetzt werden.
• – Verwendung anderer Substrate, z.B. fluorogener Substrate etc.
• – Das substratumsetzende Enzym, Meerrettichperoxidase, kann ersetzt werden durch andere Enzyme, z.B. alkalische Phosphatase, Glucosidase etc.
• – Die Antikörper können durch andere E2 bzw. EE2-bindende Antikörper ersetzt werden, die nicht aus Kaninchen stammen müssen und auch monoklonale Antikörper sein können. Auch Antikörper gegen andere Analyten sind einsetzbar.
• – Das Testformat kann in der Weise geändert werden, dass es sich nicht mehr um einen direkten Immunoassay handelt, sondern beispielsweise um einen indirekten kompetitiven ELISA.

With regard to the ELISA used, the analysis robot is based on the technique of the so-called direct, competitive, heterogeneous enzyme immunoassay. Two antibodies are used which are selective for 17β-estradiol or the synthetically produced 17α-ethinylestradiol. The signal is generated in an enzyme-catalyzed reaction. The enzyme used is the horseradish peroxidase, which together with the substrates hydrogen peroxide and 3,3 ', 5,5'-tetramethylbenzidine (TMB) gives a blue dye which, after a suitable reduction in the pH, turns to yellow and its light absorption ( so-called "optical density") at 450 nm (minus its absorption at 650 nm) represents the measurement signal. This determines the amount of a so-called "tracer" (a coupling of the analyte molecule possibly via a bridge to an enzyme molecule), which is indirectly related to the analyte concentration to be determined via the competitive situation with the analyte in the standards or in the sample. The two antibodies are selective for once the natural steroid hormone 17β-estradiol (estradiol, estradiol, E2) or the synthetically produced 17α-ethinylestradiol (ethynylestradiol, ethinylestradiol, ethinyloestradiol, EE2). The antibodies were obtained by immunization of rabbits with the aid of a suitable immunogenic analyte-protein conjugate and subsequent blood sampling. The enzyme-analyte conjugate ("tracer") and the carrier protein-analyte conjugate ("immunogen") were prepared according to the same synthesis principle. The principle of the ESTR-A-LISER to reproducibly and sensitively detect an analyte by means of an immunoassay can be modified in a variety of ways without having to change the fundamental structure of the system:

• - Measurement of the optical density at a different wavelength
• - Use of another chromogenic substrate. In addition to TMB, a variety of other substrates can be used, which are partly by peroxidase, partly by other enzymes (see point 3), implemented.
• Use of other substrates, eg fluorogenic substrates etc.
• The substrate-converting enzyme, horseradish peroxidase, can be replaced by other enzymes, eg alkaline phosphatase, glucosidase, etc.
• - The antibodies can be replaced by other E2 or EE2-binding antibodies that are not derived from rabbits and may also be monoclonal antibodies. Antibodies to other analytes can also be used.
• - The test format can be changed in such a way that it is no longer a direct immunoassay but, for example, an indirect competitive ELISA.

All Incubation and other waiting times are changeable and must ever changed according to requirements or optimized.

Claims (9)

An analysis robot, with pipetting function, Washer, Reader and shaker together with associated Software control in combination with a powerful temperature control unit, with both heating and cooling of the entire interior allows becomes. An analysis robot, consisting of a pipetting function, Washers, readers and shakers together with associated software control and temperature control, which is designed so that the individual wells the microtiter plate randomly and spatially controlled become. The robot is therefore not used for parallel processing all step, but uses that from the statistical Experimental design known principle of randomization. The idea here is that the errors that occur are averaging at least partially cancel each other out because the individual values are below different conditions are determined. In contrast, at parallel measurement the risk big that all readings, though very similar to each other, all have the same error. The said randomization becomes both spatially - on the Microtiter plate - as also in time - by a stratified randomized order of the individual cavities by means of a realized specially developed software control. An analysis robot, consisting of a pipetting function, Washers, readers and shakers together with associated software control and temperature control, which includes an evaluation module, which allows that the measurement errors resulting from different conditions can be corrected. This concerns different positions on a plate, temperature fluctuations, Measuring volumes, pipetting errors etc. by averaging the random deviations, as well as systematic corrections of temporal and spatial Effects and temperature fluctuations. An analysis robot, consisting of a pipetting function, Washers, readers and shakers together with associated software control and temperature control, which includes an evaluation module, which the inclusion of curves allows those with the ordinary ones 4-parameter function can not be detected. That's the purpose a generalized 5-parameter function possible. An analysis robot, consisting of a pipetting function, Washers, readers and shakers together with associated software control and temperature control, which includes in its evaluation module, which the detection of the detection limit based on the likelihood quotient test allows. A computer module, which is a direct acquisition the pipetting and calibration data and temperature data from the control unit and the result data from the reader, and which following functions for improved concentration determination allows: a. automatic Error correction within the calibration step (3) b. use the 5-parameter curve (4) c. Calculation of the detection limit by means of Likelihood quotient test (5) An analysis robot, consisting of a pipetting function, Washers, readers and shakers together with associated software control and temperature control, which in its software control is a special module which automatically performs an application validation according to EC 2002/657 allows. An analysis robot, consisting of pipetting function, washer, reader and shaker, together with associated software control, temperature control and control and evaluation modules, which include the following functions: a. Stratified Randomization (2) b. Automatic error correction within the calibration step (correction of random and systematic errors by inclusion of time and location point as well as temperature (3)) c. Using the 5-parameter curve (4) d. Calculation of the detection limit by means of the likelihood quotient test (5) e. Control unit enabling automated validation according to EC 2002/657 (6). An analysis robot, consisting of a pipetting function, Washers, readers and shakers together with associated software control and temperature control, which in its measuring principle on the technology the so-called direct, competitive, heterogeneous enzyme immunoassay based, with two antibodies to be used for 17β-estradiol or the synthetically produced 17α-ethinylestradiol are selective.