Profile generation device, profile generation method, and profile generation program

Abstract

Provided are a profile generation device, method, and program that can allow sampling (adaptive hyper sampling) set at arbitrary intervals instead of regular intervals and can improve a resolution or a S / N ratio, while maintaining a short distance between a sample and a detector. A profile generation device 200 for generating an X-ray scattering profile by performing sampling on X-ray scattering intensity data obtained by a pixel array detector comprises: a sampling position setting unit 223 for setting a plurality of sampling positions that are not at regular intervals in the wave number direction of a scattering vector on the basis of a specific relationship; a data extraction unit 225 for extracting intensity data in pixels overlapping the set sampling positions; and a profile generation unit 227 for generating a profile plotted on the basis of the extracted intensity data.

Description

Profile generation device, profile generation method, and profile generation program

[0001] The present invention relates to a profile generating device, method, and program for generating an X-ray scattering profile by sampling X-ray intensity data obtained by a pixel array detector.

[0002] Small-angle X-ray scattering is used to observe biopolymers, and programs for primary data processing and data analysis have been developed. To acquire the measurement data, pixel detectors with [1] high resolution and [2] high sensitivity are used, but there are limitations to their performance.

[0003] [1] The lack of resolution has been compensated for by increasing the distance between the sample and the detector from a few meters to, in some cases, 10 meters or more. However, in this case, even if a vacuum is created between the sample and the detector, the degree of vacuum is not 100%, and the X-rays are attenuated, resulting in a decrease in sensitivity.

[0004] Meanwhile, image processing methods for improving resolution have been developed outside the field of X-ray diffraction. For example, Non-Patent Document 1 discloses a method for dividing input pixels into multiple output pixels for astronomical observation image data in order to obtain high-resolution images without sacrificing the signal-to-noise ratio. Non-Patent Document 2 proposes upsampling of speckle patterns for coherent X-ray diffraction imaging, referring to the method described in Non-Patent Document 1.

[0005] [2] Insufficient sensitivity can be compensated for by pixel binning, which groups pixels together to increase the light-receiving area per pixel. Non-Patent Document 3 discloses a method for increasing sensitivity by combining multiple pixels in areas with low sensitivity and low information density into a single piece of data.

[0006] AS Fruchter and RN Hook, “Drizzle: A Method for the Linear Reconstruction of Undersampled Images”, USA, [v1] Mon, 10 Aug 1998, [v2] Fri, 19 Oct 2001, Publications of the Astronomical Society of the Pacific, 2002, Vol.114 (792), p.144-152Y. Chushkin and F. Zontone, “Upsampling speckle patterns for coherent X-ray diffraction imaging”, Journal of Applied Crystallography, (2013). 46, 319-323Tatsuya Arai, Rena Inamasu, Hiroki Yamaguchi, Daisuke Sasaki, Ayana Sato-Tomita, Hiroshi Sekiguchi, Kazuhiro Mio, Sakae Tsuda, Masahiro Kuramochi, Yuji C. Sasaki, “Laboratory diffracted x-ray blinking to monitor picometer motions of protein molecules and application to crystalline materials”, Struct Dyn, 2021 Jul 8

[0007] As mentioned above, various attempts have been made in the field of image analysis, but in the field of X-ray diffraction image analysis, the dominant approach is to simply increase the camera length, and no upsampling technique has been applied, which splits the signal from a single pixel into multiple pixels.

[0008] In small-angle X-ray scattering measurements, X-rays, which have different [1] data densities and [2] intensities depending on the detection position, are detected by a pixel array detector with uniform pixel density and sensitivity. Therefore, particularly in data acquisition by X-ray detection, there is a demand for improved resolution with low sensitivity in the low-angle scattering region (hereinafter abbreviated as "low-angle side" or "low-angle region") where X-rays are received with high intensity and high data density, and there is a demand for improved sensitivity without high resolution in the high-angle scattering region (hereinafter abbreviated as "high-angle side" or "high-angle region") where the intensity is low and the data density is low.

[0009] In particular, there is a demand for improved resolution at low angles when processing X-ray diffraction images obtained by measuring biopolymers with a size of, for example, 1000 Å to 3000 Å. Although this demand has been increasing in recent years, upsampling has only been recognized in the field of imaging as an image processing technique that can be applied uniformly regardless of position.

[0010] The present invention has been made in consideration of these circumstances, and aims to provide a profile generation device, method, and program that enables sampling at arbitrary intervals rather than fixed intervals (adaptive hypersampling), maintains a short distance between the sample and the detector, and can improve at least one of resolution and sensitivity depending on how it is applied.

[0011] (1) In order to achieve the above object, the profile generation device of the present invention is a profile generation device that generates an X-ray scattering profile by sampling X-ray scattering intensity data obtained by a pixel array detector, and is characterized by comprising: a sampling position setting unit that sets multiple sampling positions that are not spaced at regular intervals in the wave number direction of the scattering vector based on a specific relationship; a data extraction unit that extracts intensity data at pixels that overlap with the set sampling positions; and a profile generation unit that generates a plotted profile based on the extracted intensity data.

[0012] (2) In the profile generating device described in (1) above, the sampling position setting unit sets sampling positions in which the sampling interval on the high angle side is larger than the sampling interval on the low angle side.

[0013] (3) In the profile generating device described in (2) above, the sampling position setting unit sets sampling positions at a sampling interval smaller than the pixel size on the low angle side.

[0014] (4) In the profile generating device described in (2) above, the sampling position setting unit sets sampling positions at a sampling interval greater than the pixel size on the high-angle side.

[0015] (5) In the profile generating device described in (3) above, the sampling position setting unit sets sampling positions at a sampling interval greater than the pixel size on the high-angle side.

[0016] (6) Furthermore, in the profile generation device described in any one of (1) to (5) above, the sampling position setting unit is characterized in that it sets the sampling positions using a function that gives a sampling interval with respect to the wave number of the scattering vector as the specific relationship.

[0017] (7) Furthermore, in the profile generating device described in any one of (1) to (5) above, the sampling position setting unit is characterized in that, as the specific relationship, it sets the sampling positions in a relationship that gives a discrete, constant sampling width for each wavenumber range of the scattering vector.

[0018] (8) Furthermore, the profile generation device described in any one of (1) to (7) above is characterized by further comprising a first index calculation unit that calculates a first index representing the variability of data from each of the profiles generated by changing the parameters in the specific relationship, and a parameter determination unit that determines the parameters for which the first index satisfies a criterion.

[0019] (9) Furthermore, in the profile generating device described in (8) above, the first index is characterized in that it represents the irregularity of the generated profile using at least one of the standard deviation and the distribution slope of the generated profile.

[0020] (10) Furthermore, in the profile generating device described in (8) above, the first index is characterized in that it represents the occurrence of outliers in the generated profile using the standard deviation of the generated profile.

[0021] (11) Furthermore, the profile generating device described in any one of (1) to (10) above is characterized by further comprising a position shifting unit that shifts the multiple sampling positions that have already been set by a specific shift amount in the wave number direction of the scattering vector, and a shift amount determining unit that determines the specific shift amount at which the first index satisfies a criterion.

[0022] (12) The profile generating device according to any one of (1) to (5) above is characterized by further comprising: a Guinier plot executing unit that executes a Guinier plot for each profile generated by changing a parameter in the specific relationship; and a parameter determining unit that determines the parameter with the highest uniformity of the radius of gyration Rg obtained by the Guinier plot.

[0023] (13) Furthermore, in the profile generating device described in (12) above, the parameter determining unit is characterized in that it determines a parameter that specifies a sampling width in the low scattering angle region based on the uniformity of the radius of gyration Rg, and determines a parameter that specifies a sampling width in the high scattering angle region based on the type of sample.

[0024] (14) Furthermore, the profile generating device described in any one of (1) to (13) above is characterized by further comprising: a second index calculation unit that calculates a second index indicating the degree of deviation due to processing of the profile generated by changing the parameter in the specific relationship; and a parameter determination unit that determines the parameter that is suitable for a criterion using the second index.

[0025] (15) Furthermore, the profile generating device described in any one of (1) to (14) above is characterized by further comprising: a minimum wave number calculation unit that calculates the wave number of the smallest scattering vector having a significant standard deviation for a profile obtained based on each of the profiles generated by changing the parameters in the specific relationship; and a parameter determination unit that determines the parameter that is suitable for a criterion using the wave number of the calculated smallest scattering vector.

[0026] (16) Furthermore, a profile generation method of the present invention is a profile generation method for generating an X-ray scattering profile by sampling X-ray scattering intensity data obtained by a pixel array detector, characterized by including the steps of: setting a plurality of sampling positions not spaced at regular intervals in the wave number direction of the scattering vector based on a specific relationship; extracting intensity data at pixels that overlap with the set sampling positions; and generating a profile plotted based on the extracted intensity data.

[0027] (17) Furthermore, a profile generation program of the present invention is a profile generation program that generates an X-ray scattering profile by sampling X-ray scattering intensity data obtained by a pixel array detector, and is characterized in that it causes a computer to execute the following processes: a process of setting multiple sampling positions that are not spaced at regular intervals in the wave number direction of the scattering vector based on a specific relationship; a process of extracting intensity data at pixels that overlap with the set sampling positions; and a process of generating a plotted profile based on the extracted intensity data.

[0028] 1 is a schematic diagram showing a profile generation system according to the present invention; FIG. 2 is a perspective view showing an X-ray analysis apparatus; FIG. 3 is a block diagram showing the configuration of a profile generation system according to the present invention; FIG. 4 is a schematic diagram showing a measurement method; FIG. 5 is a flowchart showing a method for generating analysis data; FIG. 6 is a flowchart showing a method for generating and determining a profile according to a first embodiment; FIG. 7 is a schematic diagram showing acquisition of measurement data; FIG. 8 is a graph showing an example of sampling positions; FIG. 9 is a schematic diagram showing an example of sampling; FIG. 10 is a schematic diagram showing an example of intensity apportionment; FIG. 11 is a schematic diagram showing integration of a profile; FIG. 12 is a diagram showing an example of a setting screen for adjusting the relationship between the wave number of a scattering vector and the interval between sampling positions; FIG. 13 is a flowchart showing a method for generating and determining a profile according to a fourth embodiment; FIG. 14 is a graph showing the relationship between the number of divisions and the evaluation value of an error function; FIG. 15 is a graph showing profiles with a plurality of different numbers of divisions; (a) and (b) are graphs showing unadjusted and adjusted profiles, respectively.

[0029] Next, an embodiment of the present invention will be described with reference to the drawings. To facilitate understanding of the description, the same reference numerals are used to designate the same components in the drawings, and duplicated descriptions will be omitted.

[0030] [Principle] In this invention, a pixel-divided sampling method is used to sample the X-ray diffraction image obtained by measurement. While this method is known as a common image processing method, it has not been used to improve the resolution of X-ray diffraction images. By using this type of sampling, the present invention can generate high-resolution X-ray diffraction profiles using a measurement device that can be installed in a laboratory because the distance between the sample and the detector is small.

[0031] When observing objects that can take various forms, such as biopolymers, the importance of the acquired X-ray diffraction image varies depending on the area. For example, it may be necessary to observe the entire structure of a biopolymer as well as a portion of its surface structure. In such cases, high resolution at low angles and high sensitivity at high angles are required for the X-ray diffraction image acquired in a single measurement. This invention achieves profile generation using non-uniform sampling that matches the purpose of the observation.

[0032] (Variation range of X-ray diffraction intensity) In X-ray diffraction, the real space measured by the detector is a reciprocal lattice, and the amount of information per unit space increases inversely proportionally at small scattering angles. For example, if the origin of the scattering angle is included, the amount of information reaches infinity.

[0033] On the other hand, as the scattering angle increases, the information density decreases dramatically, and the diffraction intensity that can be obtained in principle due to the finite atomic size drops significantly. In reality, due to the existence of atomic fluctuations, the diffraction intensity decreases significantly from the low angle region to the high angle region, and the range of change is 10 ６～１０ This is a factor of 100. In other words, it far exceeds the dynamic range of imaging devices such as ordinary detection elements, photosensitive materials such as film, or CCD / CMOS. Various efforts have been made to deal with this range of change, but the effectiveness has been limited. The scattering angle is equivalent to the wave number of the scattering vector, and conversion between the two is possible. Therefore, the expression "wave number on the high-angle side or low-angle side" means "wave number corresponding to the scattering angle on the high-angle side or low-angle side."

[0034] (Physical Limits of Pixel Array Detectors) Pixel array detectors, which have revolutionized the world of X-ray diffraction, utilize semiconductor technology to miniaturize the equivalent of a single ultrafast counting detector and pack them into a single pixel, creating a one- or two-dimensional array detector. As a result, they routinely enable ultrahigh-sensitivity single-photon measurement, as well as high-precision, ultrafast measurement of extremely intense incident X-rays.

[0035] However, due to the principle that each pixel constituting a pixel array detector constitutes one counting detector (sensor, counting circuit, and digital processing device), there is a physical limit to how small the pixel size can be. Due to this physical limit, pixel array detectors are currently insufficient as detectors used for ordinary X-ray diffraction measurements.

[0036] (Long-distance measurement) Therefore, one method is to obtain diffraction data with sufficient spatial resolution by increasing the distance between the sample and the detector. This method overcomes the pixel size limitation by measuring the radially spreading diffraction pattern at a long distance. For example, in small-angle measurements called USAX (Ultra- / Small Angle X-ray Scattering), the sample-detector distance must be set to at least several meters, usually several tens of meters.

[0037] However, increasing the distance between the sample and the detector significantly reduces the versatility of the measurement device. It also poses challenges in maintaining a high vacuum in a large volume, such as the need to create a chamber that can withstand atmospheric pressure. Meanwhile, signal strength decreases due to X-ray diffusion and scattering. Long-distance measurements further reduce the S / N ratio (sensitivity) in the high-angle region, where signal strength is weaker in principle.

[0038] For example, in polymer measurement, resolution in the low-angle region determines whether measurement is possible, but the element size of a pixel array detector does not meet the required resolution. Therefore, the solution is to use a detector with the smallest pixel size available and place it far away from the sample. This comes at the expense of reduced sensitivity and extremely limited range of measurement in the high-angle region.

[0039] (Pixel Binning) Not only for pixel array detectors, but also for pixel-type detectors in general (such as CCD detectors), there is a technique called pixel binning, which increases sensitivity by treating multiple pixels as a group. In fact, pixel binning is used in measurements where measurements are mainly performed on the high-angle side and resolution on the low-angle side is not required. However, with pixel binning, the side of the pixel used is doubled or tripled, so it is not suitable for measurements in the low-angle range where resolution is insufficient.

[0040] (Upsampling) There is also a method called upsampling or oversampling, which virtually divides pixels and samples them to improve resolution. Upsampling can achieve high resolution when measuring low-angle areas. However, uniform upsampling is not suitable for analyzing high-angle areas, which have low sensitivity.

[0041] (Adaptive Hypersampling) Given the above circumstances, the inventors of the present invention came up with the idea of ​​continuously changing the interval between sampling positions, enabling upsampling and pixel binning simultaneously. This is a method of performing non-uniform and free sampling on a single measured scattering image according to an angular region based on the scattering angle, and we call it adaptive hypersampling. Adaptive hypersampling generates high-sensitivity, high-resolution profiles from the obtained diffraction data, enabling precise analysis of macromolecules in particular. The configuration and operation for realizing these functions are described in detail below.

[0042] [First Embodiment] [Profile Generation System] Fig. 1 is a schematic diagram showing a profile generation system 10. The profile generation system 10 includes an X-ray analysis apparatus 100 and a profile generation apparatus 200. The X-ray analysis apparatus 100 irradiates a sample S0 with X-rays and detects small-angle scattered X-rays. The sample S0 is preferably a polymer in solution, particularly a biopolymer. This is particularly effective when the sample is a pharmaceutical molecule, molecular complex, or structure in solution that requires structural analysis of 30 Å or less.

[0043] The profile generating device 200 is composed of a computer 210, an input device 280, and an output device 290, and controls the operation of the X-ray analyzing device 100, and also acquires and processes measurement data from the X-ray analyzing device 100.

[0044] The X-ray analysis apparatus 100 includes an X-ray generation unit 110, a sample loading mechanism 120, a pixel detector 130, and a control unit 140. The detector X-ray generation unit 110 has an X-ray source 111 and irradiates X-rays onto a sample S0. It is preferable to use Cu as the target of the X-ray source 111, but Co may also be used.

[0045] The sample loading mechanism 120 sends the target solution together with the sample holding tube to the X-ray irradiation position. Alternate irradiation of each solution is possible by switching the sample holding tube 125 by the sample loading mechanism.

[0046] The target solutions include sample solutions and reference solutions. The sample solution is a solution containing a sample, for example, a solution containing a biopolymer and a special component for retaining the biopolymer. The reference solution is a solution obtained by removing the sample from the sample solution. For example, the reference solution in the above example is a solution that does not contain a biopolymer but contains a special component.

[0047] The pixel detector 130 detects X-rays scattered by the sample S0 and transmits the resulting scattered image to the computer 210 as measurement data. The pixel detector 130 is preferably a hybrid pixel array detector. The device can be configured with the distance between the sample S0 and the pixel detector 130 set to within 1 m. For example, the X-ray analysis device 100 can be configured for desktop use with the sample-detector distance set to approximately 40 cm.

[0048] In the above configuration, one pixel detector 130 is provided for one X-ray beam emitted from the X-ray source 111, but other configurations may also be employed. For example, the X-ray analysis apparatus 100 may be configured to irradiate two equal parallel beams in the same direction using a mirror or a slit, and detect the scattered rays thereof with one pixel detector. Alternatively, the X-ray analysis apparatus 100 may be configured to irradiate two equal beams in opposite directions, and detect the scattered rays thereof with two pixel detectors, respectively.

[0049] The computer 210 is, for example, a PC, and is configured with a processor that executes processing, and a memory or hard disk that stores programs and data, etc. The computer 210 receives user input from an input device 280 such as a keyboard or mouse, and outputs a profile, etc. to an output device 290 such as a display.

[0050] The computer 210 may be a server device located on the cloud. In addition, in terms of processing load, the function of controlling the operation of the X-ray analysis apparatus 100 and the function of processing the measurement data may be separated, with the control being performed by a PC installed on-site and the data processing being performed by a server device.

[0051] [X-ray Analysis Apparatus] Fig. 2 is a perspective view showing an X-ray analysis apparatus 100. The X-ray analysis apparatus 100 includes an X-ray source 111, an optical system 115, a Kratsky block 117, a sample holding tube 125, and a pixel detector 130. The X-ray source 111 is a line radiation source or a point radiation source, and emits a diverging beam. The optical system 115 is, for example, a KB parallel type or a serial type optical system.

[0052] The pair of Kratky blocks 117 interact with the X-rays through their respective edges to define one side of the X-ray beam, thereby eliminating parasitic scattering from the irradiated X-rays. The sample holding tube 125 delivers and holds 5 to 10 μl of solution.

[0053] The pixel detector 130 detects X-rays scattered by the solution. The pixel detector 130 is a pixel array detector in which ultra-high speed counting detectors are miniaturized using semiconductor technology and densely packed into one pixel. The X-ray analysis device 100 transmits the detected scattered image to the profile generation device 200. The scattered image detected at every predetermined time t is transmitted as measurement data.

[0054] 3 is a block diagram showing the configuration of the profile generation system 10. The profile generation device 200 acquires a scattered image as measurement data from the X-ray analysis device 100, and generates a profile by sampling the scattered image. The functions of the profile generation device 200 are mainly realized by a computer 210.

[0055] The computer 210 includes an input / output control unit 211, a measurement control unit 215, a measurement data storage unit 217, a sampling position setting unit 223, a data extraction unit 225, a profile generation unit 227, a first index calculation unit 232, a minimum wave number calculation unit 236, a parameter determination unit 238, a position shift unit 241, a Guinier plot execution unit 242, a second index calculation unit 244, a shift amount determination unit 247, and an analysis data generation unit 251. Each unit can send and receive information via a control bus L.

[0056] The input / output control unit 211 accepts input from the input device 280 and controls output to the output device 290. The input / output control unit 211 can accept input of measurement conditions, for example. The measurement conditions include the intensity of the generated X-rays, the position of the sample and the solution to which the X-rays are irradiated, the arrangement of the pixel detector 130, and the measurement time t when acquiring the scattering image. The input / output control unit 211 can also output measurement data, sampling positions, indexes, profiles, etc.

[0057] The measurement control unit 215 controls the operation of the X-ray analysis apparatus 100. Controlled operations include sending out the sample, generating X-rays, moving the sample position and the pixel detector 130, and transmitting measurement data. Each part of the X-ray analysis apparatus 100 is controlled by control instructions sent to the control unit 140 within the X-ray analysis apparatus 100. This operation control causes the X-ray analysis apparatus 100 to irradiate the solution with X-rays, detect scattered X-rays with the pixel detector 130, and transmit a scattered image. In addition, the solution irradiated with X-rays is alternately changed between the reference solution and the sample solution at predetermined time intervals.

[0058] The measurement data storage unit 217 stores, as measurement data, the scattering image detected by the pixel detector 130. The stored measurement data is subjected to sampling and is used for generating a profile, etc.

[0059] The sampling position setting unit 223 sets multiple sampling positions that do not have a fixed interval with respect to the wave number of the scattering vector based on a specific relationship. The specific relationship is a relationship that determines the wave number of the scattering vector and the interval between the sampling positions, such as a function. This enables sampling to be performed at an arbitrary interval rather than a fixed interval (adaptive hypersampling). Note that the interval between the sampling positions does not need to be an integer division or an integer multiple of the basic bin defined by the pixels, but the following description will be given as an example in which the interval is an integer division or an integer multiple of the basic bin. Note that, as described below, the wave number may be divided into several areas, and the divided areas may be called bins.

[0060] For example, sampling can be performed at intervals smaller than one pixel on the low angle side of the wave number direction of the scattering vector (upsampling), or at intervals larger than one pixel on the high angle side of the wave number direction of the scattering vector (downsampling), thereby improving the spatial resolution of the readout by upsampling on the low angle side or improving the S / N ratio by downsampling on the high angle side while maintaining a short distance between the sample and the detector.

[0061] It is preferable that the sampling interval on the high angle side is larger than that on the low angle side, which allows sampling under conditions where the distribution of sampling intervals is inclined along the wave number of the scattering vector.

[0062] The sampling position setting unit 223 can set sampling positions at a sampling interval smaller than the pixel size on the low-angle side, thereby performing upsampling on the low-angle side and improving the resolution.

[0063] The sampling position setting unit 223 can also set sampling positions at a sampling interval greater than the pixel size on the high-angle side, thereby generating a profile with high sensitivity and an excellent S / N ratio even in the q range on the high-angle side where the signal strength drops sharply.

[0064] It is also possible to perform both upsampling at low angles and downsampling at high angles for a single measurement data. Particularly when observing polymers in solution, precise information about molecular structure can be easily obtained from a single scattering image.

[0065] The sampling position setting unit 223 preferably sets the sampling positions using a function that determines the sampling interval relative to the wave number of the scattering vector as a specific relationship. This allows the sampling interval to be set for each sampling position. As a result, flexible sampling is possible depending on the sample and purpose, and the sampling positions can be adjusted to avoid artifacts caused by changes in sampling conditions.

[0066] For example, a function form can be used depending on the target sample and purpose. The distribution of sampling intervals can be adjusted by changing the parameters (coefficients) of the function. Changing the parameters makes it easier to compare conditions.

[0067] The sampling position setting unit 223 can also set sampling positions in a specific relationship that provides a discrete, constant sampling width for each wave number range of the scattering vector. This enables adaptive hypersampling by changing the sampling interval for each range, even in programs that have a fixed sampling interval as the default setting. In other words, the sampling interval can be allocated in a stepped manner according to the wave number of the scattering vector.

[0068] The data extraction unit 225 extracts intensity data from pixels occupying an area including the set sampling positions. Since intensity data is extracted for each set sampling position in the direction of the wave number q of the scattering vector, sampling can be performed at any position without being restricted by pixel division or group size. In this embodiment, the sampling interval is set based on the wave number of the scattering vector, but the sampling interval may also be set based on the scattering angle 2θ.

[0069] The profile generator 227 generates a profile plotted based on the extracted intensity data against the wave number of the scattering vector, generates a profile at each fixed angle in the circumferential direction (φ direction) around the center of the scattering image, and integrates the intensity along the φ direction to generate the profile.

[0070] The first index calculation unit 232 calculates a first index representing the variability of data from each profile generated by changing parameters in a specific relationship. The first index can be expressed as a combination of indices indicating the variability of data in the generated profile or the deviation from the original value. The index can be obtained by multiplying and adding the standard deviation of the profile, the second derivative of the profile, the maximum value of the profile slope, etc. The original value refers to the value of the profile when sampled at equal intervals. The deviation from the original value includes artifacts caused by this correction.

[0071] This allows identifying a profile with reduced irregularity, resulting in a natural and smooth profile. The first index may also be an index that represents the occurrence of outliers in the generated profile using the standard deviation of the generated profile. In this case, a profile with fewer outliers can be identified.

[0072] The minimum wave number calculation unit 236 calculates the wave number of the smallest scattering vector with a significant standard deviation for the profiles obtained based on each of the profiles generated by changing the parameters in the specific relationship, thereby making it possible to identify profiles with high resolution.

[0073] The parameter determination unit 238 determines parameters that improve the resolution on the low-angle side of the profile and the sensitivity on the high-angle side. The parameter determination unit 238 can determine parameters whose first index satisfies the criterion for parameter determination. This makes it possible to identify parameters of profiles with improved S / N ratios among the generated profiles. Upsampling at low scattering angles is particularly effective because the optimal sampling position varies slightly for each sample.

[0074] When specifying the sampling position, it is determined whether the first index in the region where the range on the low-angle side of the wave number q of the scattering vector is narrow satisfies a criterion, and when determining the shift amount, it is determined whether the first index over the entire wave number q of the scattering vector satisfies a criterion. When calculating the first index, a value accompanying the data on the intensity on the vertical axis relative to q on the horizontal axis is used. Also, the region where the range on the low-angle side is narrow is, for example, 0.1 Å in the range of q. －１ The following range is meant: "0.1 Å" －１ The "range below" includes a slightly larger range and a slightly smaller range. This range may be one bin or may span several bins.

[0075] The parameter determination unit 238 can also determine parameters suitable for the criteria using the second index. This allows binning to be performed on the high-angle side, with a large sampling width, without overly smoothing the unevenness of the profile. The second index will be described later.

[0076] The parameter determination unit 238 can also determine parameters suitable for a criterion using the wave number of the calculated smallest scattering vector. In this case, a profile with high resolution can be identified. When determining parameters based on two or more criteria, it is also possible to narrow down the profiles to a certain number using one criterion and identify the parameters of the optimal profile from that set of profiles.

[0077] The parameter determination unit 238 can also determine a parameter with high uniformity of the radius of gyration Rg obtained by the Guinier plot. In this case, the standard deviation of the radius of gyration Rg can be used. This allows the low-angle sampling width to be determined without calculating the first index and without imposing a processing load. Furthermore, if the type of sample is known, a parameter specifying the sampling width in the high-angle scattering angle region can also be determined depending on the type of sample.

[0078] The position shifter 241 shifts multiple sampling positions that have already been set by a specific shift amount in the wave number direction of the scattering vector. When data is sampled at periodic intervals, beat patterns (moiré artifacts) may become prominent. Such beat patterns can be reduced by slightly shifting the phase of the periodic sampling positions. By shifting the sampling positions by a specific shift amount, it is possible to confirm how the beat patterns appear due to sampling. This type of shift is known as an origin shift. Note that while there is a method of shifting the sampling positions by a specific shift amount as described above, there is also a method of shifting the sampling positions by changing the number of divisions.

[0079] The Guinier plot execution unit 242 executes a Guinier plot for each profile generated by changing parameters in a specific relationship. Specifically, the horizontal axis represents the square of the scattering angle and the vertical axis represents the logarithm of the scattering intensity, and the measurement results are plotted to determine the radius of gyration Rg from the gradient of the approximate line.

[0080] The second index calculation unit 244 calculates a second index that indicates the degree of deviation from the profile obtained by uniform sampling. The second index can determine how gentle the binning width on the high-angle side should be. The second index may be the difference between the difference between each of the profiles generated by changing parameters in a specific relationship and the profile obtained by uniform sampling, and σ. This makes it possible to evaluate binning that does not smooth out unevenness too much.

[0081] The shift amount determination unit 247 determines a specific shift amount for which the first index satisfies the criteria for shift amount determination. This determination allows a profile with reduced beat patterns to be identified. Note that when determining the shift amount, the first index for the entire wave number q of the scattering vector is used.

[0082] The analytical data generating unit 251 adjusts the scale so that the background intensities of the sample solution profile and the reference solution profile match within the wavenumber range of the obtained predetermined scattering vector, and then subtracts the reference solution profile from the sample solution profile to generate analytical data.

[0083] [Measurement Method] Fig. 4 is a schematic diagram showing the measurement method. As shown in Fig. 4, first, a reference solution and a sample solution are prepared, and then X-rays are irradiated onto each solution to obtain a scattering image as measurement data. At this time, it is preferable to alternately obtain measurement data for the reference solution and the sample solution.

[0084] Multiple measurement data are integrated along the time axis. Then, sampling is performed on the total measurement data to generate a profile at each φ. At this time, adaptive hypersampling is performed, regardless of pixel size. The profiles are then integrated in the φ direction to generate an integrated profile. After adjusting the scale, the profile of the reference solution is subtracted from the profile of the sample solution to generate a profile of the sample.

[0085] The obtained sample profile can be used for structural analysis. The cubic volume of the real space containing the particle is represented by cubic voxels discretized into an NxNxN grid, and an electron density map can be calculated by searching for structure factors based on the measured profile.

[0086] Specifically, multiple structural models are generated from the measured X-ray scattering profile, and a calculated X-ray scattering profile is calculated from each of the multiple structural models. An index representing the degree of agreement between the calculated calculated X-ray scattering profile and the measured X-ray scattering profile is calculated, and the multiple structural models are output as an ensemble of molecular shape data based on the calculated index. Note that one representative structural model may be selected from the multiple structural models. In this way, a structural model of a polymer in solution that has a structure with dynamic fluctuations can be accurately reproduced.

[0087] [Method for Generating Analysis Data] The process from acquiring measurement data to generating analysis data in the above measurement method will be described below. Fig. 5 is a flowchart showing the method for generating analysis data. First, as shown in Fig. 5, the profile generating device 200 acquires measurement data of the reference solution by irradiating it with X-rays (step S1). Similarly, measurement data of the sample solution is also acquired (step S2).

[0088] Based on the measurement data obtained for each of the standard solution and the sample solution, sampling is performed to generate profiles, and a profile to be used for analysis is determined based on the obtained profiles (step S3), completing the series of processes. In this process, the optimal profile is determined from multiple profiles by testing each parameter. The profile to be compared may be a profile obtained by subtracting the profile of the standard solution from the profile of the sample solution after adjusting the scale. The details of generating and determining the profiles shown in step S3 will be described later.

[0089] [Profile Generation and Determination Method] Fig. 6 is a flowchart showing the profile generation and determination method (step S3 in the flowchart shown in Fig. 5 above). As shown in Fig. 6, first, the profile generation device 200 accepts input of measurement conditions including sample information and the purpose of the measurement (step T101). Then, based on the input sample information, it determines a function that represents the relationship between the wavenumber of the scattering vector and the interval between sampling positions (step T102). At this time, the function form is determined according to the sample information, but the parameters (coefficients) are not. Note that the sample information includes the type of sample (measurement target).

[0090] Next, the parameters of the function are set (step T103), measurement data is sampled at sampling positions determined by the function (step T104), and a profile is generated (step T105).

[0091] The generated profile is phase-shifted by one of multiple shift amounts (a shift of the sampling position by a single bin or less) (step T106). A first index c1 representing the data variability across the entire wavenumber q of the scattering vector is calculated for the shifted profile (step T107). It is determined whether all position shifts have been completed (step T108). If not, the process returns to step T106. If completed, the profile that has undergone position shifting by the smallest amount of first index c1 is identified from all the shifted profiles (step T109).

[0092] The minimum wavenumber qmin having a significant standard deviation for the identified profile is calculated (step T110), and a first index d1 representing the data variability for the wavenumber q of each scattering vector is calculated (step T111). Then, it is determined whether or not the generation of all profiles has been completed (step T112). Note that the number of profiles to be generated is determined because the ranges for parameter values ​​are determined for multiple parameter sets, and the loop process is continued until the maximum number of profiles is generated.

[0093] If the generation of profiles for each parameter set has not been completed, the process returns to step T3, where different parameters are set to generate profiles and calculate the minimum wave number qmin and the first index d1.

[0094] When the generation of all profiles has been completed, the parameters for which the minimum wavenumber qmin is the minimum value and the first index d1 is the smallest are selected from the multiple profiles obtained, and the profile is determined as the analysis data (step T113), thereby completing the series of processes. Note that, although the phase shift is adjusted each time a profile is generated in the above example, the phase shift may also be adjusted for the determined profile after the optimal profile is determined. Details of each process are described below.

[0095] [Acquisition of Measurement Data] Fig. 7 is a schematic diagram showing the acquisition of measurement data. As shown in Fig. 7, scattered images detected at predetermined time intervals t are acquired as measurement data obtained by irradiating a target solution with X-rays. The accumulated n pieces of measurement data are integrated. When the reference solution and the sample solution are alternately irradiated with X-rays, the respective measurement data are accumulated to obtain integrated data.

[0096] [Setting of Sampling Positions] For measurement data showing a scattering image, sampling positions are set using a specific relationship. The specific relationship is a relationship in which the sampling positions are determined with respect to the wave number of the scattering vector. FIG. 8 is a graph showing an example of sampling positions. As shown in FIG. 8, the specific relationship can be expressed by a function in which the interval between sampling positions is determined with respect to the wave number q of the scattering vector. This function is an increasing function, and the interval y also increases as the wave number q of the scattering vector increases.

[0097] As a specific example, the following function using "qmin of first region" and "qspace factor" shown in FIG. 12 described later can be given.

[0098] The function is not limited to this, and any function in which the interval y between sampling positions increases as q increases can be used. The function form is determined based on the sample to be measured and the purpose of the measurement, and the function is further determined by parameters (coefficients). Multiple settings can be generated by changing the parameters (coefficients), and the optimal setting can be determined from among them by using an evaluation function.

[0099] [Example of Evaluation Function] As the evaluation function, for example, an error function expressed by the following equation (2) can be used.

[0100] The error function E(σ-based) is expressed as the sum of squares of the second derivatives of each data point interval, the sum of the σ outliers before and after each data point, the absolute value of the maximum slope of each interval point, the average value of σ per data point, and the reciprocal of the number of bins in the first region (Nb).

[0101] The error function E (σ-based) has a total of five evaluation items. The power of the five terms constitutes the error function, and if any of the terms increases, the overall value increases and the evaluation decreases. As shown above, all terms may be used, or only any of the terms may be used.

[0102] (Item 1) Sum of squares of second derivatives for each data point interval. The more jagged the generated profile shape, the larger the second derivative value for each point. The larger the sum of squares of second derivatives over the entire profile, the greater the jaggedness of the profile.

[0103] (Item 2) Sum of σ outliers before and after each data point If sampling is done forcefully, σ will become very large in areas where it is statistically difficult to obtain data. Therefore, the extremely large whiskers that appear on the profile are due to the σ outliers before and after each data point. For example, if 1σ is set as the cutoff value for the data, the sum of all σ values ​​above 1σ is the sum of the σ outliers before and after each data point, which indicates the statistical difficulty of obtaining data.

[0104] (Item 3) Absolute value of the maximum slope at each section point The absolute value of the maximum slope at each section point indicates abrupt changes in the profile that cannot be detected by the second derivative alone. The absolute value of the maximum slope at each section point becomes large at points where the value changes drastically, and indicates, for example, whether there is a point where the value is large only in a flat straight line portion.

[0105] (Item 4) Mean σ per data point The mean σ per data point indicates how many single outliers are mixed into the overall curve.

[0106] (Item 5) Reciprocal of the number of bin divisions The error function E(σbased) in equation (2) is the product of the above four items divided by the number of bin divisions, and becomes smaller as the number of bin divisions increases. Therefore, if there is no problem when judging the profile as a whole, the error function E(σbased) will be a small value.

[0107] (Comparison of Each Item) Of the above items, items 1, 3, and 4 are basically determined empirically depending on the type of sample (target substance) such as virus, antibody, etc. Items 2 and 5 can vary slightly depending on the sample, not the type of sample.

[0108] The following equation (3) can also be used as the error function.

[0109] In the error function E (value-based) of formula (3), the sum of the outliers of the intensity data before and after each data point is multiplied instead of the sum of the outliers of σ before and after each data point in item 2 of formula (2). For example, a certain value of the data is determined as a cutoff value, and the sum of all the intensity data values ​​exceeding this value corresponds to the sum of the outliers of the intensity data before and after each data point.

[0110] [Sampling] A profile is generated by sampling at a plurality of different settings. FIG. 9 is a schematic diagram showing an example of sampling. Sampling is performed at set sampling positions b0 to bm while discretely moving the sampling axis a in the direction of the circumferential angle φ. At each sampling position b0 to bm, intensity values ​​are extracted from the pixels that overlap that position, and a profile for each φ is generated using the extracted intensity. A primary profile can then be generated by integrating the profiles for each φ over one circumference in the φ direction.

[0111] In this case, the intensity value of the extracted pixel is proportionally distributed according to the amount of overlap between the pixel and the axis along the wave number direction of the scattering vector. Fig. 10 is a schematic diagram showing an example of proportional distribution of intensity.

[0112] 10, the sampling axes specified by the circumferential angles φ1, φ2, and φ3 overlap pixel Px1 in a ratio of 1:5:2. In this case, 1 / 8, 5 / 8, or 2 / 8 of the intensity of pixel Px1 is apportioned depending on the length of overlap between pixel Px1 and the sampling axes of circumferential angles φ1, φ2, and φ3. Sampling adjusted by such intensity apportionment can reduce the S / N ratio.

[0113] [Parameter Evaluation] Fig. 11 is a schematic diagram showing the integration of profiles. The variance of the generated profile is calculated as a first index, and the most appropriate sampling position is determined based on these values. When evaluating the parameters, at least one of the minimum wavenumber and the first index is used.

[0114] The minimum wavenumber qmin is the wavenumber of the smallest scattering vector with a significant standard deviation for the profile. "Significant" means, for example, that the standard deviation is smaller than the intensity value. The minimum wavenumber qmin is calculated based on each profile generated by changing the parameters of the function, and a search is made for a minimum qmin. Among the profiles where qmin has a minimum value, a more optimal profile can be identified using the first index. This allows profiles with improved resolution to be used for analysis.

[0115] The first index can be an index indicating the overall intensity variation in the generated profile. An example of such an index is Σσ prof. Alternatively, the first index can be an index indicating the occurrence of outliers exceeding the intensity variation in the generated profile. An example of such an index is σ of σ profile. When using Σσ prof and σ of σ profile, it is preferable that both indices are small. However, if only one of them is small, it is preferable to prioritize the small σ of σ profile. By utilizing the settings determined by such parameter evaluation, it is possible to obtain a more appropriate profile that determines the optimal shift amount for beat pattern suppression, as exemplified below.

[0116] [Phase Shift (Optional)] Furthermore, as a phase shift, the sampling positions may be shifted to search for optimal sampling positions in order to reduce beat patterns. For the profile determined as described above, a shift amount is set to shift the sampling positions b0 to bm in the direction of the wave number q of the scattering vector. It is preferable that the shift amount is sufficiently smaller than the pixel size. For example, the position of the entire bin may be shifted multiple times in units of 1 / 100 of the pixel size, and conditions under which beat patterns do not occur in the profile generated by sampling may be searched for.

[0117] Phase shifting is particularly effective for upsampling, which is prone to producing beat patterns. It is possible to search for the optimal value of the shift amount to improve resolution and suppress moire. The maximum phase shift amount is the interval between sampling positions. The shift amount is the value obtained by multiplying the interval between sampling positions by a phase shift value between 0 and n-1. The error function E for each shift amount can be calculated, and the shift amount that minimizes the error function E can be identified. Note that beat patterns can also be reduced by simply adjusting the interval between sampling positions, so phase shifting is not necessarily required. Beat pattern reduction by phase shifting is not a required process, and users may omit it depending on their purpose.

[0118] By performing the above operations, a profile can be generated under optimal sampling conditions. Then, by subtracting the profile of the reference solution from the profile of the sample solution, data for analysis can be obtained.

[0119] [Second Embodiment] In the above embodiment, parameters that satisfy the criteria are identified using only the first index representing the data variability, but a second index indicating the degree of deviation between each profile generated by non-uniform sampling and a profile generated by uniform sampling may also be used. In this case, the sampling width at low scattering angles is basically determined by the first index, and the sampling width at high scattering angles is determined by the second index. The sampling width is determined by determining the parameters.

[0120] As the second index, the following evaluation function J(q) can be used.

[0121] In the above formula, proc indicates that the non-uniform sampling process of the present invention has been performed. The evaluation function J(q) is obtained by subtracting the standard deviation due to non-uniform sampling from the difference for each pixel between the profile obtained by non-uniform sampling and the profile obtained by uniform sampling, and adding the resultant value over the entire wave number q of the scattering vector. Using the evaluation function J(q), it is possible to determine whether the wave number region of the scattering vector on the high-angle side has been overly smoothed by binning.

[0122] The evaluation function J(q) may be the following formula:

[0123] [Third Embodiment] In the above-described embodiment, the sampling positions are set by determining a function as a specific relationship that represents the interval between the sampling positions and the wave number of the scattering vector. However, the sampling positions may also be set in a relationship that provides a discrete, constant sampling width for each wave number range of the scattering vector. Here, the scattering profile to be analyzed is divided into several wave number regions, and the divided wave number regions are called areas. Furthermore, the region on the lowest wave number side is called the first area.

[0124] In this case, the parameters that can be set include the type of sample (target substance), the number of regions, the number of bins or divisions in the first region, and the width of the first region (q max of first region), area width increment factor (q space factor), and division number increment factor (Divisor factor).

[0125] The number of areas is determined by the characteristics of the target substance. The number of bins in the initial area or the number of 22 divisions depends on the instrument conditions and sample conditions (concentration, structure, solution state) and is determined by an evaluation function. The width of the initial area and the area width increment factor are each determined by the characteristics of the target substance.

[0126] In this embodiment, the sampling positions are determined and profiles are generated and determined as follows: First, the division positions of the first area and the number of divisions are determined. This determines the width of the first bin. Next, the division positions of the area are determined. This determines the total number of divisions.

[0127] Next, the target material is input, and some parameters are set according to the input target material. That is, the number of areas, the width of the first area, and the area width increment factor are set. The error function is determined, and the range of q for determining the number of bins or division number of the first area is determined. Then, the area width increment factor is automatically set, and a profile is generated and output by data conversion. Furthermore, the beat pattern may be reduced by adjusting the shift amount. A plot of the error function may also be output. In this case, for example, the above formula (4) or (5) may be used as the error function.

[0128] In this case, it is preferable that the specific relationship be adjusted automatically. However, it may also be displayed on the screen so that the operator can adjust it. Fig. 12 shows an example of a settings screen for adjusting the relationship between the wavenumber of the scattering vector and the interval between sampling positions. In the example shown in Fig. 12, the minimum and maximum wavenumbers of the scattering vector and the bin width of each region are displayed, and the number of areas, the number of bins or divisions of the first area, the width of the first area, the increment factor for the area width, and the increment factor for the division number are displayed so that they can be adjusted. In addition, the wavenumber range of the scattering vector of the profile to be saved is also displayed so that it can be adjusted.

[0129] In the example shown in Fig. 12, "Number of bins first region" determines the width of the first bin, and if "Number of regions" is set to 10, the bins are divided into 10 on the upsampling side. On the other hand, since "q space factor" is 1.474, the sampling width is expanded by 1.474 times on the high-angle side.

[0130] [Fourth embodiment] When a differential profile obtained by subtracting a profile of a reference solution from a profile of a sample solution is to be evaluated, the profile may be evaluated using Rg and its standard deviation obtained by a Guinier plot instead of directly evaluating the profile using the first index. By plotting these for each parameter set, it is possible to determine the optimal parameters.

[0131] 13 is a flowchart showing a method for generating and determining a profile. As shown in FIG. 13, the profile generating device 200 first accepts input of measurement conditions, including sample information and the purpose of the measurement (step T201). Then, based on the input sample information, a function is determined that represents the relationship between the wavenumber of the scattering vector and the interval between sampling positions (step T202). At this time, the function form is determined according to the sample information, but the parameters (coefficients) are not. The sample information includes the type of sample (measurement target).

[0132] Next, parameters for the scattering angle on the low side of the function are set (step T203), measurement data is sampled at sampling positions determined by the function (step T204), and a profile is generated (step T205).

[0133] A Guinier plot is performed on the generated profile (step T206). It is determined whether or not the generation of all profiles has been completed (step T207). Note that the number of profiles to be generated is predetermined, and the loop process is repeated until the maximum number of profiles is reached.

[0134] If the generation of profiles for each parameter set at the low scattering angle has not been completed, the process returns to step T203. Then, different parameters are set, profiles are generated, and a Guinier plot is performed. The Guinier plot determines the radius of gyration Rg and its standard deviation.

[0135] When the generation of all profiles has been completed, parameters with a small standard deviation and high uniformity of the radius of gyration Rg are selected and determined (step T208). Next, it is determined whether the type of sample is known (step T209). If the type of sample is known, the parameters for the high scattering angles can be determined according to the type of sample. If parameter data for a certain sample has been accumulated and optimal parameters can be identified in advance, the parameters for the high scattering angles can be easily determined by setting them.

[0136] On the other hand, if the type of sample is unknown, parameters for the high scattering angle side are set (step T210). Measurement data is sampled at sampling positions determined by a function of the set parameters (step T211), and a profile is generated (step T212). A second index is calculated for the generated profile (step T213), and it is determined whether all profiles have been generated (step T214). As with the calculation of the first index, the number of profiles to be generated is predetermined, and the loop process is repeated until the maximum number of profiles is reached.

[0137] If the generation of profiles for each parameter set at the low scattering angle has not been completed, the process returns to step T210. Then, different parameters are set, profiles are generated, and a Guinier plot is performed. The second index calculated for each parameter set is evaluated, and the sampling width at the high scattering angle is determined (step T215), and the process ends. The sampling width is determined by the parameter determination.

[0138] In this case, the amount of shift may be further adjusted to reduce the beat pattern. In this way, profiles can be generated with an appropriate sampling width at both high and low scattering angles.

[0139] [Example] X-rays were irradiated onto a solution containing a biopolymer, and small-angle scattered X-rays were detected. The obtained measurement data was sampled by changing the sampling position settings, and a profile was generated. For each setting, the value of the error function corresponding to the first index was plotted against the number of pixel size divisions, which corresponds to the reciprocal of the interval between the sampling positions.

[0140] 14 is a graph showing the relationship between the number of divisions and the evaluation value of the error function. As shown in Fig. 14, the value of the error function is smallest when the number of divisions is 56, and in this evaluation, the setting of the sampling positions when the number of divisions is 56 was evaluated as optimal.

[0141] FIG. 15 is a graph showing profiles obtained by different numbers of divisions. In FIG. 15, each profile is intentionally shifted in the q direction for easier viewing. The measurement data used in the example shown in FIG. 15 originally had 763 pixels per side. Sampling positions were set at intervals of 1838, 1839, 1840, 1841, 1842, and 1843 for this measurement data, and sampling was performed to generate profiles P1, P2, P3, P4, P5, and P6. When σ of σ profile and Σσ prof were calculated as second indices for each of these profiles, both were found to be small for profile P2.

[0142] As described above, suitable splitting conditions were identified for the measurement data of the sample solution and the measurement data of the reference solution. Then, sampling was performed under those splitting conditions to generate adjusted profiles. These profiles were then subtracted to generate data for analysis. For comparison, unadjusted profiles and data for analysis were also generated and compared.

[0143] 16(a) and 16(b) are graphs showing unadjusted and adjusted profiles, respectively. The analysis data Su1 shown in FIG. 16(a) is data obtained by subtracting the measurement data Re1 of the reference solution from the measurement data Sa1 of the sample solution. The analysis data Su2 shown in FIG. 16(b) is data obtained by subtracting the measurement data Re2 of the reference solution from the measurement data Sa2 of the sample solution. It can be seen that the analysis data shown in FIG. 16(b) has smaller data variance at higher angles than the analysis data shown in FIG. 16(a). This demonstrates that highly accurate analysis data can be obtained by adjusting the sampling conditions.

[0144] 10 Profile generation system 100 X-ray analysis device 110 X-ray generation unit 111 X-ray source 115 Optical system 117 Kratsky block 120 Sample loading mechanism 125 Sample holding tube 130 Pixel detector 140 Control unit 200 Profile generation device 210 Computer 211 Input / output control unit 215 Measurement control unit 217 Measurement data storage unit 223 Sampling position setting unit 225 Data extraction unit 227 Profile generation unit 232 First index calculation unit 236 Minimum wavenumber calculation unit 238 Parameter determination unit 241 Position shift unit 242 Guinier plot execution unit 244 Second index calculation unit 247 Shift amount determination unit 251 Analysis data generation unit 280 Input device 290 Output device L Control bus a Sampling axis b0 to bm Sampling position P1 to P6 Profile S0 Sample qmin Minimum wave number Px1 Pixel

Claims

1. A profile generation device that generates an X-ray scattering profile by sampling X-ray scattering intensity data obtained by a pixel array detector, comprising: a sampling position setting unit that sets multiple sampling positions that are not regularly spaced in the wave number direction of the scattering vector based on a specific relationship; a data extraction unit that extracts intensity data at pixels that overlap with the set sampling positions; and a profile generation unit that generates a plotted profile based on the extracted intensity data.

2. The profile generating device according to claim 1, wherein said sampling position setting section sets sampling positions in which the sampling interval on the high angle side is greater than the sampling interval on the low angle side.

3. The profile generating device according to claim 2, wherein said sampling position setting section sets sampling positions at a sampling interval smaller than the pixel size on the low angle side.

4. The profile generating device according to claim 2, wherein said sampling position setting section sets sampling positions at a sampling interval greater than the pixel size on the high-angle side.

5. A profile generating device according to claim 3, wherein said sampling position setting section sets sampling positions at a sampling interval greater than the pixel size on the high-angle side.

6. A profile generating device according to any one of claims 1 to 5, characterized in that the sampling position setting unit sets the sampling positions using a function that gives a sampling interval for the wave number of the scattering vector as the specific relationship.

7. A profile generating device as described in any one of claims 1 to 5, characterized in that the sampling position setting unit sets the sampling positions in a relationship that gives a discrete, constant sampling width for each wave number range of the scattering vector as the specific relationship.

8. A profile generation device as described in any one of claims 1 to 5, further comprising: a first index calculation unit that calculates a first index representing the variability of data from each profile generated by changing a parameter in the specific relationship; and a parameter determination unit that determines the parameter for which the first index satisfies a criterion.

9. The profile generating device according to claim 8, wherein the first index represents the irregularity of the generated profile using at least one of the standard deviation and the slope of the distribution of the generated profile.

10. The profile generating device according to claim 8, wherein the first index represents the occurrence of outliers in the generated profile using the standard deviation of the generated profile.

11. The profile generating device according to claim 8, further comprising: a position shifting unit that shifts the plurality of sampling positions that have already been set by a specific shift amount in the wave number direction of the scattering vector; and a shift amount determining unit that determines the specific shift amount at which the first index satisfies a criterion.

12. A profile generation device according to any one of claims 1 to 5, further comprising: a Guinier plot execution unit that executes a Guinier plot for each profile generated by changing a parameter in the specific relationship; and a parameter determination unit that determines the parameter with the highest uniformity of the radius of gyration Rg obtained by the Guinier plot.

13. A profile generating device as described in claim 12, characterized in that the parameter determination unit determines a parameter that specifies the sampling width in the low-angle scattering angle region based on the uniformity of the radius of gyration Rg, and determines a parameter that specifies the sampling width in the high-angle scattering angle region based on the type of sample.

14. A profile generation device as described in any one of claims 1 to 5, further comprising: a second index calculation unit that calculates a second index indicating the degree of deviation due to processing of a profile generated by changing a parameter in the specific relationship; and a parameter determination unit that determines the parameter that is suitable for a criterion using the second index.

15. A profile generation device according to any one of claims 1 to 5, further comprising: a minimum wave number calculation unit that calculates the wave number of the smallest scattering vector having a significant standard deviation for a profile obtained based on each of the profiles generated by changing the parameters in the specific relationship; and a parameter determination unit that determines the parameter that is suitable for a criterion using the wave number of the calculated smallest scattering vector.

16. A profile generation method for generating an X-ray scattering profile by sampling X-ray scattering intensity data obtained by a pixel array detector, comprising the steps of: setting a plurality of sampling positions not spaced at regular intervals in the wave number direction of the scattering vector based on a specific relationship; extracting intensity data at pixels that overlap with the set sampling positions; and generating a plotted profile based on the extracted intensity data.

17. A profile generation program for generating an X-ray scattering profile by sampling X-ray scattering intensity data obtained by a pixel array detector, the profile generation program causing a computer to execute the following processes: setting multiple sampling positions at irregular intervals in the wave number direction of the scattering vector based on a specific relationship; extracting intensity data at pixels that overlap with the set sampling positions; and generating a plotted profile based on the extracted intensity data.

Images