JPS6250645A - Chemical analysis - Google Patents

Abstract

PURPOSE:To make the exact analysis of a reaction speed by carrying out plural times of the operation to determined the regression line of an absorbancy curve by a method of least squares from plural sets of absorbancy data and to determined the regression line by excluding the absorbancy data exceeding the standard deviation therefrom and making the analytical measurement of the reaction speed from the finally remaining data. CONSTITUTION:The regression line for the absorbancy curve is determined by the method of least squares from the absorbancy data at the plural points of the time sent from a device body 1 by a control device 2. The data of which the absolute value of the difference in the absorbancy data at the various points of the time with respect to the standard deviation thereof exceeds the prescribed coefft. times of the standard deviation is removed from the data and the regression line is determined. Such operation is carried out plural times and the exact analytical measurement of the reaction speed is made in accordance with the finally remaining data. The result thereof is displayed on a display device 3.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application 1] The present invention relates to a chemical analysis method that enables accurate reaction rate measurement.

[Prior Art] Recently, chemical analysis methods have been frequently used to analyze biological fluids such as blood and urine. This analysis method involves reacting a test liquid with a reagent and measuring the extent of this reaction optically, for example. In addition to these chemical analysis methods, there is a reaction rate analysis method that measures changes in absorbance, etc. over a certain period of time while the test solution and reagent are reacting.
As5ay) (hereinafter referred to as RRA) is used, for example, when measuring the activity value of an enzyme.

FIG. 3 shows the time-absorbance characteristic line of test liquid A when an analysis item (for example, GOT) consisting of serum is to be analyzed by RRA measurement using a chemical analyzer. Now, to perform RRA measurement, it is easy to do the following:
The purpose is to find the slope of this time-absorbance characteristic line. That is, for example, the absorbance between tl and jn (OD+, respectively) from the start of the reaction.

The next ΔOD, which is the change in OD, ), is measured.

Δ00− (ODn −OD+ )/(tn −t+
) [Problems to be Solved by the Invention] Actually, it, jz, tz from the start of the reaction.

When absorbance at tn is measured, as shown in Figure 4, the measured absorbance value is not necessarily on the ideal linear time-absorbance characteristic line. They are distributed above and below the typical time-absorbance characteristic line. Such dispersion occurs due to the S/N ratio of the optical detector being 2 during the reaction due to mechanical vibration or positional shift of the reaction vessel, or insufficient S/N ratio of the reaction itself due to side reactions.

Therefore, even though there is such a dispersion, even when measuring ΔOD,
This does not mean that an accurate RRA measurement has been performed.

The present invention aims to solve such problems.

[Means for Solving the Problems] Therefore, the present invention measures absorbance at each of a plurality of time points after the start of the reaction, and calculates an absorbance characteristic line from the absorbance data at each measured time point using the least squares method. Find a regression equation to represent the data, use this regression equation to find the standard deviation of the data, and calculate the absolute value of the difference between the value calculated by the regression equation and this absorbance data among the absorbance data at each measured time point. However, data larger than the tolerance range set based on this standard deviation was excluded, and the process of calculating a new regression equation using the least squares method from the remaining absorbance data at each time point was repeated multiple times. RRA measurements were performed based on absorbance data.

[Example] FIG. 1 is a block diagram of a chemical analysis apparatus for carrying out the chemical analysis method of the present invention.

This chemical analyzer is a container containing a test liquid.

A sampling mechanism that sucks up the test liquid from this container and sends it to the reaction container, a reagent supply mechanism that sucks up a reagent suitable for the item to be measured from the reagent container and sends it to the reaction container, and reacts the absorbance of the test liquid in this reaction container. The device main body 1 consists of a reaction and detection mechanism for detection at multiple times after the start, controls the operation of this device main body, performs various calculations based on the absorbance value detected by this device main body, and performs R.
It is composed of a control device 2 that calculates RA measurement values, etc., and a display bag M3 that displays the measured RRA measurement values according to instructions from this control device.

First, after the start of the reaction, a plurality of time points (tl.

jz, tz, tn, ts, is, tz,...
....

The control device 2 gives a command to the device main body 1 so that the absorbance is detected at the points jn-+, tn). This command causes a plurality of time points (tn, tz) after the reaction start time of the test liquid sent from the main body 1 of the apparatus.

tz, 1:4. i:s, tg, i7+...
, tTl-I.

Based on the absorbance data of in), the control device 2 performs the following calculation.

Now, each time point (tl, tz, ts) after the reaction start time of the test liquid sent from the main body 1 of the apparatus.

t4. ts, tg, tz, ......, in
-+, tn) as shown in Figure 2, the absorbance of ○D+, OD2, OD3, OD4, ODs, respectively.
, ODg, 007.

......, 0Dn-+, ODn. Here, the control device 2 assumes that the absorbance characteristic line is a straight line, performs a calculation using the least squares method based on the N pieces of absorbance data measured between t1 and tT1, and expresses the characteristic line. Find the regression line 2〇〇=a1t+b.

In addition, al in this formula is the change value of absorbance per unit time,
b is the (V) intercept of this equation.

The solid line OD in FIG. 2 is a time absorbance characteristic line representing the equation of this regression line. Then, using this regression line formula,
Each time point (tn, tz, tz, tn.

ts, js, t7-..., jT, 4. tn
), and the absorbance at each time point measured above is OD1,002.003.004.00s, O
D6,○D7,・・・・・・・・・,○Dn-10D
Difference between n and each calculated value X1. X2, X3, X4, X
s, X6, X7, ......*xn-+
l X n, and these calculations - nx2L/ (
n-1)) (where Ma=Σx/N) is calculated.

Next, considering the allowable range for this standard deviation Sx1, a coefficient k
The absolute value of the product multiplied by I SX+ ・kl and the absolute value of the difference at each time point IX+ l (i = 1 to n)
and exclude absorbance data that exceeds the dispersion tolerance. Next, at each remaining time point, for example, (NQl)
Based on the absorbance data, a new regression line equation Q[)=82i+b is determined using the least squares method. Next, using this regression line equation, the absorbance at each of the remaining time points is calculated, and the standard deviation S x z is calculated from the difference between this calculated value and the absorbance at each of the remaining time points. next,
The absolute value 1sXz·kl of this standard deviation SXz multiplied by a coefficient k taking into account the tolerance range is compared with the absolute value of the difference at each time point, and absorbance exceeding the tolerance range is excluded. And, for example, (NQ
Based on the z) pieces of absorbance data, a new regression line 2〇D-ast+b is determined, and the above calculation is repeated. In this way, by repeating the above calculation a preset number of times, the value of a in the regression line equation, that is, the absorbance change per unit time, becomes the true ΔOO value, that is, the ideal absorbance characteristic line. The slope closely approximates that of . Therefore, the value of a in the equation of the regression line finally remaining, for example, determined based on the P absorbance data, is closest to the true Δ0Dfa. Note that the number of repetitions of the above-mentioned calculation is appropriately multiplied by the number of pieces of data to be measured first.

[Effects of the Invention] According to the present invention, the S/N ratio of the optical detector is 9. The S/N ratio during the reaction due to mechanical vibration or positional shift of the reaction vessel, or the insufficient S/N ratio of the reaction itself due to side reactions, etc. Even with the dispersion that occurs, accurate RRA measurements can be made.

[Brief explanation of the drawing]

Figure 1 is a block diagram of a chemical analyzer for carrying out the chemical analysis method of the present invention, Figure 2 is to supplement the detailed explanation of the present invention, and Figure 3 is an ideal absorbance characteristic line. FIG. 4 shows the distribution of absorbance values over actual time. 1: Apparatus body 2: Control device 3: Display device drawing 1:i・No change in general content) Film correction procedure amendment (method) Showa τ,, -, r1tr '1・i 4FtfiiM 1985 December 13th 1, Display of the case Patent Application No. 191268 of 1985 2, Name of the invention Chemical analysis method 3, Person making the amendment Relationship to the case Patent applicant address 1418 Nakagami-cho, Akishima-shi, Tokyo (T.E. L
0425 (43) 1165) 4. Date of amendment order November 26, 1985 5. Drawings subject to amendment 6. Contents of amendment As per the engraving of the drawing attached to the application dated August 30, 1985 (contents (no change). that's all

Claims (1)

[Claims] After the start of the reaction, absorbance is measured at multiple time points, and a regression equation representing the absorbance characteristic line is determined using the least squares method from the absorbance data at each measured time point. Using this regression equation, the standard of the data is determined. The deviation is determined, and among the absorbance data at each measured time point, the absolute value of the difference between the value calculated by the regression formula and this absorbance data is larger than the tolerance range set based on this standard deviation. A chemical analysis in which the process of calculating a new regression equation using the least squares method from the remaining absorbance data at each time point is repeated multiple times, and finally reaction rate analysis is performed based on the remaining absorbance data. Method.