JP2005337920A - Light signal analyzing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform signal analysis with respect to a plurality of measuring points in a specimen in an almost real time by simple procedure. <P>SOLUTION: In the light signal analyzing method, a sample is irradiated with light and the light emitted from the sample is guided to a photodetector while the time series signal outputted from the photodetector is divided into a plurality of time series signals with time and signal analysis is performed at every time series signal divided with time to calculate the physical properties of the sample. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an optical signal analysis method.

Conventionally, there are the following documents as optical signal analysis methods. In these optical signal analysis methods, when estimating the autocorrelation function, a time-series signal of fluorescence intensity is acquired at only one measurement point, and self-calculation is performed using all the acquired time-series signals. The correlation function was estimated.
`` Fluorescence Correlation Spectroscopy of Triplet States in Solution: A Theoretical an Expermental Study '', J. Windengren et al., J. Phys. Chem 99, 13368-13379 (1995) "Single molecule detection by fluorescence correlation spectroscopy", Masataka Kaneshiro, Protein Nucleic Acid Enzyme, 1431-1438, Vol.44 No.9 (1999). "Single molecule level enzyme reaction analysis by fluorescence correlation spectroscopy", Nishimura, Y., Biophysics, 81-85, Vol.39 No. 2 (1999).

  When attempting to estimate the autocorrelation function at a certain time interval at multiple measurement points in the sample using the conventional method described above, first estimate the autocorrelation function at one measurement point, and then estimate the other correlation points. To estimate the autocorrelation function in the same way. Furthermore, the operation of moving to another measurement point and estimating the autocorrelation function in the same manner has to be performed sequentially for all measurement points, so that the entire estimation takes time and the work is complicated. In addition, since the measurement points are sequentially moved, the time difference between the measurement points is long and requires a long time, and it is substantially difficult to estimate the autocorrelation function for a plurality of points at the same time.

  That is, a measurement time shift (time delay) occurs at each measurement point, and fast reactions occurring inside and outside the cell, such as protein-protein interaction, intracellular signal transmission, etc., could not be measured in real time.

  The present invention has been made paying attention to such problems, and the purpose of the present invention is to perform optical signal analysis capable of performing signal analysis on a plurality of measurement points in a sample in a substantially real-time with a simple procedure. It is to provide a method.

  In order to achieve the above object, a first invention is an optical signal analysis method, comprising: a detection step of detecting light emitted from a measurement point designated with respect to a sample by a photodetector; and A time division step of time-dividing an output time-series signal into a plurality of time-series signals; and an analysis step of performing signal analysis for each of the time-division time-series signals and obtaining physical properties of the measurement points. It has.

  According to a second invention, in the optical signal analysis method according to the first invention, a plurality of points relating to the sample are sequentially selected as measurement points, and the position irradiated with the light is sequentially moved with respect to the plurality of measurement points. The time-series signal output from the photodetector is time-divided for each of the plurality of measurement points, and time-series signal analysis is performed for each of the time-divided data, Find the nature.

  According to a third aspect of the present invention, in the optical signal analysis method according to the second aspect of the invention, the time-series signal output from the photodetector is time-divided and stored for each of the plurality of measurement points. A time-series signal for each of the plurality of measurement points is read and analyzed, and physical properties for each of the plurality of measurement points regarding the sample are obtained.

  According to a fourth aspect of the present invention, in the optical signal analysis method according to the third aspect of the present invention, the position information table is created by acquiring position information of the plurality of measurement points, and the time series is referred to with reference to the position information table. The signal is analyzed to determine the physical properties of each of the plurality of measurement points related to the sample.

  The fifth invention is the optical signal analysis method according to any one of the first to fourth inventions, wherein the light emitted from the measurement point is fluorescence, phosphorescence, chemiluminescence, bioluminescence, reflected light or Scattered light.

  The sixth invention is the optical signal analysis method according to any one of the first to fifth inventions, wherein the analysis of the time-series signal output from the photodetector is an autocorrelation spectroscopy analysis or a light intensity distribution. It is analysis.

  The seventh invention is the optical signal analysis method according to any one of the first to sixth inventions, wherein the physical property of the measurement point related to the sample is a translational diffusion coefficient, a rotational diffusion coefficient of a molecule or particle, Alternatively, it is the number of molecules or particles present in the measurement region, or the fluorescence lifetime.

  According to the present invention, it is possible to simultaneously analyze the physical properties of a plurality of measurement points, freely combine the measurement order of the plurality of points, and analyze the reaction of the sample, the interaction between the samples, and the like. In particular, in the case of a multipoint FCS, a plurality of point points can be observed simultaneously as compared with a single point FCS, and various reactions inside and outside cells of a living body can be captured.

  In addition, by adopting multi-point FCS of table search method, it avoids repeated scanning of all data for channel position determination by general search method, and acquires all data at each measurement point by only one scan. be able to. By simply referring to this channel position information table at each stage of autocorrelation calculation, the data of channels other than the calculation target is skipped, and only the minimum necessary data is taken into the calculation, reducing the calculation time and making multiple measurements. Points can be measured and observed in real time. In particular, as the number of measurement points (number of channels) increases, the amount of data skipped increases, and the calculation time can be significantly shortened.

  First, the outline of the present invention will be described. In the present invention, in multi-point FCS data analysis, data obtained by repeatedly measuring a plurality of points is time-series channel divided so that only data corresponding to each point is extracted and an autocorrelation function is calculated. Here, the extraction method is a general-purpose search method that sequentially scans all data.

  In multipoint FCS data analysis, in particular, the general-purpose search method must always scan all data repeatedly to determine the channel position when calculating the correlation function. In the calculation of the correlation function, it takes a considerable time to determine the channel position, that is, the position of the measurement point. In particular, it takes a considerable amount of time to move to a channel different from the channel to be calculated and to acquire data. As an improvement method, a data table (position information table) containing the position information of all channels at all measurement points is created using all the data obtained in a single measurement, and this is referenced for each correlation function calculation. Thus, the calculation time can be greatly reduced.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a diagram showing a configuration of an optical signal analyzing apparatus to which the present invention is applied. In the present embodiment, a case where the present invention is applied to a laser scanning confocal optical microscope in order to observe a fluorescent image of a sample will be described. The basic configuration of the apparatus includes a laser scanning confocal optical microscope main body 10, a CCD camera 11, a sample stage 12, a sample stage controller 13, a computer (PC) 14, a TV monitor 15, and a laser light source 16. Depending on the wavelength of the laser light source 16, the type of fluorescent substance to be excited differs. In this embodiment, an argon laser having a wavelength of 488 nm is used as one of the excitation wavelengths, but it has been found that it is suitable for exciting the fluorescent substance rhodamine green.

  The laser light emitted from the laser light source 16 is guided to the laser scanning confocal optical microscope main body 10 and irradiated to the sample 50 on the sample stage 12, and the fluorescent light from the sample 50 is received by the CCD camera 11. Alternatively, the sample 50 is observed with a fluorescence microscope through the microscope eyepiece.

  Two stepping motors 20 are respectively attached to the sample stage 12 of the laser scanning confocal optical microscope body 10 in a 90 ° direction, and each stepping motor 20 is connected to the sample stage controller 13. In response to a command from the computer 14, the two stepping motors 20 are driven to move the sample stage 12 in the X direction and the Y direction.

  Next, the configuration of the scanning system will be described. The laser light emitted from the laser light source 16 is passed through an expansion optical system (beam expander) composed of two lenses 23-1 and 23-2 to expand the beam diameter to be collimated light. To the dichroic mirror 25.

  The dichroic mirror 25 is formed so as to reflect light having a wavelength of about 488 nm or less and transmit light having a wavelength of about 488 nm or more. The laser beam reflected by the dichroic mirror 25 enters the XY scanner 22. The laser light is two-dimensionally scanned in the X-axis and Y-axis directions by the XY scanner 22 and guided to the objective lens 21. The laser beam is scanned in the sample 50 at random, and the fluorescent substance existing in the sample 50 can be excited.

  In order to form a minute confocal region in the sample 50, the objective lens 21 having an NA (numerical aperture) of 1.0 is used here. The size of the confocal region thus obtained is a substantially cylindrical shape with a central portion with a diameter of about 0.6 μm and a length of about 2 μm, and the average number of fluorescent molecules present in the measurement region is about several. Fluorescence emitted from a desired fluorescent substance in the sample 50 passes through the same optical path as the excitation light, is separated from the incident optical path by the dichroic mirror 25, and then received by the photodetector 28 via the pinhole 26 and the half mirror 27. Is done. The photodetector 28 may be a weak photodetector such as an avalanche photodiode (APD) or a photomultiplier tube. The pinhole has a diameter of 200 μm.

  A fluorescent output signal received by the photodetector 28 is guided to an image processing device (not shown), and after image processing such as contrast enhancement and contour enhancement, a two-dimensional image of the sample 50 is displayed on the TV monitor 15 through the computer 14. Alternatively, a three-dimensional image is obtained.

  The XY scanners 22 are arranged in vertical directions so that they can be scanned in directions orthogonal to each other, and the rotational movement is accurately controlled by respective XY scanner driving devices (not shown) controlled by the computer 14. The

  The XY scanner driving device is equipped with an XY scanner angle detection device (not shown), and the scanning angle is accurately detected in real time and feedback controlled by the computer 14. In this embodiment, a galvano scanner is used as the XY scanner 22. The means for XY scanning with laser light, that is, the XY scanner is not limited to a galvano scanner, and an acousto-optic modulation element (AOM), a polygon mirror, a hologram scanner, or the like may be used.

  The light intensity signal incident on the photodetector 28 is converted into an electrical signal, shaped in a waveform by a signal processing device (not shown), converted into an on-off binary pulse, and guided to the computer 14. The computer 14 performs a correlation operation on the input binary pulse signal to calculate an autocorrelation function. Furthermore, from the obtained autocorrelation function, the translational diffusion rate of the fluorescent substance, the change in the number of fluorescent molecules present in the measurement region, and the like are obtained.

  When the light intensity signal incident on the photodetector 28 is relatively large, the electrical signal output from the photodetector 28 is an analog time-series signal. In this case, A / D conversion is performed by the signal processing device, waveform shaping is performed, and the signal is converted into a binary pulse signal as before.

  Further, by moving the objective lens 21 in the optical axis direction, the focus position of the laser beam can be moved up and down along the optical axis direction. Thereby, a three-dimensional image can be generated on the TV monitor 15.

  Note that the fluorescent substance is not limited to rhodamine / green, but using, for example, TMR (Tetrame-hylrhodamine), Cyfive (Cy5), FITC (Fluorescein-isothiocyanate), TOTO1, Acridine-Orange, Texas-Red, etc. Also good.

(First embodiment)
FIG. 2 is a flowchart for explaining the procedure of multipoint FCS calculation by the general-purpose search method. First, the measurement data about the sample 50 is acquired by the configuration shown in FIG. 1 (step S1). Thereby, a time series signal as shown in FIG. Next, this time-series signal is time-divided into multi-channels (here, channels ch0 to ch4) (step S2, FIG. 3B). In FIG. 3B, ch0 is a random measurement data range during movement of the galvano scanner as means for XY scanning the laser light. ch1 is the measurement data range of measurement point 1. ch2 is the measurement data range of measurement point 2. ch3 is the measurement data range of measurement point 3. ch4 is the measurement data range of the measurement point 4. FIG. 3C shows an example of each measurement point 1 to 4.

Next, calculation is started by substituting 1 as an initial value for channel number CH_NO (steps S3 and S4). Next, as the first part of the FCS calculation, statistical calculation of the number of effective data Ma for average value calculation by the general-purpose search method is performed using the following equation (2-1) (step S5).

Here, all the measurement data are simply scanned, and in channel 1, +1 statistical calculation is performed only when the channel data is 1. The effective data used for the actual calculation for channel 1 is shown in FIG.

Only the part corresponding to. In order to extract this portion, all the measured data M _max (see FIG. 4) are sequentially scanned to determine the channel number. That is, in the channel 1, while sequentially scanning all data M _max, each time a channel number is 1, statistics calculated by +1 the value of M _a, if it is determined that the other channel numbers by scanning , M _a is not statistically calculated.

Similar processing is also in the channels 2, 3 and 4 is performed, and consequently, 4 statistical calculations channels multipoint FCS generic search method Mean value calculation valid data number M _a of one of M _tau (correlation calculation Scan all measured data M _max for a total of 4 times. That is, for p (total number of delay times during correlation calculation) _Mτ , M _max data is scanned p × 4 times to obtain a result.

Next, as a second part of the FCS calculation, the average value is calculated by the general search method using the following equation (2-2) (step S6).

For channel 1, the valid data used in the actual calculation is shown in FIG.

Only the part corresponding to. In order to extract this part, all the data M _max (see FIG. 4) measured in the same manner are sequentially scanned to determine the channel number. In other words, while sequentially scanning all data M _max, each time a channel number is 1, the value of M _a total calculated, if it is determined that the other channel numbers by the scan, the total calculation of M _a is performed Absent. The same processing is performed for channels 2, 3, and 4. As a result, the average value of the 4-channel multipoint FCS general-purpose search method is calculated for all measured data M _max for one M _τ . Scan a total of 4 times each. That is, for p _Mτ , M _max data is scanned p × 4 times to obtain a result.

Next, a third portion of the FCS calculation, the statistical calculation of the autocorrelation function calculation valid data number M _c, performed using the following equation (2-3) (step S7). This part is a part that statistics the number of data necessary for calculating the autocorrelation function.

For channel 1, the valid data used in the actual calculation is shown in FIG.

Only the part corresponding to. Autocorrelation calculation is

The results obtained by multiplying each piece of data by M _τ are summed. The value of M _c is the number of data used for autocorrelation calculation. In M _c, the original data D [i] used in the multiplication, M _tau shifted by data D _[i + M τ] must be the data of the same channel.

However, in order to extract this portion, first, all the measured data M _max (see FIG. 4) are sequentially scanned to determine the channel number. That is, in channel 1, while sequentially scanning all data M _max , every time the channel number of D [i] and D [i + M _τ ] becomes 1, the value of M _c is calculated by +1, and scanning is performed. If it is determined that either D [i] or D [i + M _τ ] is another channel number, the statistical calculation of M _c is not performed. The same processing is performed for channels 2, 3, and 4. As a result, the statistical calculation of the effective data Mc for autocorrelation function calculation in the 4-channel multipoint FCS general-purpose search method is performed for one _Mτ . All measured data M _max is scanned a total of 4 times. That is, for p _Mτ , M _max data is scanned p × 4 times to obtain a result.

Next, as a fourth part of the FCS calculation, an autocorrelation function is calculated using the following equation (2-4) (step S8).

Similar to the calculation of M _{c in} step S7 above, for channel 1, valid data used for the actual calculation is shown in FIG.

Only the part corresponding to. Autocorrelation calculation is

The results obtained by multiplying each piece of data by M _τ are summed. Similarly, the original data D [i] used in the multiplication, M _tau shifted by data D _[i + M τ] must be the data of the same channel. However, in order to extract this portion, first, all the measured data M _max (see FIG. 4) are sequentially scanned to determine the channel number.

That is, in the channel 1, while sequentially scanning all data M _max, and D [i], D each time a _[i + M τ] channel number is 1, and D [i], D [i + M τ] Multiply and calculate the total.

When one of D [i] and D [i + M _τ ] is determined to be another channel number by scanning, correlation calculation (multiplication and total calculation) is not performed.

The same processing is performed for channels 2, 3, and 4. As a result, the calculation of the autocorrelation function of the 4-channel multipoint FCS general-purpose search method uses all measured data for one _Mτ . Scan a total of 4 times each. That is, for p _Mτ , M _max data is scanned p × 4 times to obtain a result.

Next, as a fifth part of the FCS calculation, the result obtained in steps S5 to S8 is substituted into the following equation (2-5) to obtain a comprehensive calculation (final result) (step S9).

Statistical calculation of the average value calculation valid data number M _a in step S5, the average value calculated in step S6, statistical calculations and the autocorrelation function calculation in step S8 in the valid data number M _c for the auto-correlation function calculated in step S7 In each case, all the measured data M _max are scanned a total of four times for one M _τ . That is, for p _Mτ , a total of _Mmax total measurement data is scanned p × 4 × 4 times to obtain a result.

Specifically, when handling data measured for 10 seconds with a bin time set to 12.5 μsec, the total number of measurement data for calculation reaches 7.3 × 10 ⁵ or more.

4 For multipoint FCS channels for one M _tau, all data 7.3 × 10 ⁵ as measured by scanning a total of 4 × 4 times, for p number of M _tau, 7. The result is obtained by scanning 3 × 10 ⁵ data p × 4 × 4 times.

  The above calculation is performed while incrementing the channel number CH_NO by 1 (step S10), but when CH_NO = 4 (step S11), the FCS calculation ends (step S12).

(Second Embodiment)
FIG. 5 is a flowchart for explaining the procedure of multipoint FCS calculation by the table search method. First, the measurement data about the sample 50 is acquired by the configuration according to FIG. 1 (step S21). Thereby, a time series signal as shown in FIG. 6A is obtained. Next, this time-series signal is time-divided into multi-channels (here, channels ch0 to ch4) (step S22, FIG. 6B). In FIG. 6B, ch0 to ch4 are as described with reference to FIG.

  Next, a channel position information table is created based on the multi-channel time-series signal (FIG. 7) (step S22-1). The contents of each parameter in FIG. 7 are as shown in FIG.

  The channel position information table here can be created by one-time scanning of the data file after dividing the channel. First, a position information conversion table as shown in FIG. 9 is created, and each parameter (see FIG. 7) of the position information conversion table is statistically calculated and stored as an array. Become.

Next, calculation is started by substituting 1 as an initial value for channel number CH_NO (steps S23 and S24). Next, as a first part of the FCS calculation, using the following equation (2-6), performs statistical calculation of valid data number M _a for average calculation by table lookup method (step S25).

For channel 1, each variable in the equation is read by the position information data table (FIG. 10),

Only the data of is used for the average calculation. Other

The data corresponding to other channels, that is, portions corresponding to other channels are skipped and statistical calculation is not performed. Here, a sequential scan of measurement data having a long array as in the multipoint FCS general-purpose search method is not required. Therefore, calculation time can be shortened.

Next, as a second part of the FCS calculation, the average value is calculated by the table search method using the following equation (2-7) (step S26).

In channel 1, each variable in the equation is read by the position information data table (FIG. 10),

Calculate the total of the measured values. At the same time, the portion of data corresponding to the other channels,

The data is skipped. Thereby, calculation time can be shortened similarly to said step S25.

Next, a third portion of the FCS calculation, the statistical calculation of the autocorrelation function calculation valid data number M _c, performed using the following equation (2-8) (step S27). This part is a part that statistics the number of data necessary for calculating the autocorrelation function.

In channel 1, each variable in the equation is read by the position information data table (FIG. 10). In channel 1, the autocorrelation function is calculated

The results obtained by multiplying each piece of data by M _τ are summed. The value of M _c is the number of data used for autocorrelation calculation. In M _c, original data D [i] and M _tau shifted by data D used for the multiplication _[i + M τ] must be the data of the same channel. When performing the multiplication, the above-mentioned position information data table (FIG. 10) is referred to as needed, and the data of the same channel is obtained by the equation (2-8).

Only statistic is _counted as Mc. And the part corresponding to other channels

The data is skipped. In other words, it corresponds to other channels for one _Mτ .

Skip the pieces of data, for the p number of M _τ, also corresponds to the other channel

Data will be skipped. Therefore, it can be shortened effect of computation time as the number of M _tau often.

Next, as a fourth part of the FCS calculation, an autocorrelation function is calculated using the following equation (2-9) (step S28). Calculation of the autocorrelation function is similar to the time of statistical calculation of M _c.

Each variable in the equation is read by the position information data table (FIG. 10). In channel 1, the autocorrelation function is calculated

The results obtained by multiplying each piece of data by M _τ are summed. Similarly original data D used for the multiplication [i] and only M _tau shifting data D _[i + M τ] must be the data of the same channel.

In channel 1, by referring to the position information table (FIG. 10), data of the same channel as shown in equation (2-9)

Handles only for calculation, equivalent to other channels

Individual data can be skipped.

In other words, it corresponds to other channels for one _Mτ .

It is possible to skip the number of pieces of data and reduce the number of multiplications that require the most calculation time.

In the FCS computation, the error number is enough to curve fitting many of _M τ is small. That is, the accuracy of curve fitting increases because the data density increases. However, by increasing the M _tau, thereby repeating the calculation of steps S25 to S28.

Therefore, using all data for calculation leads to an increase in calculation time. Skipping the other channel data portions is very effective in increasing the calculation speed. In other words, it corresponds to other channels for one _Mτ .

To skip the data, for p _Mτ ,

Will be to skip the other channel data of the individual, it can be shortened calculation time as the number of M _τ.

Next, as a fifth part of the FCS calculation, the result obtained in steps S25 to S28 is substituted into the following equation (2-10) to obtain a comprehensive calculation (final result) (step S29).

Average statistical calculation of calculation valid data number M _a in step S25, the average value calculated in step S26, statistical calculations and the autocorrelation function calculation in step S28 in the valid data number M _c for the auto-correlation function calculated in step S27 In channel 1, all correspond to other channels for one _Mτ .

Pieces of data will be skipped, and for p pieces of _Mτ ,

The data of other channels will be skipped.

Specifically, when handling data measured for 10 seconds, if the bin time is set to 12.5 μsec, the total number of measurement data for calculation reaches 7.3 × 10 ⁵ or more.

In the case of a 4-channel multipoint FCS, a position information table (FIG. 10) can be created by scanning only all measured data 7.3 × 10 ⁵ even for p M _τ . That's it. The calculation speed is at least p × 4 × 4 times faster than the general-purpose search method in which all 7.3 × 10 ⁵ measurement data are repeatedly scanned p × 4 × 4 times.

  FIG. 11 is an FCS graph showing the result of the FCS calculation.

(Third embodiment)
FIG. 12 is a flowchart for explaining the procedure of the FCCS calculation using two measurement data. First, two measurement data are acquired about the sample 50 by the structure by FIG. 1 (step S41). Thereby, a time series signal as shown in FIGS. 13A and 13B is obtained. Next, this time-series signal is time-divided into multi-channels (here, channels ch0 to ch4) (step S42, FIGS. 14A and 14B). 14A and 14B, ch0 is a random measurement data range during movement of the galvano scanner as means for XY scanning laser light. ch1 is the measurement data range of measurement point 1. ch2 is the measurement data range of measurement point 2. ch3 is the measurement data range of measurement point 3. ch4 is the measurement data range of the measurement point 4.

Next, calculation is started by substituting 1 as an initial value for the channel number CH_NO (steps S43 and S44). Next, as a first part of the FCCS calculation, using the following equation (3-1), performs statistical calculation of valid data number M _a for average calculation by a general purpose search method (Step S45).

Here, it is a part which calculates the number of effective data M _Aa and M _Ba for each average value by the following method in the two measurement data. Here, calculation of M _Aa will be described as an example.

  First, all the measurement data is simply scanned, and for channel 1, only +1 statistical calculation is performed only when the channel data is 1.

The statistical calculation of the average value calculation effective data number M _Aa is performed by the equation (3-1). For channel 1, the effective data used in the actual calculation is shown in FIG.

Only the part corresponding to. In order to extract this portion, all the measured data M _{A max} (see FIG. 14A) are sequentially scanned to determine the channel number. In other words, in channel 1, every data M _{A max} is sequentially scanned, and each time the channel number becomes 1, the value of M _{A a} is incremented by 1, and statistical calculation is performed, and other channel numbers are determined by scanning. In this case, the statistical calculation of M _{A a} is not performed.

The same processing is performed for channels 2, 3, and 4. As a result, the statistical calculation of the average number of effective data M _{A a} for calculating the average value in the 4-channel multipoint FCCS general-purpose search method is one M _τ (correlation All measured data M _{A max} are scanned a total of 4 times for the delay time at the time of calculation. That is, for p (total number of delay times during correlation calculation) _Mτ , M _{A max} data is scanned p × 4 times, and the result is obtained.

Next, as the second part of the FCCS calculation, an average value of the two measurement data is calculated. In the measurement data A, calculation is performed using the following formula (3-2) (step S46).

For channel 1, the effective data used in the actual calculation is shown in FIGS.

Only the part corresponding to. In order to extract this portion, all the data M _max (see FIGS. 15 and 16) measured in the same manner are sequentially scanned to determine the channel number. That is, in channel 1, while sequentially scanning all the data M _{A max} , every time the channel number becomes 1, the value of M _{A a} is summed up, and if it is determined by scanning that it is another channel number, The total calculation of M _{A a} is not performed. The same processing is performed for channels 2, 3, and 4. As a result, the average value of the 4-channel multipoint FCCS general-purpose search method is calculated for all measured data M _A for one M _τ . Scan _max four times each. That is, for p _Mτ , M _{A max} data is scanned p × 4 times to obtain a result.

Next, a third portion of the FCCS calculations, statistical calculations of the cross-correlation function calculation valid data number M _c, performed using the following equation (3-3) (step S47). This part is a part that statistics the number of data necessary for calculating the cross-correlation function.

For channel 1, the effective data used in the actual calculation is shown in FIGS.

Only the part corresponding to. Cross-correlation calculation

The results obtained by multiplying each piece of data by M _τ are summed. The value of M _ABc is the number of data used for cross-correlation calculation. In M _ABc, the original data D _A [i] to be used for multiplication, M _tau shifted by data D _{B [i} + M τ] must be the data of the same channel.

In this part extraction, all measured data are sequentially scanned to determine the channel number. That is, in channel 1, every time the channel number of all D _A [i] and D _B [i + M _τ ] becomes 1, the value of M _c is calculated by +1, and D _A [i] is obtained by scanning. If any of D _B [i + M _τ ] is determined to be another channel number, the statistical calculation of M _Abc is not performed.

The same processing is performed for channels 2, 3, and 4. As a result, the statistical calculation of the effective data M _ABc for calculating the cross-correlation function in the 4-channel multipoint FCCS general-purpose search method is performed for one M _τ . All the measured data M _Amax (M _Bmax ) are scanned a total of 4 times. That is, for p number of M _tau, obtaining results M _max (M _Bmax) pieces of data each scan p × 4 times.

Next, as a fourth part of the FCCS calculation, a cross-correlation function is calculated using the following equation (3-4) (step S48).

Similar to the calculation of M _{Abc in} step S47 above, for channel 1, the effective data used for the actual calculation is shown in FIGS.

Only the part corresponding to. Cross-correlation calculation is

The results obtained by multiplying each piece of data by M _τ are summed. Similarly, the original data D _A [i] to be used for multiplication, M _tau shifted by data D _{B [i} + M τ] must be the data of the same channel.

To extract this portion, first, all the measured data M _Amax (M _Bmax ) (see FIG. 15) are sequentially scanned to determine the channel number.

That is, in the channel 1, while sequentially scanning all data M _Amax, and D _A [i], each time the D channel number _{B [i} + M _τ] is 1, and D _A [i], D _B Multiply [i + M _τ ] to calculate the total.

When it is determined by scanning that either D _A [i] or D _B [i + M _τ ] is another channel number, correlation calculation (multiplication and total calculation) is not performed.

Similar processing is also in the channels 2, 3 and 4 is performed, and consequently, the calculation of the cross correlation function of 4-channel multi-point FCCS generic search policy, for one M _tau, all data M measured _Amax (M _Bmax ) is scanned 4 times in total. That is, for p number of M _tau, get results, respectively M _Amax and (M _Bmax) pieces of data by scanning p × 4 times.

Next, as a fifth part of the FCCS calculation, the result obtained in steps S45 to S48 is substituted into the following equation (3-5) to obtain a comprehensive calculation (final result) (step S49).

Average statistical calculation of calculation valid data number M _a in step S45, the average value calculation, the cross-correlation function of the statistical calculation and step S48 of the cross-correlation function calculation valid data number M _c in step S47 in step S46 Substituting the calculation result into Expression (3-5), the total calculation (final result) is obtained.

  According to the embodiment described above, in multi-point FCS and FCCS data analysis, autocorrelation functions and cross-correlation functions at a plurality of measurement points are estimated at the same time, and the order of measurement at a plurality of points is freely combined, and the cell response The interaction between proteins can be analyzed. In particular, when compared with single point FCS and FCCS, in the case of multipoint FCS and FCCS, a plurality of point points can be observed at the same time, and various reactions inside and outside the living cells can be captured.

  In addition, by adopting table search method multipoint FCS and FCCS, it avoids repeated scanning of all data for channel position judgment by general search method, and acquires all data at each measurement point with only one scan. It becomes possible to do. By simply referring to this channel position information table at each stage of autocorrelation and cross-correlation calculation, the data of channels other than the calculation target can be skipped, and only the minimum necessary data can be taken into the calculation, thereby shortening the calculation time. Multiple measurement points can be measured and observed in real time. In particular, as the number of measurement points (number of channels) increases, the amount of data skipped increases, and the calculation time can be significantly shortened.

  Furthermore, with a measuring device configured based on the same principle as the conventional method of estimating the autocorrelation function and cross-correlation function at a single point, it is possible to perform alternating measurement at multiple measurement points for a short time and analyze the sequential measurement data. Although it is a single measurement point, it is possible to capture molecular behavior, reaction, etc. at a plurality of measurement points in the specimen in real time. In other words, the points to be measured are sequentially moved inside and outside the cell, and the obtained measurement data is time-divided into a series of time-series data for each measurement point, and autocorrelation functions and cross-correlation functions are set for each time-division measurement data. The values of the autocorrelation function and the cross-correlation function at a plurality of measurement points can be estimated almost simultaneously.

  Furthermore, according to the present invention, light intensity distribution analysis can be performed from the obtained time series signal. As a result, the binding state and density distribution of different fluorescent molecules can be obtained.

It is a figure which shows the structure of the optical signal analyzer to which this invention is applied. It is a flowchart for demonstrating the procedure of the multipoint FCS calculation by a general purpose search system. (A) shows the measurement data (time series signal) about the sample, (B) shows the time series signal being time-divided into multi-channels, and (C) is an example of each measurement point 1-4. FIG. It is a figure used for the statistical calculation of the effective data number Ma for average value calculation. It is a flowchart for demonstrating the procedure of the multipoint FCS calculation by a table search system. (A) shows the measurement data (time series signal) about the sample, and (B) is a diagram showing the time series signal being time-divided into multi-channels. It is a figure used for channel position information table preparation. It is a figure which shows each parameter in a channel position information table. It is a figure which shows a position information variable table. It is a figure which shows a positional infomation data table. It is a FCS graph which shows the result of FCS calculation. It is a flowchart for demonstrating the procedure of the general purpose system FCCS calculation using two measurement data. It is a figure which shows the measurement data (time series signal) about a sample. FIG. 14 is a diagram illustrating a state in which the time series signal illustrated in FIG. It is a figure used for statistical calculation of effective data number M _Aa for average value calculation about two measurement data. It is a figure used for statistical calculation of effective data number _MBB for average value calculation about two measurement data.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Laser scanning type confocal optical microscope main body, 11 ... CCD camera, 12 ... Sample stage, 13 ... Sample stage controller, 14 ... Computer (PC), 15 ... TV monitor, 16 ... Laser light source.

Claims (7)

A detection step of detecting light emitted from a measurement point designated with respect to the sample by a photodetector;
A time division step of time-dividing the time-series signal output from the photodetector into a plurality of time-series signals;
Analyzing the time-divided time series signal for each time series, and analyzing the physical properties of the measurement points;
An optical signal analysis method comprising: A plurality of points related to the sample are sequentially selected as measurement points, the position irradiated with the light is sequentially moved with respect to the plurality of measurement points, and a time-series signal output from the photodetector is detected for each of the plurality of measurement points. 2. The optical signal analysis method according to claim 1, wherein time-division signal analysis is performed for each of the time-divided data, and a physical property for each of the plurality of measurement points is obtained. The time series signal output from the photodetector is time-divided and stored for each of the plurality of measurement points, and the time series signal for each of the plurality of measurement points is read and analyzed as necessary. 3. The optical signal analysis method according to claim 2, wherein a physical property for each of a plurality of measurement points is obtained. Obtain position information of the plurality of measurement points to create a position information table, analyze the time-series signal with reference to the position information table, and obtain physical properties of the sample for each of the plurality of measurement points. The optical signal analysis method according to claim 3. The optical signal analysis method according to claim 1, wherein the light emitted from the measurement point is fluorescence, phosphorescence, chemiluminescence, bioluminescence, reflected light, or scattered light. The optical signal analysis method according to claim 1, wherein the analysis of the time-series signal output from the photodetector is autocorrelation spectroscopy analysis or light intensity distribution analysis. The physical property of the measurement point with respect to the sample is a translational diffusion coefficient, a rotational diffusion coefficient of molecules or particles, or the number of molecules or particles present in the measurement region, or a fluorescence lifetime. The optical signal analysis method according to any one of 1 to 6.

Images