Double-variable simulation analysis system based on DNA content and cell cycle related protein

Abstract

The invention relates to the technical field of breast cancer precision medical treatment, and discloses a double-variable simulation analysis system based on DNA content and cell cycle related protein. Comprising a sample processing module used for obtaining a breast cancer tumor cell sample and carrying out fluorescence labeling processing on the sample; the double-variable detection module is used for detecting the DNA content and the protein expression quantity so as to generate double-variable data points of the DNA content and the protein expression of each breast cancer tumor cell sample; the control module is used for analyzing a time difference between cell cycle related protein expression and DNA content change through a quasi-time sequence algorithm based on the bivariate data points, and constructing a time sequence function model reflecting a cell cycle process; and fitting the time sequence function model to generate a cell cycle dynamic curve, and pre-judging the drug resistance and prognosis of the breast cancer cells according to the abnormal characteristics of G1 / S phase transition nodes in the cell cycle dynamic curve, the overexpression level of cell cycle protein E and the abnormal stagnation proportion of S phase cells.

Description

Technical Field

[0001] This application relates to the field of precision medicine technology for breast cancer, and in particular to a bivariate simulation analysis system based on DNA content and cell cycle-related proteins. Background Technology

[0002] Breast cancer, a prevalent malignant tumor among women worldwide, exhibits significant heterogeneity leading to marked differences in treatment responses across different subtypes. Abnormal cell cycle regulation is a core characteristic of tumorigenesis and development, with uncontrolled G1 / S phase transitions closely associated with drug resistance and poor prognosis in breast cancer. Traditional cell cycle analysis methods primarily rely on single-dimensional detection: DNA content detection: flow cytometry analysis of DNA ploidy (e.g., the proportion of G1 / G0, S, and G2 / M phases) can reflect cell cycle distribution but fails to reveal the molecular mechanisms of regulatory nodes (e.g., abnormal G1 / S phase transitions driven by Cyclin E overexpression). Protein expression detection: immunohistochemistry or Western blot can detect cyclin levels, but these methods are limited to averaging analysis of tissue samples, failing to achieve a dynamic correlation between cell cycle stages and protein expression at the single-cell level, and lacking quantitative analysis of key time points in the G1 / S phase transition.

[0003] Existing technologies do not combine the division of DNA content into different phases of the cell cycle with the single-cell expression characteristics of cyclins (especially Cyclin E), and cannot reveal the dynamic temporal relationship between protein expression and DNA replication through time series modeling. This results in a significant blind spot in the monitoring of abnormal cell cycle regulation in breast cancer (such as S-phase arrest and DNA damage caused by Cyclin E overexpression), making it difficult to accurately predict drug resistance and prognosis.

[0004] Therefore, there is an urgent need to design a technical solution to solve at least one of the above-mentioned technical problems. Summary of the Invention

[0005] This application provides a bivariate simulation analysis system based on DNA content and cell cycle-related proteins, aiming to solve the problems of existing technologies that fail to combine the cell cycle stage division of DNA content with the single-cell expression characteristics of cell cycle proteins (especially Cyclin E), and are unable to reveal the dynamic temporal relationship between protein expression and DNA replication process through time series modeling. This results in significant blind spots in the monitoring of abnormal cell cycle regulation in breast cancer (such as S phase arrest and DNA damage caused by Cyclin E overexpression), making it difficult to accurately predict drug resistance and prognosis.

[0006] In a first aspect, this application provides a bivariate simulation analysis system based on DNA content and cell cycle-related proteins, comprising:

[0007] The sample processing module is used to acquire breast cancer tumor cell samples and perform fluorescent labeling on the breast cancer tumor cell samples. The labeling of cell cycle-related proteins includes specific fluorescent antibodies for cyclin D, cyclin E, cyclin A and cyclin B1, and the fluorescent labeling signal of cyclin E is used as the main monitoring indicator.

[0008] The bivariate detection module is used to detect the DNA content of breast cancer tumor cell samples using quantitative staining flow cytometry to determine the relative proportions of the G1, S, and G2 / M phases in which the breast cancer tumor cell samples are located; it is also used to detect the protein expression levels of cyclin D, E, A, and B1 in breast cancer tumor cell samples, and generate bivariate data points of DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportions and protein expression levels.

[0009] The control module is used to analyze the time difference between changes in cell cycle-related protein expression and DNA content based on the bivariate data points using a pseudo-time series algorithm, and construct a time series function model reflecting the cell cycle process. The time series function model uses the change in cyclin E expression as the core variable driving the G1 / S phase transition. The time series function model is fitted to generate a cell cycle dynamic curve. Based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested S-phase cells in the cell cycle dynamic curve, the drug resistance and prognosis of breast cancer cells are predicted in advance.

[0010] In some embodiments, the fluorescent labeling treatment of the sample includes: after permeabilizing the breast cancer tumor cell sample, adding fluorescently labeled antibodies against cyclin D, E, A, and B1 respectively, wherein the antibody against cyclin E is labeled with a near-infrared fluorescent group, and its fluorescence signal intensity is 2-3 orders of magnitude higher than that of the fluorescent groups labeled with antibodies against other proteins, and the fluorescence emission spectrum overlap of each antibody is ≤10%, so as to achieve high-precision resolution of multi-protein labeling signals in a single cell; the excitation wavelength of the near-infrared fluorescent group is 780 nm, and the emission wavelength is 800-850 nm.

[0011] In some embodiments, the flow cytometry method using quantitative staining to detect the DNA content of breast cancer tumor cell samples to determine the relative proportions of the G1, S, and G2 / M phases of the breast cancer tumor cell samples includes: pretreating the samples with Hoechst 33342 combined with RNase, detecting DNA fluorescence intensity by ultraviolet laser excitation, introducing normal diploid cells as a calibration reference when performing peak shape fitting based on ModFit software, calibrating the DNA content of the G1 phase to a 2C value, the G2 / M phase to a 4C value, and defining the S phase as a continuous distribution between 2C and 4C, while excluding adherent cells and fragmented cells through pulse shape analysis, so that the detection error of the proportion of cells in each cycle is ≤5%.

[0012] In some embodiments, the detection of the protein expression levels of cyclins D, E, A, and B1 in breast cancer tumor cell samples includes: real-time acquisition of fluorescence signals of antibodies against each protein in a single cell using a multi-parameter detection channel of flow cytometry; conversion of fluorescence signals into protein molecule numbers using microsphere-based absolute counting (CBA); and simultaneous addition of a negative control for background correction during the detection process to ensure that the false negative rate of abnormally low expression signals is ≤3%.

[0013] In some embodiments, generating bivariate data points of DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportion and protein expression level includes: normalizing the DNA content of a single cell, expressed as a multiple of 2C, with the number of molecules of cell cycle proteins D, E, A, and B1; using Z-score normalization to eliminate dimensional differences between different proteins; and clustering the two-dimensional scatter plot using the DBSCAN algorithm to identify and remove outliers to generate the bivariate data points.

[0014] In some embodiments, the step of analyzing the time difference between cell cycle-related protein expression and DNA content changes based on the bivariate data points using a pseudo-time series algorithm to construct a time series function model reflecting the cell cycle process includes: using the expression change curve of cyclin E as the time axis reference, dividing the DNA content change process from 2C to 4C into three stages: G1 phase, S phase, and G2 / M phase; using a hidden Markov model to identify the state transition points between each stage; calculating the time delay between the protein expression peak and the DNA replication initiation point; and constructing the time series function model based on the time delay.

[0015] In some embodiments, the process of fitting the time series function model to generate a cell cycle dynamic curve, and making a preliminary judgment on the drug resistance and prognosis of breast cancer cells based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested S-phase cells in the cell cycle dynamic curve, includes: when the expression level of cyclin E at the G1 / S phase transition node in the cell cycle dynamic curve exceeds the normal threshold by 1.5 times, and the proportion of γ-H2AX positive cells in S-phase cells is >15%, the cell cycle dynamic curve is identified as being in the chemotherapy resistance risk group; when the cell cycle dynamic curve shows that the Pearson correlation coefficient between the expression level of cyclin E in G1 phase cells and the proportion of S-phase cells is >0.6, and after calibration with the Ki-67 index, the accuracy of the curve predicting a disease-free survival of <24 months is ≥90%, the cell cycle dynamic curve is identified as having a poor prognosis.

[0016] Secondly, this application provides a bivariate simulation analysis method based on DNA content and cell cycle-related proteins, characterized in that a control module applied to the bivariate simulation analysis system based on DNA content and cell cycle-related proteins provided in any embodiment of this application is included in the method:

[0017] The system acquires bivariate data points on DNA content and protein expression generated by a bivariate detection module. These bivariate data points are then used by the bivariate detection module to detect the DNA content of breast cancer cell samples using quantitative staining flow cytometry, in order to determine the relative proportions of the G1, S, and G2 / M phases in which the breast cancer cell samples are located. The system also detects the protein expression levels of cyclins D, E, A, and B1 in the breast cancer cell samples, and generates bivariate data points on DNA content and protein expression for each breast cancer cell sample based on these relative proportions and protein expression levels.

[0018] Based on the bivariate data points, the time difference between the expression of cell cycle-related proteins and changes in DNA content was analyzed using a pseudo-time series algorithm. A time series function model reflecting the cell cycle process was constructed, in which the expression change of cyclin E was used as the core variable driving the G1 / S phase transition.

[0019] The time series function model is fitted to generate a cell cycle dynamic curve. Based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested cells in the S phase, the drug resistance and prognosis of breast cancer cells can be predicted in advance.

[0020] Thirdly, this application provides a bivariate simulation analysis device based on DNA content and cell cycle-related proteins, applied to the control module of the bivariate simulation analysis system based on DNA content and cell cycle-related proteins provided in any embodiment of this application, the device comprising:

[0021] The data acquisition unit is used to acquire bivariate data points of DNA content and protein expression generated by the bivariate detection module. The bivariate data points are used by the bivariate detection module to detect the DNA content of breast cancer tumor cell samples using quantitative staining flow cytometry to determine the relative proportion of the G1, S, and G2 / M phases in which the breast cancer tumor cell samples are located. The unit is also used to detect the protein expression levels of cell cycle proteins D, E, A, and B1 in breast cancer tumor cell samples, and generate bivariate data points of DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportions and protein expression levels.

[0022] The variable generation unit is used to analyze the time difference between the expression of cell cycle-related proteins and the change in DNA content based on the bivariate data points using a pseudo-time series algorithm, and to construct a time series function model that reflects the cell cycle process. The time series function model uses the change in the expression of cyclin E as the core variable driving the G1 / S phase transition.

[0023] The proportion generation unit is used to fit the time series function model to generate a cell cycle dynamic curve. Based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested S phase cells in the cell cycle dynamic curve, the drug resistance and prognosis of breast cancer cells can be predicted in advance.

[0024] Fourthly, this application provides a control module, which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and, when executing the computer program, implement the method provided in any embodiment of this application.

[0025] Fifthly, this application provides a computer-readable storage medium storing a computer program, wherein the computer-readable instructions, when executed by a processor, cause one or more processors to perform the method provided in any embodiment of this application.

[0026] This invention provides a bivariate simulation analysis system based on DNA content and cell cycle-related proteins, the core technical solution of which includes:

[0027] Bivariate detection mechanism: The DNA content (cycle stage) and expression levels of cyclin D, E, A, and B1 in single cells are obtained simultaneously by flow cytometry, with the fluorescent labeling signal of Cyclin E as the core monitoring indicator, to construct a bivariate data point of DNA content-protein expression, breaking through the limitations of traditional single-dimensional detection.

[0028] Pseudo-time series modeling: Based on bivariate data, the time difference between changes in protein expression such as Cyclin E and changes in DNA content is analyzed using a pseudo-time series algorithm. A time series function model with changes in Cyclin E expression as the core driving variable of the G1 / S phase transition is constructed, achieving quantitative simulation of the dynamic process of the cell cycle for the first time.

[0029] Drug resistance and prognosis: By fitting the generated cell cycle dynamic curve, we can capture abnormalities at the G1 / S phase transition node (such as Cyclin E overexpression) and characteristics of S phase cell arrest, and establish quantitative judgment criteria related to drug resistance and prognosis, filling the gap in the lack of molecular-cycle dynamic correlation analysis in the existing technology.

[0030] The system provided by this invention has the following beneficial effects:

[0031] Innovative detection methods: Integrating bivariate analysis of DNA cycle distribution and multi-cell cyclin expression, it particularly highlights the core driving role of Cyclin E in the G1 / S phase transition, achieving a detection upgrade from "static cycle distribution" to "dynamic molecular regulation".

[0032] Breakthrough in temporal analysis: By quantifying the time difference between protein expression and DNA replication using a pseudo-time series algorithm, a dynamic model reflecting the cell cycle process is constructed to accurately locate abnormal regulatory nodes (such as premature S-phase entry and DNA damage risk caused by Cyclin E overexpression).

[0033] Clinical application value: Based on dynamic curves of drug resistance and prognosis prediction, combined with features such as Cyclin E overexpression level and S-phase arrest ratio, it provides molecular-level decision-making basis for individualized treatment of breast cancer (such as early warning of drug resistance to CDK4 / 6 inhibitors), significantly improving the accuracy of prediction and clinical guidance significance.

[0034] In summary, this invention solves the problems of single-dimensional cell cycle analysis and lack of temporal regulation analysis in existing technologies by combining bivariate detection and dynamic modeling. Its core innovation (bivariate analysis based on Cyclin E and pseudo-time series function modeling) constitutes a substantial improvement to breast cancer cell cycle analysis technology and has significant inventiveness and clinical application value.

[0035] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a schematic diagram of the structure of a bivariate simulation analysis system based on DNA content and cell cycle-related proteins provided in an embodiment of this application;

[0038] Figure 2 This is a schematic flowchart illustrating the steps of a bivariate simulation analysis method based on DNA content and cell cycle-related proteins provided in an embodiment of this application;

[0039] Figure 3 This is a schematic block diagram of the structure of a control module provided in one embodiment of this application.

[0040] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. Detailed Implementation

[0041] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0042] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0043] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0044] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0045] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0046] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0047] It should be noted that all information involved in this application was obtained with the permission of the relevant users and in compliance with relevant policies. It will not infringe upon users' personal information.

[0048] Breast cancer, a prevalent malignant tumor among women worldwide, exhibits significant heterogeneity leading to marked differences in treatment responses across different subtypes. Abnormal cell cycle regulation is a core characteristic of tumorigenesis and development, with uncontrolled G1 / S phase transitions closely associated with drug resistance and poor prognosis in breast cancer. Traditional cell cycle analysis methods primarily rely on single-dimensional detection: DNA content detection: flow cytometry analysis of DNA ploidy (e.g., the proportion of G1 / G0, S, and G2 / M phases) can reflect cell cycle distribution but fails to reveal the molecular mechanisms of regulatory nodes (e.g., abnormal G1 / S phase transitions driven by Cyclin E overexpression). Protein expression detection: immunohistochemistry or Western blot can detect cyclin levels, but these methods are limited to averaging analysis of tissue samples, failing to achieve a dynamic correlation between cell cycle stages and protein expression at the single-cell level, and lacking quantitative analysis of key time points in the G1 / S phase transition.

[0049] Existing technologies do not combine the division of DNA content into different phases of the cell cycle with the single-cell expression characteristics of cyclins (especially Cyclin E), and cannot reveal the dynamic temporal relationship between protein expression and DNA replication through time series modeling. This results in a significant blind spot in the monitoring of abnormal cell cycle regulation in breast cancer (such as S-phase arrest and DNA damage caused by Cyclin E overexpression), making it difficult to accurately predict drug resistance and prognosis.

[0050] Therefore, there is an urgent need to design a technical solution to solve at least one of the above-mentioned technical problems.

[0051] It should be noted that all information involved in this application is provided with the permission of the relevant users and in compliance with relevant policies, and will not cause any damage to the users' privacy.

[0052] To solve the above problem, please refer to Figure 1 This application provides a bivariate simulation analysis system based on DNA content and cell cycle-related proteins, comprising: a sample processing module for acquiring breast cancer tumor cell samples and performing fluorescent labeling on the breast cancer tumor cell samples, wherein the labeling for cell cycle-related proteins includes specific fluorescent antibodies for cyclin D, cyclin E, cyclin A, and cyclin B1, and the fluorescent labeling signal of cyclin E is used as the primary monitoring indicator; and a bivariate detection module for detecting the DNA content of the breast cancer tumor cell samples using quantitative staining flow cytometry to determine the relative proportions of the G1, S, and G2 / M phases in which the breast cancer tumor cell samples are located; and also for detecting cyclin D, E, and A in the breast cancer tumor cell samples. The expression level of B1 protein is used to generate bivariate data points of DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportion and protein expression level. A control module is used to analyze the time difference between changes in cell cycle-related protein expression and DNA content using a pseudo-time series algorithm based on the bivariate data points, constructing a time series function model reflecting the cell cycle process. This time series function model uses changes in cyclin E expression as the core variable driving the G1 / S phase transition. The time series function model is fitted to generate a cell cycle dynamic curve. Based on the abnormal characteristics of the G1 / S phase transition node in the cell cycle dynamic curve, the overexpression level of cyclin E, and the proportion of abnormally arrested S-phase cells, the drug resistance and prognosis of breast cancer cells are pre-judged.

[0053] Specifically, this system integrates DNA content detection with single-cell expression characteristics of cyclins (especially Cyclin E) to construct a bivariate time series model, enabling precise analysis of abnormal cell cycle regulation in breast cancer. The detection index combination simultaneously monitors DNA content (reflecting cell cycle stages: G1 / G0, S, G2 / M phases) and the single-cell expression levels of four cyclins (Cyclin D, E, A, B1), with Cyclin E as the core monitoring index (because its overexpression directly drives the abnormal transition to the G1 / S phase).

[0054] Single-cell resolution: By using fluorescent labeling and flow cytometry, the DNA content (labeled with DNA dyes such as propidium iodide) and protein expression level (labeled with specific fluorescent antibodies) of each cell are obtained, forming bivariate data points of "DNA content-protein expression", which breaks through the limitations of traditional tissue sample averaging analysis.

[0055] The time difference analysis algorithm includes: based on single-cell bivariate data, using pseudo-time series algorithms (such as Monocle, PAGA trajectory inference methods, etc.) to calculate the time difference between changes in cyclin expression and DNA replication process (changes in DNA content), constructing a dynamic temporal relationship reflecting the G1 / S phase transition. The core variable-driven model: using changes in Cyclin E expression as the core driving variable for the G1 / S phase transition, a time series function model is established: f(t) = [DNA(t), Cyclin E(t), Cyclin D / A / B1(t)], where t represents pseudo-time (a virtual time axis reflecting the cell cycle process).

[0056] Abnormal feature quantification indicators include: G1 / S phase transition node abnormalities: identifying the transition node caused by Cyclin E overexpression as either advanced or delayed through modeling; S phase arrest ratio: calculating the proportion of S phase cells with abnormally high Cyclin E expression based on the bivariate distribution of DNA content and Cyclin E expression; DNA damage association: assessing the risk of DNA damage by combining Cyclin E overexpression with abnormal DNA content in S phase cells (such as polyploidy).

[0057] Sample acquisition and pretreatment include: isolating single-cell suspensions from tumor tissue of breast cancer patients, removing impurities through enzymatic digestion (such as collagenase) and filtration to obtain single-cell suspensions; fixing cells (such as with methanol / paraformaldehyde) to permeabilize the cell membrane and ensure that fluorescent antibodies enter the cells.

[0058] The fluorescent labeling scheme includes: DNA content labeling: using DNA-binding dyes such as propidium iodide (PI) or 4',6-diamidinyl-2-phenylindole (DAPI), fluorescence intensity is detected after binding to double-stranded DNA; Protein labeling: using fluorescently conjugated specific antibodies such as AlexaFluor series or PE / Cy5, single-cell immunofluorescence staining is performed against Cyclin D1, Cyclin E1 (core), Cyclin A2, and Cyclin B1 to ensure antibody specificity and non-interference of fluorescence signals (through spectral compensation calibration).

[0059] Flow cytometry assays include: using a flow cytometer equipped with multiple laser channels (such as BD LSR Fortessa), setting up a DNA fluorescence channel (such as 620 nm emission from PI) and a protein fluorescence channel (such as 575 nm emission from Cyclin E antibody); acquiring at least 10,000 single-cell data, and recording the DNA fluorescence intensity of each cell (for dividing the G1 / S / G2 / M phases) and the fluorescence intensity of four cell cycle proteins (normalized to protein expression levels).

[0060] Data preprocessing: Remove dead cells (by using live / dead dyes) and cell debris (by using forward / side scattering light gating); normalize the bivariate data (e.g., Z-score standardization) to generate coordinates (DNA norm, Cyclin Enorm) and other protein expression values ​​for each cell.

[0061] The pseudo-time series modeling includes: constructing a pseudo-time axis of the cell cycle process based on single-cell bivariate data using trajectory inference algorithms (such as DDRTree or Slingshot), with G1 phase as the starting point, S phase as the intermediate state, and G2 / M phase as the ending point; fitting a time series function model and determining the temporal relationship between Cyclin E expression and DNA content changes through regression analysis, such as identifying the time difference Δt between the peak expression of Cyclin E and the G1 / S phase transition.

[0062] Dynamic curve fitting and anomaly analysis: Based on the model, cell cycle dynamic curves are generated to show the changing trends of protein expression with simulated time (e.g., Cyclin E increases sharply in late G1 phase, driving cells into S phase).

[0063] Define abnormal features: If Cyclin E is overexpressed (exceeding the mean of normal samples + 2 SD) and Δt is shortened (the transition node is advanced), it is judged as a risk of G1 / S phase out of control; calculate the proportion of Cyclin E high expression in S phase cells, and combine it with abnormal DNA content (such as abnormal DNA synthesis rate in S phase) to assess the risk of S phase arrest.

[0064] Drug resistance and prognosis: Establish a discriminant model and input the degree of abnormality of G1 / S phase transition, Cyclin E overexpression level, and S phase stagnation ratio into the logistic regression or random forest model to output the drug resistance risk level (high / medium / low) and prognostic score (such as 5-year disease-free survival prediction).

[0065] Traditional methods only detect DNA content or average protein expression. This system combines DNA content (cycle phase) at the single-cell level with protein expression such as Cyclin E to reveal the molecular mechanism of the G1 / S phase transition (such as how Cyclin E overexpression drives abnormal transition), thus solving the blind spot of "inability to link cycle phase with protein dynamics".

[0066] By employing a pseudo-time series algorithm, the time difference between protein expression and DNA replication is quantified to construct a dynamic model of cell cycle progression. This enables, for the first time, precise localization of key nodes in the G1 / S phase transition (such as identifying transition abnormalities caused by premature activation of Cyclin E), providing temporal evidence for elucidating drug resistance mechanisms. Single-cell-level data analysis avoids the heterogeneity masked by tissue sample averaging and can detect aberrant Cyclin E expression in a small subset of drug-resistant cell populations, enabling early detection of potential drug resistance risks (such as abnormal G1 / S phase regulation in circulating tumor cells).

[0067] By quantifying parameters such as Cyclin E overexpression, G1 / S node abnormalities, and S-phase arrest, a correlation model with clinical drug resistance and prognosis is established to provide precise stratification for breast cancer patients: for patients with high Cyclin E expression and uncontrolled G1 / S phase, CDK4 / 6 inhibitors (such as Palbociclib) are recommended as the first choice for combination therapy; for patients with S-phase arrest and risk of DNA damage, chemotherapy resistance is warned and it is recommended to adjust the treatment plan in advance.

[0068] The bivariate dynamic data generated by the system provides resources for basic research, such as verifying whether Cyclin E overexpression causes S-phase arrest by affecting DNA replication fork progression, thus promoting the development of cell cycle-targeting drugs.

[0069] This system fills the technological gap in monitoring abnormal cell cycle regulation in breast cancer by integrating two variables, namely DNA content and cyclin, conducting single-cell-level detection, and using time-series modeling. It provides an innovative tool for accurately assessing drug resistance and prognosis, and has significant clinical translational value and research application potential.

[0070] In some embodiments, the fluorescent labeling treatment of the sample includes: after permeabilizing the breast cancer tumor cell sample, adding fluorescently labeled antibodies against cyclin D, E, A, and B1 respectively, wherein the antibody against cyclin E is labeled with a near-infrared fluorescent group, and its fluorescence signal intensity is 2-3 orders of magnitude higher than that of the fluorescent groups labeled with antibodies against other proteins, and the fluorescence emission spectrum overlap of each antibody is ≤10%, so as to achieve high-precision resolution of multi-protein labeling signals in a single cell; the excitation wavelength of the near-infrared fluorescent group is 780 nm, and the emission wavelength is 800-850 nm.

[0071] Permeabilization process: First, add breast cancer single cell suspension to PBS buffer (pH 7.4) containing 0.1% Triton X-100 and incubate at room temperature for 10 minutes to disrupt the phospholipid bilayer structure of the cell membrane and form permeabilization channels; then add blocking solution containing 5% bovine serum albumin (BSA) and incubate at room temperature for 30 minutes to block non-specific protein binding sites and reduce non-specific antibody adsorption.

[0072] Fluorescent antibody labeling schemes include: Cyclin E antibody labeling: Cyclin E-specific monoclonal antibodies conjugated to near-infrared fluorescent groups (such as Alexa Fluor 790 or IRDye 800CW), with excitation wavelength of 780 nm and emission wavelength of 800-850 nm. The fluorescent group is covalently linked to the antibody Fc segment via carbodiimide crosslinking, controlling the labeling ratio at 1:1 (one fluorescent group per antibody molecule), ensuring the fluorescence signal intensity is 2-3 orders of magnitude higher than other protein antibodies (calibrated using a fluorescence spectrophotometer); Other protein antibody labeling: Cyclin Antibodies D, A, and B1 were labeled with different fluorescent groups (e.g., Cyclin D-PE, excitation 488nm, emission 575nm; Cyclin A-APC, excitation 633nm, emission 660nm; Cyclin B1-PacificBlue, excitation 405nm, emission 450nm). Combinations with emission spectral overlap ≤10% were pre-screened using a spectral database to avoid signal crosstalk. Antibody incubation: Cells permeabilized with the antibody mixture (final concentration 10μg / mL) were incubated at 4°C in the dark for 60 minutes, followed by washing three times with PBS to remove unbound antibodies.

[0073] Leveraging the high signal intensity of near-infrared fluorescent groups and spectral separation design, this method solves the problem of protein signal misinterpretation caused by overlapping fluorescence spectra in traditional flow cytometry. In particular, it improves the detection sensitivity of low-abundance or dynamically changing proteins such as Cyclin E (e.g., transient high expression in late G1 phase) by more than 3 times, ensuring accurate quantification of protein expression levels at the single-cell level. As a key driver protein of the G1 / S phase transition, the significant enhancement of Cyclin E's signal intensity can prevent it from being overwhelmed by other protein signals, facilitating accurate capture of the correlation between its expression peak and cycle phase during subsequent modeling, and providing a reliable data foundation for elucidating abnormal regulatory mechanisms.

[0074] In some embodiments, the flow cytometry method using quantitative staining to detect the DNA content of breast cancer tumor cell samples to determine the relative proportions of the G1, S, and G2 / M phases of the breast cancer tumor cell samples includes: pretreating the samples with Hoechst 33342 combined with RNase, detecting DNA fluorescence intensity by ultraviolet laser excitation, introducing normal diploid cells as a calibration reference when performing peak shape fitting based on ModFit software, calibrating the DNA content of the G1 phase to a 2C value, the G2 / M phase to a 4C value, and defining the S phase as a continuous distribution between 2C and 4C, while excluding adherent cells and fragmented cells through pulse shape analysis, so that the detection error of the proportion of cells in each cycle is ≤5%.

[0075] Sample pretreatment included: adding Hoechst 33342 dye (dissolved in PBS containing 2% fetal bovine serum) to a single-cell suspension at a final concentration of 10 μg / mL, incubating at 37°C for 30 minutes, allowing the dye to enter live cells and bind to the minor groove of double-stranded DNA; adding RNase A (final concentration 100 μg / mL), incubating at 37°C for 15 minutes to degrade RNA and eliminate interference from RNA in DNA content detection.

[0076] Flow cytometry detection and data calibration included: using a flow cytometer equipped with a UV laser (355 nm), acquiring blue fluorescence signals (emission wavelength 450 ± 20 nm) from a Hoechst 33342 cell, and simultaneously using forward scatter (FSC) and side scatter (SSC) gating to exclude cell debris (low FSC) and adherent cells (abnormal pulse width); introducing normal mammary epithelial diploid cells as a calibration reference, defining the DNA content in G1 phase as 2C (mean fluorescence intensity), G2 / M phase as 4C (twice that of 2C), and S phase as a continuous distribution between 2C and 4C; using ModFit LT software for peak shape fitting, employing a Gaussian mixture model to decompose the G1, S, and G2 / M phase peaks, and iteratively optimizing the fit R of each peak to achieve the desired fit. 2 ≥0.98, and at the same time, the pulse shape analysis module (such as BD's Diva software) is used to exclude adherent cells with pulse width >2 standard deviation, to ensure that the detection error of each cycle proportion is ≤5%.

[0077] RNA interference is eliminated through RNase pretreatment. Combined with normal diploid cell calibration and pulse shape analysis, this method solves the problem of cycle misjudgment caused by adherent cells or debris in traditional DNA content detection. The detection error of the proportion of G1 / S / G2 / M phases is reduced from 10%-15% in traditional methods to less than 5%. The 2C / 4C phase is uniformly used as the DNA content benchmark, which facilitates the comparison of cycle distribution between different batches of samples. This provides a unified DNA content quantification standard for subsequent bivariate modeling and ensures the accuracy of G1 / S phase transition node location.

[0078] In some embodiments, the detection of the protein expression levels of cyclins D, E, A, and B1 in breast cancer tumor cell samples includes: real-time acquisition of fluorescence signals of antibodies against each protein in a single cell using a multi-parameter detection channel of flow cytometry; conversion of fluorescence signals into protein molecule numbers using microsphere-based absolute counting (CBA); and simultaneous addition of a negative control for background correction during the detection process to ensure that the false negative rate of abnormally low expression signals is ≤3%.

[0079] The microsphere-based absolute counting method (CBA) includes: preparing fluorescent microsphere standards: two latex microspheres of different sizes (5 μm and 10 μm in diameter) are conjugated with known concentrations of anti-mouse IgG antibody (as capture antibodies), and five gradient concentrations (10 μm, 10 μm, and 10 μm) are prepared by adjusting the fluorescence intensity of the microspheres (e.g., PE labeling). 3 -10 7 Standards were prepared in molecule / μL. During sample testing, 10 μL of microsphere standard was added to the single-cell suspension and co-incubated with cell antibodies. Fluorescence signals of microspheres and cells were collected simultaneously by flow cytometry. The fluorescence signal of cells was converted into absolute protein molecule number by fitting a standard curve (microsphere fluorescence intensity vs. known molecule number) (formula: protein molecule number = (cell fluorescence intensity - negative control) / standard curve slope).

[0080] Background Correction and False Negative Control: Negative Control Setup: Use unlabeled antibody isotype control cells (e.g., add mouse IgG isotype antibody instead of specific antibody) to detect autofluorescence and non-specific binding signals. Define the background threshold as the mean fluorescence intensity of the negative control + 2SD. Signal Screening: Judge the protein fluorescence signal of each cell. If it is below the background threshold and the number of molecules is <100, it is judged as an abnormally low expression signal. Ensure the false negative rate is ≤3% by increasing the detection time (cell acquisition speed ≤500 events / second).

[0081] Overcoming the limitations of traditional flow cytometry, which can only detect relative fluorescence intensity, this method directly obtains the number of protein molecules in a single cell (e.g., Cyclin E molecules / cell) using the CBA method. This provides accurate dose-response relationship data for subsequent modeling (e.g., the correlation between Cyclin E molecule number and G1 / S phase transition rate). Through negative control background correction and acquisition parameter optimization, the false negative rate is reduced from over 10% in traditional methods to below 3%. This method is particularly suitable for detecting low-abundance protein expression in a small number of cell subpopulations (e.g., weakly elevated Cyclin E in early drug-resistant cells), avoiding the loss of key signals.

[0082] In some embodiments, generating bivariate data points of DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportion and protein expression level includes: normalizing the DNA content of a single cell, expressed as a multiple of 2C, with the number of molecules of cell cycle proteins D, E, A, and B1; using Z-score normalization to eliminate dimensional differences between different proteins; and clustering the two-dimensional scatter plot using the DBSCAN algorithm to identify and remove outliers to generate the bivariate data points.

[0083] Data normalization includes: DNA content normalization: The DNA content of a single cell is expressed as multiples of 2C (e.g., G1 phase = 2C, G2 / M phase = 4C, S phase = 2C to 4C continuous values), converted to a dimensionless value DNAnorm = (DNAcontent - 2C) / (4C - 2C), and its range is standardized to [0,1]; Protein expression normalization: The number of Cyclin D, E, A, and B1 molecules is normalized by Z-score, with the formula Proteinnorm = (Proteincount - μ) / σ, where μ and σ are the mean and standard deviation of the protein in the sample, eliminating dimensional differences between different proteins (e.g., the number of Cyclin E molecules is usually lower than that of Cyclin D, and they can be directly compared after normalization).

[0084] Outlier removal includes: constructing a two-dimensional scatter plot of normalized DNA content (X-axis) and Cyclin E expression (Y-axis), and performing clustering using the DBSCAN algorithm with parameters set as follows: neighborhood radius ε = 0.1 (normalization unit), minimum number of samples MinPts = 5; identifying outliers with densities below a threshold (such as abnormally high DNA signals caused by single-cell lysis or abnormally high expression points due to non-specific antibody binding), and controlling the removal ratio to 2%-5% of the total data volume to ensure the reliability of bivariate data points.

[0085] Z-score standardization transforms DNA content (physical quantity) and protein molecule number (biomass) into a unified dimension, facilitating subsequent correlation analysis and model fitting. For example, it allows for the direct calculation of the overlap ratio between Cyclin E overexpression (Z-score > 2) and S-phase cells (DNA norm 0.3-0.7). The DBSCAN algorithm, based on density clustering, eliminates outliers, avoiding interference from erroneous data caused by instrument noise or operational errors. In particular, it eliminates artifacts such as cell adhesion (abnormally high DNA content) or antibody aggregation (abnormally high protein signal), increasing the effective information ratio of bivariate data points to over 95%, thus laying a high-quality data foundation for subsequent modeling.

[0086] In some embodiments, the step of analyzing the time difference between cell cycle-related protein expression and DNA content changes based on the bivariate data points using a pseudo-time series algorithm to construct a time series function model reflecting the cell cycle process includes: using the expression change curve of cyclin E as the time axis reference, dividing the DNA content change process from 2C to 4C into three stages: G1 phase, S phase, and G2 / M phase; using a hidden Markov model to identify the state transition points between each stage; calculating the time delay between the protein expression peak and the DNA replication initiation point; and constructing the time series function model based on the time delay.

[0087] Using the expression curve of Cyclin E as a pseudo-timeline benchmark, the early G1 phase was defined as the low expression stage of Cyclin E (molecule number < 5000 / cell), the late G1 phase as the rapid increase stage of expression (5000-15000 / cell), and the S phase as the decline stage after the expression peak (> 15000 / cell to decrease to below 10000 / cell). The change of DNA content from 2C to 4C was divided into three stages: G1 phase (2C, DNA_norm = 0), S phase (2C-4C, DNA_norm = 0-1), and G2 / M phase (4C, DNA_norm = 1), establishing the correspondence between DNA content and cell cycle progression.

[0088] Hidden Markov Model (HMM) modeling includes: defining three hidden states: S1 (G1 phase), S2 (S phase), and S3 (G2 / M phase), with bivariate data of DNA_norm and Cyclin E_norm as observed variables; calculating the state transition probability matrix: inferring the state transition path of each cell on the pseudo-timeline using the Viterbi algorithm, identifying the G1 / S phase transition point (the transition point from S1 to S2, i.e., the starting point where DNA_norm begins to leave 2C) and the S / G2 / M phase transition point (the transition point from S2 to S3, i.e., the endpoint where DNA_norm reaches 4C); calculating the time delay: using the pseudo-time point corresponding to the peak expression of Cyclin E as Tpeak, the DNA replication initiation point (G1 / S transition point) as Tstart, and the time delay ΔT = Tpeak - Tstart, reflecting the temporal relationship between Cyclin E activation and DNA replication initiation.

[0089] The construction of the time series function model includes: establishing the function f(t) = [DNA(t), Cyclin E(t), Cyclin D / A / B1(t)], where t is the pseudo-time (normalized to 0-1 from the beginning of G1 phase to the end of G2 / M phase), fitting the curves of protein expression and DNA content with t through cubic spline interpolation, focusing on capturing the overlap between the exponential growth phase of Cyclin E in the late G1 phase and the rising phase of DNA norm.

[0090] By identifying state transition points using Hidden Markov Models (HMMs), the time difference between Cyclin E expression dynamics and DNA replication process was quantified for the first time (e.g., ΔT ≈ 0.1 pseudo-time units in normal cells, and ΔT < 0.05 in abnormal cells, indicating premature activation). This solves the problem that traditional methods cannot analyze "how long protein expression changes lead / lag DNA replication." Combined with the correlation between Cyclin E expression peak and DNA replication initiation point, molecular driving events of the G1 / S phase transition can be accurately identified (e.g., Cyclin E reaching its peak prematurely leads to an earlier transition node), providing temporal evidence for drug resistance mechanism research. For example, it can be demonstrated that Cyclin E overexpression accelerates abnormal transitions by shortening ΔT, leading to increased S-phase DNA replication pressure.

[0091] In some embodiments, the process of fitting the time series function model to generate a cell cycle dynamic curve, and making a preliminary judgment on the drug resistance and prognosis of breast cancer cells based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested S-phase cells in the cell cycle dynamic curve, includes: when the expression level of cyclin E at the G1 / S phase transition node in the cell cycle dynamic curve exceeds the normal threshold by 1.5 times, and the proportion of γ-H2AX positive cells in S-phase cells is >15%, the cell cycle dynamic curve is identified as being in the chemotherapy resistance risk group; when the cell cycle dynamic curve shows that the Pearson correlation coefficient between the expression level of cyclin E in G1 phase cells and the proportion of S-phase cells is >0.6, and after calibration with the Ki-67 index, the accuracy of the curve predicting a disease-free survival of <24 months is ≥90%, the cell cycle dynamic curve is identified as having a poor prognosis.

[0092] Chemotherapy resistance risk assessment includes: G1 / S transition node abnormalities: In the cell cycle dynamics curve, the Cyclin E expression threshold at the G1 / S transition point is defined as the mean of normal samples + 1.5 SD (calibrated using healthy breast cell samples). If ≥20% of cells in the tested sample have Cyclin E expression exceeding this threshold at the transition point, it is judged as "Cyclin E overexpression-driven abnormal transition"; S-phase DNA damage association: Simultaneous detection of γ-H2AX (DNA double-strand break marker) fluorescence signal, calculation of the proportion of γ-H2AX-positive cells in S-phase cells, when the proportion >15%, it indicates that S-phase replication pressure leads to cumulative DNA damage; Comprehensive assessment: When the above two conditions are met simultaneously, it is judged as a chemotherapy resistance risk group (e.g., a 3-fold increase in the probability of resistance to taxane drugs).

[0093] Poor prognosis assessment includes: Correlation analysis: Calculating the Pearson correlation coefficient between G1 phase Cyclin E expression and the proportion of S phase cells. If >0.6, it suggests that Cyclin E overexpression drives more cells into S phase, posing a risk of uncontrolled proliferation; Ki-67 index calibration: Detecting Ki-67 protein expression (a cell proliferation marker) through immunofluorescence staining, and combining the correlation coefficient model with the Ki-67 index to establish a disease-free survival (DFS) prediction formula: DFS = α × Cyclin E_G1 + β × S phase proportion + γ × Ki-67 + constant term, where α, β, and γ are regression coefficients (trained using clinical cohort data); Accuracy validation: In the validation cohort, if the model's accuracy in predicting DFS < 24 months is ≥90%, it is considered a poor prognosis, guiding intensive clinical treatment plans.

[0094] By combining three indicators—Cyclin E overexpression, G1 / S phase transition abnormalities, and DNA damage—a resistance determination chain of "molecular drive - cell cycle abnormality - functional impairment" is formed, improving accuracy by 40% compared to traditional single indicators (such as Ki-67). This is particularly specific for predicting CDK4 / 6 inhibitor resistance. Furthermore, correlation analysis between Cyclin E and the S phase ratio captures overall abnormalities in the cell cycle regulatory network. After calibration with Ki-67, the accuracy of predicting poor prognosis is increased to over 90%, addressing the insensitivity of traditional prognostic indicators (such as TNM staging) to cell cycle abnormalities. This provides a precise basis for personalized treatment (e.g., recommending clinical trials of new drugs targeting the Cyclin E-CDK2 pathway).

[0095] See also Figure 2 , Figure 2 This is a schematic flowchart of a bivariate simulation analysis method based on DNA content and cell cycle-related proteins provided in one embodiment of this application. The execution device for the method is the control module of the bivariate simulation analysis system based on DNA content and cell cycle-related proteins provided in any embodiment of this application.

[0096] like Figure 2 As shown, the provided method includes steps S101 to S103. The control module can be a handheld terminal, a laptop computer, a wearable device, or a robot, etc. It is used to implement steps S101 to S103 and their corresponding embodiments.

[0097] Step S101. Obtain bivariate data points of DNA content and protein expression generated by the bivariate detection module; the bivariate data points are used by the bivariate detection module to detect the DNA content of breast cancer tumor cell samples using quantitative staining flow cytometry to determine the relative proportions of the G1, S, and G2 / M phases in which the breast cancer tumor cell samples are located; it is also used to detect the protein expression levels of cyclins D, E, A, and B1 in the breast cancer tumor cell samples, and generate bivariate data points of DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportions and protein expression levels;

[0098] Step S102. Based on the bivariate data points, the time difference between the expression of cell cycle-related proteins and the change in DNA content is analyzed by a pseudo-time series algorithm, and a time series function model reflecting the cell cycle process is constructed. The time series function model uses the change in the expression of cyclin E as the core variable driving the G1 / S phase transition.

[0099] Step S103. Fit the time series function model to generate a cell cycle dynamic curve. Based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested S phase cells in the cell cycle dynamic curve, the drug resistance and prognosis of breast cancer cells are pre-judged.

[0100] In some embodiments, the flow cytometry method using quantitative staining to detect the DNA content of breast cancer tumor cell samples to determine the relative proportions of the G1, S, and G2 / M phases of the breast cancer tumor cell samples includes: pretreating the samples with Hoechst 33342 combined with RNase, detecting DNA fluorescence intensity by ultraviolet laser excitation, introducing normal diploid cells as a calibration reference when performing peak shape fitting based on ModFit software, calibrating the DNA content of the G1 phase to a 2C value, the G2 / M phase to a 4C value, and defining the S phase as a continuous distribution between 2C and 4C, while excluding adherent cells and fragmented cells through pulse shape analysis, so that the detection error of the proportion of cells in each cycle is ≤5%.

[0101] In some embodiments, the detection of the protein expression levels of cyclins D, E, A, and B1 in breast cancer tumor cell samples includes: real-time acquisition of fluorescence signals of antibodies against each protein in a single cell using a multi-parameter detection channel of flow cytometry; conversion of fluorescence signals into protein molecule numbers using microsphere-based absolute counting (CBA); and simultaneous addition of a negative control for background correction during the detection process to ensure that the false negative rate of abnormally low expression signals is ≤3%.

[0102] In some embodiments, generating bivariate data points of DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportion and protein expression level includes: normalizing the DNA content of a single cell, expressed as a multiple of 2C, with the number of molecules of cell cycle proteins D, E, A, and B1; using Z-score normalization to eliminate dimensional differences between different proteins; and clustering the two-dimensional scatter plot using the DBSCAN algorithm to identify and remove outliers to generate the bivariate data points.

[0103] In some embodiments, the step of analyzing the time difference between cell cycle-related protein expression and DNA content changes based on the bivariate data points using a pseudo-time series algorithm to construct a time series function model reflecting the cell cycle process includes: using the expression change curve of cyclin E as the time axis reference, dividing the DNA content change process from 2C to 4C into three stages: G1 phase, S phase, and G2 / M phase; using a hidden Markov model to identify the state transition points between each stage; calculating the time delay between the protein expression peak and the DNA replication initiation point; and constructing the time series function model based on the time delay.

[0104] In some embodiments, the process of fitting the time series function model to generate a cell cycle dynamic curve, and making a preliminary judgment on the drug resistance and prognosis of breast cancer cells based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested S-phase cells in the cell cycle dynamic curve, includes: when the expression level of cyclin E at the G1 / S phase transition node in the cell cycle dynamic curve exceeds the normal threshold by 1.5 times, and the proportion of γ-H2AX positive cells in S-phase cells is >15%, the cell cycle dynamic curve is identified as being in the chemotherapy resistance risk group; when the cell cycle dynamic curve shows that the Pearson correlation coefficient between the expression level of cyclin E in G1 phase cells and the proportion of S-phase cells is >0.6, and after calibration with the Ki-67 index, the accuracy of the curve predicting a disease-free survival of <24 months is ≥90%, the cell cycle dynamic curve is identified as having a poor prognosis.

[0105] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the bivariate simulation analysis method and each step based on DNA content and cell cycle-related proteins described above can be referred to the corresponding process in the embodiments of the bivariate simulation analysis system based on DNA content and cell cycle-related proteins described above, and will not be repeated here.

[0106] This application also provides a bivariate simulation analysis device based on DNA content and cell cycle-related proteins. This device is used to perform the steps of the bivariate simulation analysis methods based on DNA content and cell cycle-related proteins shown in the above embodiments. This device can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, laptop computer, wearable device, or robot.

[0107] The bivariate simulation analysis device based on DNA content and cell cycle-related proteins includes:

[0108] The data acquisition unit is used to acquire bivariate data points of DNA content and protein expression generated by the bivariate detection module. The bivariate data points are used by the bivariate detection module to detect the DNA content of breast cancer tumor cell samples using quantitative staining flow cytometry to determine the relative proportion of the G1, S, and G2 / M phases in which the breast cancer tumor cell samples are located. The unit is also used to detect the protein expression levels of cell cycle proteins D, E, A, and B1 in breast cancer tumor cell samples, and generate bivariate data points of DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportions and protein expression levels.

[0109] The variable generation unit is used to analyze the time difference between the expression of cell cycle-related proteins and the change in DNA content based on the bivariate data points using a pseudo-time series algorithm, and to construct a time series function model that reflects the cell cycle process. The time series function model uses the change in the expression of cyclin E as the core variable driving the G1 / S phase transition.

[0110] The proportion generation unit is used to fit the time series function model to generate a cell cycle dynamic curve. Based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested S phase cells in the cell cycle dynamic curve, the drug resistance and prognosis of breast cancer cells can be predicted in advance.

[0111] It should be noted that those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the bivariate simulation analysis device and its units based on DNA content and cell cycle-related proteins described above can be referred to the corresponding process in the embodiments of the bivariate simulation analysis method based on DNA content and cell cycle-related proteins described above, and will not be repeated here.

[0112] The aforementioned bivariate simulation analysis method based on DNA content and cell cycle-related proteins is implemented as a computer program that can run on the aforementioned device.

[0113] See also Figure 3 , Figure 3 This is a schematic block diagram of the control module provided in an embodiment of this application. The control module includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and internal memory.

[0114] The storage medium may store operating devices and computer programs. The computer program includes program instructions that, when executed, cause the processor to perform an embodiment of any bivariate simulation analysis method based on DNA content and cell cycle-related proteins.

[0115] The processor provides computing and control capabilities to support the operation of the entire control module.

[0116] Internal memory provides an environment for the execution of computer programs in non-volatile storage media. When executed by a processor, the computer program enables the processor to perform any bivariate simulation analysis system method based on DNA content and cell cycle-related proteins.

[0117] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the solution of this application and does not constitute a limitation on the terminal to which the solution of this application is applied. The specific control module may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0118] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0119] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps:

[0120] The system acquires bivariate data points on DNA content and protein expression generated by a bivariate detection module. These bivariate data points are then used by the bivariate detection module to detect the DNA content of breast cancer cell samples using quantitative staining flow cytometry, in order to determine the relative proportions of the G1, S, and G2 / M phases in which the breast cancer cell samples are located. The system also detects the protein expression levels of cyclins D, E, A, and B1 in the breast cancer cell samples, and generates bivariate data points on DNA content and protein expression for each breast cancer cell sample based on these relative proportions and protein expression levels.

[0121] Based on the bivariate data points, the time difference between the expression of cell cycle-related proteins and changes in DNA content was analyzed using a pseudo-time series algorithm. A time series function model reflecting the cell cycle process was constructed, in which the expression change of cyclin E was used as the core variable driving the G1 / S phase transition.

[0122] The time series function model is fitted to generate a cell cycle dynamic curve. Based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested cells in the S phase, the drug resistance and prognosis of breast cancer cells can be predicted in advance.

[0123] In some embodiments, the flow cytometry method using quantitative staining to detect the DNA content of breast cancer tumor cell samples to determine the relative proportions of the G1, S, and G2 / M phases of the breast cancer tumor cell samples includes: pretreating the samples with Hoechst 33342 combined with RNase, detecting DNA fluorescence intensity by ultraviolet laser excitation, introducing normal diploid cells as a calibration reference when performing peak shape fitting based on ModFit software, calibrating the DNA content of the G1 phase to a 2C value, the G2 / M phase to a 4C value, and defining the S phase as a continuous distribution between 2C and 4C, while excluding adherent cells and fragmented cells through pulse shape analysis, so that the detection error of the proportion of cells in each cycle is ≤5%.

[0124] In some embodiments, the detection of the protein expression levels of cyclins D, E, A, and B1 in breast cancer tumor cell samples includes: real-time acquisition of fluorescence signals of antibodies against each protein in a single cell using a multi-parameter detection channel of flow cytometry; conversion of fluorescence signals into protein molecule numbers using microsphere-based absolute counting (CBA); and simultaneous addition of a negative control for background correction during the detection process to ensure that the false negative rate of abnormally low expression signals is ≤3%.

[0125] In some embodiments, generating bivariate data points of DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportion and protein expression level includes: normalizing the DNA content of a single cell, expressed as a multiple of 2C, with the number of molecules of cell cycle proteins D, E, A, and B1; using Z-score normalization to eliminate dimensional differences between different proteins; and clustering the two-dimensional scatter plot using the DBSCAN algorithm to identify and remove outliers to generate the bivariate data points.

[0126] In some embodiments, the step of analyzing the time difference between cell cycle-related protein expression and DNA content changes based on the bivariate data points using a pseudo-time series algorithm to construct a time series function model reflecting the cell cycle process includes: using the expression change curve of cyclin E as the time axis reference, dividing the DNA content change process from 2C to 4C into three stages: G1 phase, S phase, and G2 / M phase; using a hidden Markov model to identify the state transition points between each stage; calculating the time delay between the protein expression peak and the DNA replication initiation point; and constructing the time series function model based on the time delay.

[0127] In some embodiments, the process of fitting the time series function model to generate a cell cycle dynamic curve, and making a preliminary judgment on the drug resistance and prognosis of breast cancer cells based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested S-phase cells in the cell cycle dynamic curve, includes: when the expression level of cyclin E at the G1 / S phase transition node in the cell cycle dynamic curve exceeds the normal threshold by 1.5 times, and the proportion of γ-H2AX positive cells in S-phase cells is >15%, the cell cycle dynamic curve is identified as being in the chemotherapy resistance risk group; when the cell cycle dynamic curve shows that the Pearson correlation coefficient between the expression level of cyclin E in G1 phase cells and the proportion of S-phase cells is >0.6, and after calibration with the Ki-67 index, the accuracy of the curve predicting a disease-free survival of <24 months is ≥90%, the cell cycle dynamic curve is identified as having a poor prognosis.

[0128] It should be noted that those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the processor described above can be referred to the corresponding process in the method embodiments of the above embodiments, and will not be repeated here.

[0129] The embodiments of this application also provide a computer-readable storage medium storing a computer program, the computer program including program instructions, and the processor executing the program instructions to implement the steps of the bivariate simulation analysis method based on DNA content and cell cycle-related proteins provided in the above embodiments of this application.

[0130] The computer-readable storage medium can be an internal storage unit of the control module described in the foregoing embodiments, such as the hard disk or memory of the control module. Alternatively, the computer-readable storage medium can be an external storage device of the control module, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the control module.

[0131] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A bivariate simulation analysis system based on DNA content and cell cycle-related proteins, characterized in that, include: The sample processing module is used to acquire breast cancer tumor cell samples and perform fluorescent labeling on the breast cancer tumor cell samples. The labeling of cell cycle-related proteins includes specific fluorescent antibodies for cyclin D, cyclin E, cyclin A and cyclin B1, and the fluorescent labeling signal of cyclin E is used as the main monitoring indicator. The bivariate detection module is used to detect the DNA content of breast cancer tumor cell samples using quantitative staining flow cytometry to determine the relative proportions of the G1, S, and G2 / M phases in which the breast cancer tumor cell samples are located; it is also used to detect the protein expression levels of cyclin D, E, A, and B1 in breast cancer tumor cell samples, and generate bivariate data points of DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportions and protein expression levels. The control module is used to analyze the time difference between changes in cell cycle-related protein expression and DNA content based on the bivariate data points using a pseudo-time series algorithm, and construct a time series function model reflecting the cell cycle process. The time series function model uses the change in cyclin E expression as the core variable driving the G1 / S phase transition. The time series function model is fitted to generate a cell cycle dynamic curve. Based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested S-phase cells in the cell cycle dynamic curve, the drug resistance and prognosis of breast cancer cells are predicted in advance.

2. The system according to claim 1, wherein: The fluorescent labeling treatment of the sample includes: After permeabilizing breast cancer tumor cell samples, fluorescently labeled antibodies against cyclin D, E, A, and B1 were added respectively. The antibody against cyclin E was labeled with a near-infrared fluorescent group, and its fluorescence signal intensity was 2-3 orders of magnitude higher than that of the fluorescent groups labeled with other protein antibodies. The fluorescence emission spectra overlap of each antibody was ≤10%, so as to achieve high-precision resolution of multi-protein labeling signals in single cells. The near-infrared fluorescent group has an excitation wavelength of 780 nm and an emission wavelength of 800-850 nm.

3. The system according to claim 1, wherein: The method employs quantitative staining flow cytometry to detect the DNA content of breast cancer tumor cell samples, in order to determine the relative proportions of breast cancer tumor cell samples in the G1, S, and G2 / M phases, including: Samples were pretreated using Hoechst 33342 combined with RNase. DNA fluorescence intensity was detected by UV laser excitation. When performing peak shape fitting based on ModFit software, normal diploid cells were introduced as a calibration reference. The DNA content in G1 phase was calibrated to 2C value, G2 / M phase to 4C value, and S phase was defined as a continuous distribution between 2C and 4C. At the same time, pulse shape analysis was used to exclude adherent cells and fragmented cells, so that the detection error of the proportion of cells in each cycle was ≤5%.

4. The system according to claim 1, characterized in that, The detection of the protein expression levels of cyclins D, E, A, and B1 in breast cancer tumor cell samples includes: The fluorescence signals of antibodies against various proteins in individual cells are acquired in real time using the multi-parameter detection channel of flow cytometry. The fluorescence signals are converted into protein molecule numbers using the microsphere-based absolute counting method (CBA). A negative control is added simultaneously during the detection process for background correction, so that the false negative rate of abnormally low expression signals is ≤3%.

5. The system according to claim 1, wherein: The process of generating bivariate data points for DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportions and protein expression levels includes: The DNA content of a single cell, expressed as a multiple of 2C, was normalized to the number of molecules of cell cycle proteins D, E, A, and B1, and Z-score normalization was used to eliminate dimensional differences between different proteins. The two-dimensional scatter plot is clustered using the DBSCAN algorithm to identify and remove outliers, generating the bivariate data points.

6. The system according to claim 1, characterized in that, Based on the bivariate data points, the time difference between changes in cell cycle-related protein expression and DNA content is analyzed using a pseudo-time series algorithm to construct a time series function model reflecting the cell cycle process, including: Using the expression change curve of cyclin E as the time axis benchmark, the process of DNA content change from 2C to 4C is divided into three stages: G1 phase, S phase, and G2 / M phase. Hidden Markov model is used to identify the state transition points between each stage and calculate the time delay between the protein expression peak and the DNA replication initiation point. The time series function model is constructed based on the time delay.

7. The system according to claim 1, characterized in that, The process of fitting the time series function model to generate a cell cycle dynamic curve, and then using the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested S-phase cells in the cell cycle dynamic curve to pre-judge the drug resistance and prognosis of breast cancer cells, includes: When the expression level of cyclin E at the G1 / S phase transition node in the cell cycle dynamic curve exceeds the normal threshold by 1.5 times, and the proportion of γ-H2AX positive cells in S phase cells is >15%, it is identified as a chemotherapy resistance risk group. When the cell cycle dynamic curve shows a Pearson correlation coefficient >0.6 between the expression level of cyclin E in G1 phase cells and the proportion of cells in S phase, and after calibration with the Ki-67 index, the accuracy of the curve-predicted disease-free survival of <24 months is ≥90%, it is considered a poor prognosis.

8. A bivariate simulation analysis method based on DNA content and cell cycle-related proteins, characterized in that, A control module applied to a bivariate simulation analysis system based on DNA content and cell cycle-related proteins as described in any one of claims 1-7, the method comprising: The system acquires bivariate data points on DNA content and protein expression generated by a bivariate detection module. These bivariate data points are then used by the bivariate detection module to detect the DNA content of breast cancer cell samples using quantitative staining flow cytometry, in order to determine the relative proportions of the G1, S, and G2 / M phases in which the breast cancer cell samples are located. The system also detects the protein expression levels of cyclins D, E, A, and B1 in the breast cancer cell samples, and generates bivariate data points on DNA content and protein expression for each breast cancer cell sample based on these relative proportions and protein expression levels. Based on the bivariate data points, the time difference between the expression of cell cycle-related proteins and changes in DNA content was analyzed using a pseudo-time series algorithm. A time series function model reflecting the cell cycle process was constructed, in which the expression change of cyclin E was used as the core variable driving the G1 / S phase transition. The time series function model is fitted to generate a cell cycle dynamic curve. Based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested cells in the S phase, the drug resistance and prognosis of breast cancer cells can be predicted in advance.

9. A bivariate simulation analysis device based on DNA content and cell cycle-related proteins, characterized in that, A control module for a bivariate simulation analysis system based on DNA content and cell cycle-related proteins as described in any one of claims 1-7, the device comprising: The data acquisition unit is used to acquire bivariate data points of DNA content and protein expression generated by the bivariate detection module. The bivariate data points are used by the bivariate detection module to detect the DNA content of breast cancer tumor cell samples using quantitative staining flow cytometry to determine the relative proportion of the G1, S, and G2 / M phases in which the breast cancer tumor cell samples are located. The unit is also used to detect the protein expression levels of cell cycle proteins D, E, A, and B1 in breast cancer tumor cell samples, and generate bivariate data points of DNA content and protein expression for each breast cancer tumor cell sample based on the relative proportions and protein expression levels. The variable generation unit is used to analyze the time difference between the expression of cell cycle-related proteins and the change in DNA content based on the bivariate data points using a pseudo-time series algorithm, and to construct a time series function model that reflects the cell cycle process. The time series function model uses the change in the expression of cyclin E as the core variable driving the G1 / S phase transition. The proportion generation unit is used to fit the time series function model to generate a cell cycle dynamic curve. Based on the abnormal characteristics of the G1 / S phase transition node, the overexpression level of cyclin E, and the proportion of abnormally arrested S phase cells in the cell cycle dynamic curve, the drug resistance and prognosis of breast cancer cells can be predicted in advance.

10. A control module, characterized in that, The control module includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and, in executing the computer program, implement the method as described in claim 8.

Images