WO2018188395A1 - 一种基于拉曼光谱测量的细胞培养液质量检测方法 - Google Patents

一种基于拉曼光谱测量的细胞培养液质量检测方法 Download PDF

Info

Publication number
WO2018188395A1
WO2018188395A1 PCT/CN2018/072393 CN2018072393W WO2018188395A1 WO 2018188395 A1 WO2018188395 A1 WO 2018188395A1 CN 2018072393 W CN2018072393 W CN 2018072393W WO 2018188395 A1 WO2018188395 A1 WO 2018188395A1
Authority
WO
WIPO (PCT)
Prior art keywords
cell culture
signal
culture fluid
raman
raman spectral
Prior art date
Application number
PCT/CN2018/072393
Other languages
English (en)
French (fr)
Inventor
赵一雷
梁波
宣黎明
孔令印
刘国宁
Original Assignee
苏州贝康医疗器械有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 苏州贝康医疗器械有限公司 filed Critical 苏州贝康医疗器械有限公司
Priority to US16/499,941 priority Critical patent/US11561182B2/en
Priority to EP18784206.7A priority patent/EP3611495A4/en
Publication of WO2018188395A1 publication Critical patent/WO2018188395A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/65—Raman scattering
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/01—Arrangements or apparatus for facilitating the optical investigation
    • G01N21/03—Cuvette constructions
    • G01N21/0303—Optical path conditioning in cuvettes, e.g. windows; adapted optical elements or systems; path modifying or adjustment
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/65—Raman scattering
    • G01N2021/651—Cuvettes therefore
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2201/00—Features of devices classified in G01N21/00
    • G01N2201/06—Illumination; Optics
    • G01N2201/061—Sources
    • G01N2201/06113—Coherent sources; lasers
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2201/00—Features of devices classified in G01N21/00
    • G01N2201/06—Illumination; Optics
    • G01N2201/063—Illuminating optical parts
    • G01N2201/0636—Reflectors

Definitions

  • the invention belongs to the field of cell culture liquid detection, and particularly relates to a cell culture liquid quality detecting method based on Raman spectroscopy measurement.
  • Raman spectroscopy combined with chemometric methods can be used to develop a rapid, efficient and non-invasive assay for evaluating cell cultures of complex components to determine cell growth status.
  • Raman spectroscopy has significant advantages in the detection of multi-component aqueous solutions because complex sample preparation is not required and water has no response signal.
  • Spectral technology is easy to operate and low in cost. It has been widely used in the medical field and has breakthroughs in application concepts and technological innovation.
  • this paper analyzes the data by using the support vector machine method to detect the Raman spectra of the same type of cells under the same culture conditions, and establishes the perfect data acquisition, preprocessing and data.
  • the modeling process enables efficient and non-invasive cell growth activity identification, which can be further promoted for clinical applications.
  • the present invention aims to provide a cell culture liquid quality detecting method based on Raman spectroscopy with high accuracy, simple detection process and low cost.
  • a method for detecting cell culture liquid quality based on Raman spectroscopy comprising the following steps:
  • Raman spectroscopy technique is used to measure the original Raman spectral signal of the metabolite in the cell culture solution obtained in the step (1); if the original Raman spectral signal is judged to be qualified, the process proceeds to step (3). , otherwise, the cell culture solution is repeatedly subjected to Raman spectroscopy measurement;
  • the cell culture solution taken in the step (1) is 7 ⁇ l.
  • the 7 microliters of the cell culture medium was placed in a micro-solution detection cell.
  • the micro-solution detection pool includes a hemispherical mirror, or a mirror set composed of a hemispherical mirror and a cylindrical mirror, the hemispherical mirror and the mirror group each having a highly reflective mirror surface.
  • the mirror plated dielectric film or an inert metal film is formed by the same method as that described above.
  • the period of time is 3-4 days.
  • the step (2) determines whether the original Raman spectral signal is qualified as follows: the absolute peak intensity CV ⁇ 5% of the scattering peak of the 0.5% ethanol solution at 880 cm -1 and the fluctuation range of the Raman displacement ⁇ 2 cm -1 .
  • step of performing data signal processing on the acquired original Raman spectrum signal in the step (3) includes:
  • the best function matching achieves data correction by minimizing the sum of the squares of the distances between the sample points and the fitted curve
  • the signal from which the background of the fluorescent signal is removed is homogenized to obtain an analyzable signal.
  • the present invention detects the difference signal in the cell culture fluid by Raman spectroscopy, detects the quality of the cell culture fluid, and achieves the purpose of non-invasively evaluating the cell growth state, thereby having important value in various fields, for example, In clinical applications, it can be extended to non-invasive detection of embryo quality.
  • real-time online detection of cell growth and protein expression status can be performed.
  • the invention is convenient, effective, low-cost, suitable for large-scale promotion, and is not subject to geographical restrictions and lack of professional personnel, and can be industrialized and streamlined on a large scale.
  • Figure 1 is a flow chart of the method of the present invention
  • FIG. 2 is a schematic view of a micro solution detection tank in the present invention
  • FIG. 3 is a schematic view of another micro-solution detection tank in the present invention.
  • Figure 5 is a diagram showing the relationship between K value and frequency in the embodiment of the present invention.
  • FIG. 8 is a schematic diagram of 100 points where the difference in P values between LIST1 and LIST2 is the largest in the embodiment of the present invention.
  • a method for detecting cell culture liquid quality based on Raman spectroscopy comprises the following steps:
  • Raman spectroscopy technique is used to measure the original Raman spectral signal of the metabolite in the cell culture solution obtained in the step (1); if the original Raman spectral signal is judged to be qualified, the process proceeds to step (3). , otherwise, the cell culture solution is repeatedly subjected to Raman spectroscopy measurement;
  • the process of judging whether the original Raman spectral signal is qualified is: 0.5% ethanol solution at 880 cm -1
  • the best function matching achieves data correction by minimizing the sum of the squares of the distances between the sample points and the fitted curve
  • the signal from which the background of the fluorescent signal is removed is homogenized to obtain an analyzable signal.
  • the micro-solution detection pool is composed of a hemispherical mirror and a cylindrical mirror, or is composed of a hemispherical mirror, and the metal aluminum or copper is processed by a diamond car to obtain high reflection.
  • the mirror surface has a high reflection of both the probe light and the Raman scattered light; to prevent oxidation or corrosion of the mirror-coated dielectric film or inert metal film.
  • the Raman test uses a 785 nm laser and a micro-solution detection cell consisting of a hemispherical mirror and a cylindrical mirror. The test cell is placed in a predetermined fixture such that the center of the cylindrical mirror is concentric with the laser beam and the focus of the laser beam coincides with the center of the hemispherical mirror.
  • Raman spectroscopy technique is used to measure the original Raman spectral signal of the metabolite in the cell culture solution obtained in the step (1); if the original Raman spectral signal is judged to be qualified, the process proceeds to step (3). Otherwise, the cell culture solution was repeatedly subjected to Raman spectroscopy measurement. Due to the nature of the spectral signal, multiple measurements are required for the same sample to improve the detection accuracy of the spectral signal. In this embodiment, the number of samples for data modeling and testing is as shown in Table 1.
  • the theoretical standard deviation is a constant ⁇ .
  • the S value approaches ⁇ instead of approaching zero.
  • Appropriate increase in the number of measurements can improve the precision of the arithmetic mean and also facilitate the discovery of large errors.
  • the standard deviation of the average decreases slowly with the increase of the number of measurements.
  • S(V i ) is the experimental standard deviation
  • n is the number of repeated measurements
  • V is the arithmetic mean of n measurements.
  • the difference curve in FIG. 4 indicates the SD obtained by random sampling minus the SD value of all samples of LIST2;
  • the difference trend curve indicates the difference between adjacent two sampling points after increasing the sample size. The value can reflect that as the sample size increases, the sampled SD becomes closer to the SD value of the real sample. It can be seen from Fig. 4 that after the ninth sampling, the range of variation of the difference trend curve is more and more stable, and the SD value is getting closer and closer to the true value, so 9 can be selected as the minimum number of measurements.
  • Step (3) Processing of Raman spectral signal: Data signal processing is performed on the qualified original Raman spectral signal obtained in step (2) to obtain an analyzable signal. Since the cell culture medium has a volume of only 7 ⁇ L and there is strong background interference in the signal (about 99.9% of the signal comes from the background of the culture medium), it is expected that the fluctuation signal caused by the cell metabolism to the culture solution is less than 0.1%.
  • the background signal cancellation algorithm is used to achieve the purpose of reducing the noise signal and enhancing the target signal.
  • Data signal processing is required for the acquired raw Raman signal, and the steps include: 1) data correction; 2) removal of the fluorescent signal background; 3) homogenization.
  • the best function matching achieves data correction by minimizing the sum of the squares of the distances between the sample points and the fitted curve
  • the peak position is determined accurately by the continuous wavelet pattern matching method of the Mexican hat wavelet as the parent function. Then, the continuous wavelet derivation method with Haar wavelet as the parent function is used to determine the starting position of the peak. Finally, the penalty least squares method is used. Fit a smooth adjustable background;
  • the signal from which the background of the fluorescent signal is removed is homogenized to obtain an analyzable signal.
  • the difference between the data in List1 and the data in List2 is counted (P ⁇ 0.05).
  • the optimal parameter K we need to calculate the frequency of all abnormal points under each parameter K (Formula 1), so that we The frequency of each parameter K below LIST1 is obtained, as shown in Table 2.
  • Equation 1 illustrates:
  • the processed signal can be obtained by using the background signal elimination algorithm, and the signal comparison before and after processing is as shown in FIG. 6.
  • the processed LIST1 and LIST2 signals are compared by a two-sample U test to find the 100 data points with the largest difference (or data points with a p-value ⁇ 0.05), which is required for subsequent SVM classification. Difference point.
  • SVM Support Vector Machine
  • SVM Normal cell culture fluid
  • SVM Abnormal cell culture fluid
  • SVM Normal cell culture fluid (original) 72 28 Abnormal cell culture fluid (original) 16 84
  • the invention adopts the collection of the cell culture liquid; the acquisition, processing and analysis of the Raman spectral signal; the Raman spectroscopy technique is used to measure the original Raman spectral signal of the metabolite in the cell culture solution; whether the original Raman spectral signal is qualified or not is qualified.
  • the original Raman spectral signal is processed by the data signal to obtain an analyzable signal; the difference signal is analyzed by differential statistical analysis to obtain the difference signal, and the difference signal is used for modeling, and the difference signal is classified by the support vector machine to normal and abnormal.
  • the cell culture fluid spectral signals are differentiated to obtain cell culture fluid quality results.
  • the invention detects the difference signal in the cell culture liquid by Raman spectroscopy, detects the quality of the cell culture liquid, thereby achieving the purpose of non-invasive evaluation of the cell growth state, and the invention is convenient, effective, low-cost, and can be industrialized and streamlined in a large scale.

Landscapes

  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)

Abstract

一种基于拉曼光谱测量的细胞培养液质量检测方法，包括如下步骤：细胞培养液的采集；拉曼光谱信号的采集、处理和分析；使用拉曼光谱技术测量细胞培养液中代谢产物的原始拉曼光谱信号；判断原始拉曼光谱信号是否合格，对合格的原始拉曼光谱信号进行数据信号处理，得到可分析信号；再对可分析信号进行差异统计分析得到差异信号，利用差异信号建模，并利用支持向量机对差异信号进行分类，对正常和异常的细胞培养液光谱信号进行区分，得到细胞培养液质量结果。通过拉曼光谱检测细胞培养液中差异信号，检测出细胞培养液质量，从而达到无创评价细胞生长状态的目的，且该方法便捷有效、低成本，可大规模产业化、流程化。

Description

一种基于拉曼光谱测量的细胞培养液质量检测方法 技术领域
本发明属于细胞培养液检测领域，尤其涉及一种基于拉曼光谱测量的细胞培养液质量检测方法。
背景技术
使用拉曼光谱技术结合化学计量学方法可开发出一种快速、高效且无创的检测方法，用于评价复杂组分的细胞培养液，从而判断细胞生长状态。由于不需要进行复杂的样本制备，且水没有响应信号，拉曼光谱技术在对多组分水溶液检测上具有显著的优势。光谱技术操作方便，成本低廉，已广泛应用于医学领域，在应用理念和技术创新方面具有突破性。
在实际应用中，迫切需要一种能够快速、准确且成本较低的细胞活性鉴定技术。低分子量代谢物作为细胞调控进程中的最终产物，可以揭示生物系统对营养及环境因素改变的应答，更快速地反映细胞活力。因此，可以通过测量细胞培养液中的代谢产物变化和培养基消耗偏好来对细胞生长质量进行评估。
对于细胞培养液影响细胞发育的机理方面，早期有较多研究集中在分析某几种明确的代谢物，以作为生物标记物来表征细胞发育潜力，但由于代谢物水平存在时序性和多样性，目前并无明确的生物标记物可以在所有的培养条件和培养过程中适用，哪些成分对细胞生长活性起核心作用也没有定论。近期有研究关注于分析整个代谢指纹图谱，即对特定生理阶段或发育期细胞中所有低分子量化合物(相对分子量<1000)的动态定量分析。
当前使用拉曼光谱法检测细胞培养液的研究尚处于发展阶段。现有技术中有使用AVALON拉曼光谱仪检测了五种化学成分清晰的混合物溶液，如多种氨基酸、多种有机酸或无机酸等，使用主成分分析法和独立软模式类簇法进行数据分析比较，所建立的分析模型可准确的进行培养基质量鉴定。还有使用拉曼光谱仪检测多种CHO细胞培养液，使用最小二乘法进行数据分析，建立了拉曼光谱法在线无创实时测定培养液中葡萄糖和乳酸含量的方法。本文在已有研究的基础上，通过检测同一类型细胞在同等培养条件下的培养液的拉曼光谱，使用支持向量机法对数据进行分类建模，建立了完善的数据采集、预处理及数据建模流程，实现了高 效的且无创的细胞生长活性鉴定，可进一步推广做临床应用。
发明内容
发明目的：本发明旨在提供一种准确率高、检测流程简便、成本低廉的基于拉曼光谱测量的细胞培养液质量检测方法。
技术方案：一种基于拉曼光谱测量的细胞培养液质量检测方法，包括如下步骤：
(1)细胞培养液的采集：取经细胞培养一段时间的细胞培养液；
(2)拉曼光谱信号的采集：使用拉曼光谱技术测量步骤(1)所得细胞培养液中代谢产物的原始拉曼光谱信号；判断原始拉曼光谱信号是否合格，是则进入步骤(3)，否则，重复对该细胞培养液进行拉曼光谱技术测量；
(3)拉曼光谱信号的处理：对步骤(2)获取的合格原始拉曼光谱信号进行数据信号处理，得到可分析信号；
(4)拉曼光谱信号的分析：对步骤(3)得到的可分析信号进行差异统计分析得到差异信号，利用差异信号建模，并利用支持向量机对差异信号进行分类，对正常和异常的细胞培养液光谱信号进行区分，得到细胞培养液质量结果。
进一步的，所述步骤(1)中所取的细胞培养液为7微升。
进一步的，所述7微升细胞培养液放置在微量溶液检测池中。
进一步的，所述微量溶液检测池包括半球面反射镜，或者由半球面反射镜和圆柱面反射镜构成反射镜组，所述半球面反射镜和反射镜组都具有高反射的镜面。
进一步的，所述镜面镀介质膜或惰性金属膜。
进一步的，所述步骤(1)中一段时间为3-4天。
进一步的，所述步骤(2)判断原始拉曼光谱信号是否合格过程为：0.5％的乙醇溶液在880cm -1处散射峰绝对峰强CV≤5％，拉曼位移波动范围≤2cm -1。
进一步的，所述步骤(3)中对获取的原始拉曼光谱信号进行数据信号处理步骤包括：
1)对获取的原始拉曼光谱信号进行数据校正；
利用最小二乘算法，通过最小化误差的平方和寻找数据的最佳函数匹配，最佳函数匹配通过让采样点跟拟合曲线的距离平方和最小实现数据校正；
2)将经数据校正后的信号进行去除荧光信号背景；
a利用墨西哥帽小波为母函数的连续小波模式匹配方法准确确定峰位置；
b.继续利用以Haar小波为母函数的连续小波求导方法确定峰的起始位置；
c.利用惩罚最小二乘法拟合出平滑可调背景；
3)对去除荧光信号背景的信号进行均一化处理
基于Stouffer's Z-score算法对去除荧光信号背景的信号进行均一化处理，得到可分析信号。
有益效果：相对于现有技术，本发明通过拉曼光谱检测细胞培养液中差异信号，检测出细胞培养液质量，从而达到无创评价细胞生长状态的目的，从而在多个领域有重要价值，例如在临床应用中可推广至胚胎质量无创检测，在医药重组蛋白生产时，可进行实时在线检测细胞生长和蛋白表达状态。本发明便捷有效、低成本、适宜大规模推广，且不受地域限制、专业人员缺乏的限制，可大规模产业化、流程化。
附图说明
图1为本发明方法流程图；
图2为本发明中一种微量溶液检测池示意图；
图3为本发明中另一种微量溶液检测池示意图；
图4为本发明实施例中最佳重复测量次数对比图；
图5为本发明实施例中K值与频率关系图；
图6为本发明实施例中信号处理前后对比图；
图7为本发明实施例中LIST1和LIST2的U检验T值分布图；
图8为本发明实施例中LIST1和LIST2的P值差异最大的100个点示意图。
具体实施方式
下面将结合附图，对本发明进行详细的描述：
如图1所示，本发明所述的一种基于拉曼光谱测量的细胞培养液质量检测方法，包括如下步骤：
(1)细胞培养液的采集：取经细胞培养一段时间的细胞培养液；
(2)拉曼光谱信号的采集：使用拉曼光谱技术测量步骤(1)所得细胞培养液中代谢产物的原始拉曼光谱信号；判断原始拉曼光谱信号是否合格，是则进入 步骤(3)，否则，重复对该细胞培养液进行拉曼光谱技术测量；
判断原始拉曼光谱信号是否合格过程为：0.5％的乙醇溶液在880cm -1处散射峰绝对峰强CV≤5％，拉曼位移波动范围≤2cm -1。
(3)拉曼光谱信号的处理：对步骤(2)获取的合格原始拉曼光谱信号进行数据信号处理，得到可分析信号；对获取的原始拉曼光谱信号进行数据信号处理步骤具体为：
1)对获取的原始拉曼光谱信号进行数据校正；
利用最小二乘算法，通过最小化误差的平方和寻找数据的最佳函数匹配，最佳函数匹配通过让采样点跟拟合曲线的距离平方和最小实现数据校正；
2)将经数据校正后的信号进行去除荧光信号背景；
a利用墨西哥帽小波为母函数的连续小波模式匹配方法准确确定峰位置；
b.继续利用以Haar小波为母函数的连续小波求导方法确定峰的起始位置；
c.利用惩罚最小二乘法拟合出平滑可调背景；
3)对去除荧光信号背景的信号进行均一化处理
基于Stouffer's Z-score算法对去除荧光信号背景的信号进行均一化处理，得到可分析信号。
(4)拉曼光谱信号的分析：对步骤(3)得到的可分析信号进行差异统计分析得到差异信号，利用差异信号建模，并利用支持向量机对差异信号进行分类，对正常和异常的细胞培养液光谱信号进行区分，得到细胞培养液质量结果。
下面结合具体实施例详述本发明所述的基于拉曼光谱测量的细胞培养液质量检测方法，其包括如下步骤：
(1)细胞培养液的采集：取经按标准方法培养的细胞培养液。
使用经接种后第3-4天的细胞培养液7微升，使用拉曼光谱技术测定细胞培养液中代谢产物的光谱信号。由于溶液是极微量的，所以该过程必须在为此技术设计的微量溶液检测池中方能进行信号检测。
如图2、3所示，所述微量溶液检测池是由半球面反射镜和圆柱面反射镜构成，或者由半球面反射镜组成，通过采用金刚车对金属铝或铜进行加工，获得高反射的镜面，对探测光和拉曼散射光都有很高的反射；为防止氧化或者腐蚀所述镜面可镀介质膜或惰性金属膜。拉曼检测使用785nm激光和由半球面反射镜和圆 柱面反射镜构成的微量溶液检测池。检测池放入预设好的固定装置中，使圆柱面反射镜中心与激光束同心并使激光束焦点与半球面反射镜圆心重合。
(2)拉曼光谱信号的采集：使用拉曼光谱技术测量步骤(1)所得细胞培养液中代谢产物的原始拉曼光谱信号；判断原始拉曼光谱信号是否合格，是则进入步骤(3)，否则，重复对该细胞培养液进行拉曼光谱技术测量。由于光谱信号的特性，对于同一样本需要进行多次重复测量，以此提高光谱信号的检测精度。本实施例中，数据建模和测试的样本数如表1所示。
表1 数据建模和测试的样本数
Figure PCTCN2018072393-appb-000001
根据贝塞尔公式，理论上标准偏差是一个常量δ，随着测量次数的增多，S值逼近δ，而不是趋近于零。适当增多测量次数，可以提高算数平均值的精密度，也便于发现出大误差。但是测量次数达到一定值后，比如10次以后，平均值的标准差随着测量次数的增加减少的很慢。
贝塞尔公式：
Figure PCTCN2018072393-appb-000002
其中：S(V i)为实验标准差；n为重复测量次数；V为n次测量结果的算术平均值。
根据这一结论，我们从LIST2的文件列表中随机抽取2-30个样本计算其SD1值，并且比较其SD1与全部的LIST2计算的到的SD2之间的差值大小。具体如图4所示，图4中差值曲线：表示随机抽样得到的SD减去LIST2全部样本的SD值；差值变化趋势曲线：表示增加样本量后相邻两个取样点之间的差值，可反映随着样本量的增加，抽样得到的SD越来越趋近与真实样本的SD值。从图4可知在，在第9次抽样后，差值变化趋势曲线的变化范围越来越稳定，其SD值越来越趋近与真实值，所以可以选择9为最少的测量次数。
(3)拉曼光谱信号的处理：对步骤(2)获取的合格原始拉曼光谱信号进行 数据信号处理，得到可分析信号。由于细胞培养液体积只有7微升，信号中有较强的背景干扰(大约99.9％的信号来自于培养液背景)，可以预见由于细胞代谢对培养液引起的涨落信号不到0.1％。利用背景信号消除算法，从而到达降低噪声信号，增强目标信号的目的。对获取的原始拉曼信号需要进行数据信号处理，该步骤包括：1)数据校正；2)去除荧光信号背景；3)均一化。
1)对获取的原始拉曼光谱信号进行数据校正；
利用最小二乘算法，通过最小化误差的平方和寻找数据的最佳函数匹配，最佳函数匹配通过让采样点跟拟合曲线的距离平方和最小实现数据校正；
2)将经数据校正后的信号进行去除荧光信号背景；
首先，利用墨西哥帽小波为母函数的连续小波模式匹配方法准确确定峰位置；然后，继续利用以Haar小波为母函数的连续小波求导方法确定峰的起始位置；最后，利用惩罚最小二乘法拟合出平滑可调背景；
3)对去除荧光信号背景的信号进行均一化处理
为了使多组数据能够进行比较，需要对数据进行均一化，基于Stouffer's Z-score算法对去除荧光信号背景的信号进行均一化处理，得到可分析信号。
为了增强目标信号，我们使用的是Stouffer's Z-score算法，这就需要确认K值的最佳解。
按照不同的参数K统计List1中数据与List2数据中的差异点(P＜0.05)，为了选择最优化的参数K，我们需计算每个参数K下所有异常点的频率(公式1)，这样我们得到LIST1下面各个参数K的频率，如表2所示。
公式1：
Figure PCTCN2018072393-appb-000003
公式1说明：
A、常数24：
我们对K进行5-51，step＝2进行取值，总共进行了24组异常点数据；
B、该位点在不同参数K中出现异常次数：
在当前K值下的某个异常位点在所有24次分析中出现的次数；
C、当前K值下异常点数目：
在当前K值下，所有P<0.05的点的数目；
公式1计算得到的在当前K值下Freq1频率分布，如图3所示，选取频率最大的K值做为最优K。可以发现当K＝29时，频率达到最高。
表2 LIST1下面各个参数K的频率表
K值 5 7 9 11 13 15
频率 0.527 0.630 0.656 0.684 0.717 0.749
K值 17 19 21 23 25 27
频率 0.752 0.763 0.758 0.765 0.772 0.776
K值 29 31 33 35 37 39
频率 0.778 0.774 0.773 0.772 0.759 0.743
K值 41 43 45 47 49 51
频率 0.736 0.723 0.719 0.705 0.689 0.674
确定最优K值后，利用背景信号消除算法，可以得到经过处理后的信号，其处理前后信号比较如图6所示。
如图7、8所示，对处理后LIST1和LIST2信号进行双样本U检验比较，找出差异最大的100个数据点(或者p值<0.05的数据点)，作为后续SVM分类所需的明显差异点。
(4)拉曼光谱信号的分析：对步骤(3)得到的可分析信号进行差异统计分析得到差异信号，利用差异信号建模，并利用支持向量机对差异信号进行分类，对正常和异常的细胞培养液光谱信号进行区分，得到细胞培养液质量结果。
(5)利用支持向量机(Support Vector Machine，SVM)算法对细胞培养液进行分类：由于细胞培养液正常/异常分类使用的是SVM算法，所以对正常细胞培养液信号和异常细胞培养液信号进行分组(训练集和预测集)，训练集包含样本409例，预测集中包含样本200例(100例正常细胞培养液、100例异常细胞培养液)。经SVM算法分析后，结果显示在细胞培养液中信号识别率为78％。
表3 SVM分类结果统计
  正常细胞培养液(SVM) 异常细胞培养液(SVM)
正常细胞培养液(原始) 72 28
异常细胞培养液(原始) 16 84
本发明通过细胞培养液的采集；拉曼光谱信号的采集、处理和分析；使用拉曼光谱技术测量细胞培养液中代谢产物的原始拉曼光谱信号；判断原始拉曼光谱信号是否合格，对合格的原始拉曼光谱信号进行数据信号处理，得到可分析信号；再对可分析信号进行差异统计分析得到差异信号，利用差异信号建模，并利用支 持向量机对差异信号进行分类，对正常和异常的细胞培养液光谱信号进行区分，得到细胞培养液质量结果。本发明通过拉曼光谱检测细胞培养液中差异信号，检测出细胞培养液质量，从而达到无创评价细胞生长状态的目的，且本发明便捷有效、低成本，可大规模产业化、流程化。

Claims (8)

  1. 一种基于拉曼光谱测量的细胞培养液质量检测方法，其特征在于，包括如下步骤：
    (1)细胞培养液的采集：取经细胞培养一段时间的细胞培养液；
    (2)拉曼光谱信号的采集：使用拉曼光谱技术测量步骤(1)所得细胞培养液中代谢产物的原始拉曼光谱信号；判断原始拉曼光谱信号是否合格，是则进入步骤(3)，否则，重复对该细胞培养液进行拉曼光谱技术测量；
    (3)拉曼光谱信号的处理：对步骤(2)获取的合格原始拉曼光谱信号进行数据信号处理，得到可分析信号；
    (4)拉曼光谱信号的分析：对步骤(3)得到的可分析信号进行差异统计分析得到差异信号，利用差异信号建模，并利用支持向量机对差异信号进行分类，对正常和异常的细胞培养液光谱信号进行区分，得到细胞培养液质量结果。
  2. 根据权利要求1所述的基于拉曼光谱测量的细胞培养液质量检测方法，其特征在于，所述步骤(1)中所取的细胞培养液为7微升。
  3. 根据权利要求2所述的基于拉曼光谱测量的细胞培养液质量检测方法，其特征在于，所述7微升细胞培养液放置在微量溶液检测池中。
  4. 根据权利要求3所述的基于拉曼光谱测量的细胞培养液质量检测方法，其特征在于，所述微量溶液检测池包括半球面反射镜，或者由半球面反射镜和圆柱面反射镜构成反射镜组，所述半球面反射镜和反射镜组都具有高反射的镜面。
  5. 根据权利要求4所述的基于拉曼光谱测量的细胞培养液质量检测方法，其特征在于，所述镜面镀介质膜或惰性金属膜。
  6. 根据权利要求1所述的基于拉曼光谱测量的细胞培养液质量检测方法，其特征在于，所述步骤(1)中一段时间为3-4天。
  7. 根据权利要求1所述的基于拉曼光谱测量的细胞培养液质量检测方法，其特征在于，所述步骤(2)判断原始拉曼光谱信号是否合格过程为：0.5％的乙醇溶液在880cm -1处散射峰绝对峰强CV≤5％，拉曼位移波动范围≤2cm -1。
  8. 根据权利要求1所述的基于拉曼光谱测量的细胞培养液质量检测方法，其特征在于，所述步骤(3)中对获取的原始拉曼光谱信号进行数据信号处理步骤包括：
    1)对获取的原始拉曼光谱信号进行数据校正；
    利用最小二乘算法，通过最小化误差的平方和寻找数据的最佳函数匹配，最佳函数匹配通过让采样点跟拟合曲线的距离平方和最小实现数据校正；
    2)将经数据校正后的信号进行去除荧光信号背景；
    a利用墨西哥帽小波为母函数的连续小波模式匹配方法准确确定峰位置；
    b.继续利用以Haar小波为母函数的连续小波求导方法确定峰的起始位置；
    c.利用惩罚最小二乘法拟合出平滑可调背景；
    3)对去除荧光信号背景的信号进行均一化处理
    基于Stouffer's Z-score算法对去除荧光信号背景的信号进行均一化处理，得到可分析信号。
PCT/CN2018/072393 2017-04-10 2018-01-12 一种基于拉曼光谱测量的细胞培养液质量检测方法 WO2018188395A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US16/499,941 US11561182B2 (en) 2017-04-10 2018-01-12 Method for detecting quality of cell culture fluid based on Raman spectral measurement
EP18784206.7A EP3611495A4 (en) 2017-04-10 2018-01-12 METHOD OF DETECTING THE QUALITY OF A CELL CULTURE LIQUID BASED ON RAMAN SPECTRAL MEASUREMENT

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710227963.9 2017-04-10
CN201710227963.9A CN107132208B (zh) 2017-04-10 2017-04-10 一种基于拉曼光谱测量的细胞培养液质量检测方法

Publications (1)

Publication Number Publication Date
WO2018188395A1 true WO2018188395A1 (zh) 2018-10-18

Family

ID=59715761

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/072393 WO2018188395A1 (zh) 2017-04-10 2018-01-12 一种基于拉曼光谱测量的细胞培养液质量检测方法

Country Status (4)

Country Link
US (1) US11561182B2 (zh)
EP (1) EP3611495A4 (zh)
CN (1) CN107132208B (zh)
WO (1) WO2018188395A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111289489A (zh) * 2020-03-05 2020-06-16 长春长光辰英生物科学仪器有限公司 一种基于拉曼光谱的微生物单细胞生长检测方法
US11358984B2 (en) 2018-08-27 2022-06-14 Regeneran Pharmaceuticals, Inc. Use of Raman spectroscopy in downstream purification
US11609120B2 (en) 2017-10-06 2023-03-21 Lonza Ltd Automated control of cell culture using Raman spectroscopy

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107132208B (zh) * 2017-04-10 2018-11-30 苏州贝康医疗器械有限公司 一种基于拉曼光谱测量的细胞培养液质量检测方法
CN110874548A (zh) * 2018-08-31 2020-03-10 天津理工大学 一种基于拉曼光谱结合svm的肺癌细胞与正常细胞的识别方法
CN113252628B (zh) * 2021-05-12 2023-06-16 中国科学技术大学 一种荧光光谱水质监测装置及其监测方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101082585A (zh) * 2007-07-19 2007-12-05 清华大学 反射式近场拉曼光谱仪测量头
US20120035442A1 (en) * 2010-08-05 2012-02-09 Ishan Barman Portable raman diagnostic system
CN103901014A (zh) * 2014-03-10 2014-07-02 华南师范大学 多元线性回归拟合获得真实的细胞拉曼光谱的方法
CN103940801A (zh) * 2014-04-17 2014-07-23 中国科学院青岛生物能源与过程研究所 一种单细胞水平快速鉴定微藻的方法
CN104634771A (zh) * 2015-01-28 2015-05-20 浙江大学 基于微藻油脂特征峰的氮胁迫随时间变化的监测方法
WO2016003371A1 (en) * 2014-07-02 2016-01-07 National University Of Singapore Raman spectroscopy system, apparatus, and method for analyzing, characterizing, and/or diagnosing a type or nature of a sample or a tissue such as an abnormal growth
CN107132208A (zh) * 2017-04-10 2017-09-05 苏州贝康医疗器械有限公司 一种基于拉曼光谱测量的细胞培养液质量检测方法

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE29824644U1 (de) * 1997-09-17 2001-12-06 Glaxo Group Ltd Vorrichtung zur Durchführung photometrischer Tests
WO2008011046A2 (en) * 2006-07-17 2008-01-24 The H.Lee Moffitt Cancer And Research Institute, Inc. Computer systems and methods for selecting subjects for clinical trials
US20140106469A1 (en) * 2012-10-11 2014-04-17 The Regents Of The University Of California Nanoparticle Plasmonic Sensor for Localized Surface Plasmon Resonance
DK3314007T3 (da) * 2015-06-25 2024-04-22 Native Microbials Inc Fremgangsmåder, apparater og systemer til analyse af mikroorganismestammer fra komplekse heterogene miljøer til forudsigelse og identificering af funktionelle forbindelser og interaktioner derimellem og udvælgelse og syntese af mikrobielle ensembler baseret derpå

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101082585A (zh) * 2007-07-19 2007-12-05 清华大学 反射式近场拉曼光谱仪测量头
US20120035442A1 (en) * 2010-08-05 2012-02-09 Ishan Barman Portable raman diagnostic system
CN103901014A (zh) * 2014-03-10 2014-07-02 华南师范大学 多元线性回归拟合获得真实的细胞拉曼光谱的方法
CN103940801A (zh) * 2014-04-17 2014-07-23 中国科学院青岛生物能源与过程研究所 一种单细胞水平快速鉴定微藻的方法
WO2016003371A1 (en) * 2014-07-02 2016-01-07 National University Of Singapore Raman spectroscopy system, apparatus, and method for analyzing, characterizing, and/or diagnosing a type or nature of a sample or a tissue such as an abnormal growth
CN104634771A (zh) * 2015-01-28 2015-05-20 浙江大学 基于微藻油脂特征峰的氮胁迫随时间变化的监测方法
CN107132208A (zh) * 2017-04-10 2017-09-05 苏州贝康医疗器械有限公司 一种基于拉曼光谱测量的细胞培养液质量检测方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3611495A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11609120B2 (en) 2017-10-06 2023-03-21 Lonza Ltd Automated control of cell culture using Raman spectroscopy
US11358984B2 (en) 2018-08-27 2022-06-14 Regeneran Pharmaceuticals, Inc. Use of Raman spectroscopy in downstream purification
CN111289489A (zh) * 2020-03-05 2020-06-16 长春长光辰英生物科学仪器有限公司 一种基于拉曼光谱的微生物单细胞生长检测方法

Also Published As

Publication number Publication date
CN107132208A (zh) 2017-09-05
US11561182B2 (en) 2023-01-24
US20210116382A1 (en) 2021-04-22
CN107132208B (zh) 2018-11-30
EP3611495A4 (en) 2020-12-16
EP3611495A1 (en) 2020-02-19

Similar Documents

Publication Publication Date Title
WO2018188395A1 (zh) 一种基于拉曼光谱测量的细胞培养液质量检测方法
CN101713731B (zh) 一种药物制剂包衣质量的鉴别方法
CN109001180B (zh) 一种拉曼光谱结合人工智能高通量单细胞分析鉴定方法
CN106778893B (zh) 一种基于降维与聚类的高光谱样本选择方法
CN103543123A (zh) 一种掺假牛奶的红外光谱识别方法
JP2022547771A (ja) 微生物を同定する方法およびシステム
CN111289489A (zh) 一种基于拉曼光谱的微生物单细胞生长检测方法
CN113310930A (zh) 高温灭菌奶、巴氏杀菌奶和掺加高温灭菌奶的巴氏杀菌奶的光谱鉴定方法
Fenn et al. Raman spectroscopy utilizing Fisher‐based feature selection combined with Support Vector Machines for the characterization of breast cell lines
CN115078331B (zh) 一种光谱学和人工智能交互的血清分析方法及其应用
CN112730312A (zh) 一种基于近红外光谱技术的掺杂牛初乳定性识别方法
WO2012059748A1 (en) Method, apparatus and software for identifying cells
Xu et al. Fermentation process monitoring of broad bean paste quality by NIR combined with chemometrics
CN108342447B (zh) 一种筛选与已知菌株表型相似的菌株的方法
CN110987856B (zh) 基于配方体系和指纹图谱的化妆品质量快速鉴定方法
CN111220575A (zh) 一种基于太赫兹近场光谱的细胞检测方法
CN115586159A (zh) 基于中红外光谱检测技术的白酒陈化缔合程度评定方法
Yan et al. Rapid detection of cAMP content in red jujube using near-infrared spectroscopy
Schulze et al. Fully Automated Decomposition of Raman Spectra into Individual Pearson's Type VII Distributions Applied to Biological and Biomedical Samples
CN106932377B (zh) 一种利用拉曼增强光谱检测食品中致病菌的方法
CN110320308A (zh) 蜂蜜中2-甲氧基苯乙酮的测定方法
CN110609030B (zh) 一种基于检出概率模型的拉曼快检性能综合评价方法
CN114152557A (zh) 基于图像分析的血细胞计数方法和系统
CN112683900A (zh) 一种手性多巴检测专用纸芯片及检测分析方法
CN112067577A (zh) 基于支持向量机的奶油色素超标鉴别方法、装置及设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18784206

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2018784206

Country of ref document: EP

Effective date: 20191111