Quantitative optimization analysis method and system applied to nucleic acid modification site

Abstract

The application provides a quantitative optimization analysis method and system applied to a nucleic acid modification site, and comprises the following steps: acquiring real-time fluorescence detection data and amplification configuration characteristics of each reaction well of qPCR, constructing a detection system combination key according to the amplification configuration characteristics, performing feature grouping and median absolute deviation clipping to obtain an initial reference object set; constructing a double-thermodynamic boundary set according to the initial reference object set, performing soft membership weight calculation and modification rate estimation to obtain an initial quantitative result set; calculating the amplification credibility of each reaction well, and performing credibility labeling on the initial quantitative result set to obtain a labeled quantitative result set; iteratively updating the initial reference object set by using the labeled quantitative result set, and re-performing soft membership weight calculation, modification rate estimation and iterative convergence analysis to obtain a modification site quantitative result set; and the application can improve the accuracy and reliability of quantitative analysis of the nucleic acid modification site.

Description

Technical Field

[0001] This invention relates to the field of DNA detection and molecular quantitative analysis technology, and in particular to a quantitative optimization analysis method and system applied to nucleic acid modification sites. Background Technology

[0002] In the fields of epigenetics and precision medicine research, precise quantitative analysis of nucleic acid modification sites is a crucial technological foundation for revealing gene expression regulation mechanisms, disease development processes, and conducting early screening and personalized treatment. Among these methods, nucleic acid modification detection based on real-time quantitative polymerase chain reaction (qPCR) has been widely applied to the quantitative analysis of various nucleic acid modification sites, such as DNA methylation and RNA methylation, due to its advantages of high detection sensitivity, short detection cycle, mature equipment, and ease of deployment. In recent years, with the development of modification-specific primers, dual-probe detection systems, and melting curve analysis techniques, the use of amplification curves combined with melting curves for quantitative analysis of modification sites has gradually become an important technological direction for nucleic acid modification detection.

[0003] However, existing methods for quantitative analysis of nucleic acid modification sites based on qPCR platforms still have significant technical limitations. First, current methods typically establish reference systems using a uniform background baseline, fixed fluorescence thresholds, or a small number of negative controls. This makes it difficult to fully consider the impact of different targets, fluorescence channels, and flanking sequences of different modification motifs on amplification efficiency, background fluorescence, and melting behavior, easily leading to baseline confusion and background correction errors between different detection systems. Second, most existing methods rely on Ct values, melting peak temperatures, or fixed thresholds for hard determination of methylation or unmethylation of samples. They lack effective characterization capabilities for overlapping double peaks, broad peaks, or continuous transition states within the melting rain region, failing to fully utilize the overall thermodynamic information of the melting curve, and easily causing errors in modification rate estimation. In addition, qPCR experiments are generally affected by factors such as fluctuations in amplification efficiency, differences in background between wells, non-specific amplification, contaminated wells, and amplification failure wells. Existing methods lack a comprehensive quality evaluation mechanism that combines amplification curve quality, melting curve morphology, and replicate consistency. They also lack the self-calibration capability to optimize the reference system based on quantitative results, which makes the reference baseline susceptible to contamination by abnormal samples, limiting the accuracy, robustness, and batch consistency of nucleic acid modification site quantitative analysis results. Summary of the Invention

[0004] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a quantitative optimization analysis method and system for nucleic acid modification sites. It has the advantages of equivalence class grouping of detection system, robust baseline estimation of median absolute deviation, soft membership weight quantification of dual thermodynamic boundaries, multi-dimensional amplification reliability assessment, and self-consistent iterative reference system calibration. It solves the problems of low accuracy, insufficient sensitivity, and poor cross-batch consistency in the quantitative analysis of modification sites caused by the difficulty in adapting the global baseline to the target and the loss of information in the melting overlap region due to hard threshold cutting in the existing nucleic acid modification qPCR analysis process.

[0005] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: This invention provides a quantitative optimization analysis method for nucleic acid modification sites, comprising the following steps: Real-time fluorescence detection data and amplification configuration characteristics of each reaction well in qPCR are obtained. Based on the amplification configuration characteristics, detection system combination bonds are constructed and feature groups are performed to obtain fluorescence equivalence class sets. The fluorescence equivalence class sets are then truncated by median absolute deviation to obtain an initial reference object set. Based on the initial reference object set, a dual thermodynamic boundary set corresponding to the fluorescence equivalence class set is constructed, and soft membership weights and modification rates are calculated and estimated on the real-time fluorescence detection data to obtain an initial quantitative result set. The amplification confidence level of each reaction well is calculated based on the real-time fluorescence detection data. The amplification confidence level is then used to label the initial quantitative result set to obtain a labeled quantitative result set. The initial reference object set is updated using the labeled quantitative result set to obtain a standard reference object set; Based on the aforementioned standard reference object set, the soft membership weights were recalculated, the modification rate was estimated, and iterative convergence analysis was performed to obtain a quantitative result set of modification sites.

[0006] According to a preferred embodiment of the present invention, the process of obtaining real-time fluorescence detection data and amplification configuration characteristics of each qPCR reaction well includes: According to the preset experimental well layout table, add the pre-configured nucleic acid samples to be tested, template-free control samples, unmodified control samples, and internal control samples into the corresponding reaction wells; Simultaneous PCR cycle amplification was performed on each reaction well, and the real-time fluorescence signal of each reaction well was collected during the cycle amplification to obtain the amplification curve; Temperature data and fluorescence signals of each reaction well were collected after the cycle amplification was completed, and melting curves were constructed. The amplification curves and melting curves were then combined to form real-time fluorescence detection data. The target identifier, probe fluorescence channel identifier, modified motif flanking sequence type, and well purpose marker corresponding to each reaction well are extracted from the experimental well layout table and the preset amplification system configuration file, and then compiled into amplification configuration features.

[0007] According to another preferred embodiment of the present invention, the process of constructing detection system combination bonds and performing feature grouping based on the amplification configuration features includes: The target identifier, probe fluorescence channel identifier, and modified motif flanking sequence type in the amplification configuration features are concatenated into a detection system combination bond; The real-time fluorescence detection data corresponding to each reaction well are grouped according to the combination key of the detection system to obtain the fluorescence equivalence class set.

[0008] According to another preferred embodiment of the present invention, the process of performing median absolute bias clipping on the fluorescence equivalence class set includes: One by one, the fluorescence equivalence classes in the fluorescence equivalence class set are selected as target equivalence classes. Real-time fluorescence detection data corresponding to well positions marked as template-free control wells and unmodified control wells are selected from the target equivalence classes as control fluorescence detection data. The fluorescence intensity values ​​corresponding to the control fluorescence detection data in the early stage of cyclic amplification are compiled into a control fluorescence value set; The median and median absolute deviation of the control fluorescence value set were calculated and compiled into a statistical summary; Based on the statistical summary, outlier values ​​in the control fluorescence value set are removed to obtain the clean control fluorescence value set; The truncated mean value of the fluorescence intensity values ​​in the clean control fluorescence value set is calculated to obtain the local baseline fluorescence value of the target equivalence class; Calculate the reference dispersion corresponding to the clean control fluorescence value set, encapsulate the local baseline fluorescence value and the reference dispersion as an initial reference object, and aggregate the initial reference objects of all target equivalence classes into an initial reference object set.

[0009] According to another preferred embodiment of the present invention, the process of constructing the dual thermodynamic boundary set corresponding to the fluorescence equivalence class set based on the initial reference object set includes: Each fluorescence equivalence class in the fluorescence equivalence class set is selected as the target equivalence class, and the initial reference object corresponding to the target equivalence class in the initial reference object set is taken as the target initial object. The melting curve set is extracted from the real-time fluorescence detection data of the target equivalence class, and the local baseline fluorescence value and the corresponding reference scattering are extracted from the target initial object. The melting curves are superimposed and averaged to calculate the baseline melting spectrum. The peaks of the baseline melting spectrum are located based on the local baseline fluorescence value and the reference dispersion to obtain the boundary between the characteristic peaks of the methylated probe and the characteristic peaks of the unmethylated probe. The temperature range between the boundary of the methylated probe characteristic peak and the boundary of the unmethylated probe characteristic peak is defined as the melting rain region. The boundary of the methylated probe characteristic peak, the boundary of the unmethylated probe characteristic peak, and the melting rain region are encapsulated as a dual thermodynamic boundary, and all dual thermodynamic boundaries are aggregated into a dual thermodynamic boundary set.

[0010] According to another preferred embodiment of the present invention, the process of calculating soft membership weights and estimating modification rate for the real-time fluorescence detection data includes: The reaction wells marked as sample wells in the amplification configuration features are used as target reaction wells, and the real-time fluorescence detection data corresponding to the target reaction wells are used as target fluorescence detection data. The dual thermodynamic boundary corresponding to the target fluorescence detection data is taken as the target dual boundary; The target temperature-fluorescence derivative spectrum corresponding to the melting curve is extracted from the target fluorescence detection data. The soft membership weight is calculated at each temperature point on the target temperature-fluorescence derivative spectrum based on the target double boundary. The modification rate is estimated based on the soft membership weight to obtain the initial modification rate. The initial modification rate of the target reaction wells is encapsulated into an initial quantitative result, and the initial quantitative results of all target reaction wells are aggregated into an initial quantitative result set.

[0011] According to another preferred embodiment of the present invention, the process of calculating the amplification confidence of each reaction well based on the real-time fluorescence detection data includes: The reaction wells marked as sample wells in the amplification configuration features are designated as target reaction wells, the real-time fluorescence detection data corresponding to the target reaction wells are designated as target fluorescence detection data, and the fluorescence equivalence class corresponding to the target fluorescence detection data in the fluorescence equivalence class set is designated as the target equivalence class. The exponential phase rise slope is extracted from the amplification curve of the target fluorescence detection data, and the slope anomaly analysis is performed on the exponential phase rise slope according to the target equivalence class to obtain the exponential phase slope anomaly degree. The half-peak width of the main peak is extracted from the melting curve of the target fluorescence detection data, and the half-peak width of the main peak is subjected to threshold anomaly analysis based on the target equivalence class to obtain the peak width anomaly degree. Select the Ct values ​​corresponding to the sample complex group to which the target reaction well belongs in the target equivalence class, aggregate them into a target Ct value group, and calculate the Ct value dispersion corresponding to the target Ct value group; The amplification credibility is obtained by comprehensively evaluating the exponential slope anomaly, peak width anomaly, and Ct value dispersion.

[0012] According to another preferred embodiment of the present invention, the process of using the amplified confidence level to label the initial quantitative result set includes: Each initial quantitative result in the initial quantitative result set is selected as the target initial quantitative result, and the initial modification rate of the target initial quantitative result is taken as the target initial modification rate. The amplification confidence level of the reaction well corresponding to the initial quantitative result of the target is taken as the target confidence level; The target initial modification rate is labeled using the target credibility and then jointly encapsulated to obtain labeled quantitative results. All labeled quantitative results are then aggregated into a labeled quantitative result set.

[0013] According to another preferred embodiment of the present invention, the process of updating the initial reference object set using the labeled quantitative result set includes: One by one, the fluorescence equivalence classes in the fluorescence equivalence class set are selected as target equivalence classes. The labeled quantitative results corresponding to the target equivalence classes in the labeled quantitative results set are collected as target labeled results into a target labeled result group. The initial reference objects corresponding to the target equivalence classes are extracted from the initial reference object set as the current reference objects. Traverse the target annotation results in the target annotation result group, and determine one by one whether the amplification confidence of the target annotation result is less than a preset confidence threshold, or whether the initial modification rate of the target annotation result is greater than a preset modification rate threshold. If so, the real-time fluorescence detection data corresponding to the target annotation result will be removed from the target equivalence class as abnormal fluorescence detection data; If not, return to the step of selecting the target annotation results in the target annotation result group one by one; Until all target standard results in the target annotation result group have been traversed, the target equivalence class after removing all abnormal fluorescence detection data will be taken as the target standard equivalence class; The local baseline fluorescence value and reference scatter of the target standard equivalence class are recalculated, the current reference object is updated, a standard reference object is obtained, and all standard reference objects are aggregated into a standard reference object set.

[0014] To achieve at least one of the above-mentioned objectives, the present invention further provides a quantitative optimization analysis system for nucleic acid modification sites, the system comprising an inter-class trimming module, an initial quantification module, a confidence weighting module, an object update module, and a result iteration module; The inter-class trimming module acquires real-time fluorescence detection data and amplification configuration features of each qPCR reaction well, constructs detection system combination bonds based on the amplification configuration features and performs feature grouping to obtain a fluorescence equivalence class set, and trims the fluorescence equivalence class set by median absolute deviation to obtain an initial reference object set; The initial quantification module constructs a dual thermodynamic boundary set corresponding to the fluorescence equivalence class set based on the initial reference object set, and performs soft membership weight calculation and modification rate estimation on the real-time fluorescence detection data to obtain the initial quantification result set. The confidence weighting module calculates the amplification confidence of each reaction well based on the real-time fluorescence detection data, and uses the amplification confidence to label the initial quantitative result set to obtain a labeled quantitative result set. The object update module updates the initial reference object set using the labeled quantitative result set to obtain a standard reference object set; The result iteration module recalculates the soft membership weights, estimates the modification rate, and performs iterative convergence analysis based on the standard reference object set to obtain a quantitative result set of modification sites.

[0015] (III) Beneficial Effects Compared with existing technologies, this invention provides a quantitative optimization analysis method for nucleic acid modification sites, which has the following beneficial effects: This quantitative optimization analysis method applied to nucleic acid modification sites acquires real-time fluorescence detection data and amplification configuration characteristics, and uses detection system combination bonds to complete the characteristic grouping of different detection systems. This avoids mutual interference between different targets, different modification types, and different fluorescence channels, and avoids baseline crosstalk and statistical confusion between different detection systems. By screening early cycle data of control wells within equivalence classes and using median absolute deviation for robust outlier pruning, it can effectively remove occasional noise, contaminated wells, and abnormal background signals, and improve the stability of local background estimation.

[0016] This quantitative optimization analysis method for nucleic acid modification sites constructs dual thermodynamic boundaries corresponding to each fluorescence equivalence class based on an initial reference set and analyzes them in conjunction with the thermodynamic characteristics of the melting curve. It fully utilizes the amplification process and melting information to achieve continuous determination of modified and unmodified templates. Through a soft membership weight calculation method for the melting rain zone, it avoids boundary misjudgments caused by the traditional hard threshold method, improving the stability of determination when the modification ratio changes continuously. This enhances the robustness and accuracy of quantitative optimization of nucleic acid modification sites. By extracting the exponential rise slope, main peak half-width, and Ct dispersion of the test samples, it combines the stability of PCR amplification, the specificity of amplification products, and the reproducibility of the technique for joint reliability analysis. This comprehensively evaluates the reliability of the detection results of each reaction well, effectively identifies and labels abnormal or low-quality data, and improves the stability and accuracy of quantitative analysis results for nucleic acid modification sites.

[0017] This quantitative optimization analysis method applied to nucleic acid modification sites updates the initial reference set using a labeled quantitative result set. This removes amplification data from positive reaction wells with low confidence or high modification rates from the calculation of the reference set, thereby gradually bringing the local baseline fluorescence values ​​in the reference set closer to the fluorescence values ​​of the true unmodified state. This reduces the contamination effect of positive signals on the negative reference system caused by sample cross-contamination or incomplete chemical transformation, ensuring the accuracy and consistency of the quantitative results. Through iterative soft membership weight calculation and modification rate estimation, it ensures that the reference set gradually approaches the fluorescence values ​​of the true unmodified state in multiple iterations. By combining root mean square error and iteration number threshold for convergence analysis, it ensures that the output quantitative result set of modification sites comes from a well-purified and stable reference set, improving the quantitative accuracy and batch reproducibility of low-abundance modified samples in complex backgrounds. Attached Figure Description

[0018] Figure 1 The diagram shows a flowchart of a quantitative optimization analysis method for nucleic acid modification sites according to the present invention.

[0019] Figure 2 The diagram shows the method of constructing the dual thermodynamic boundary set corresponding to the fluorescence equivalence class set according to the present invention.

[0020] Figure 3 The diagram shows the flowchart of the method for calculating the amplification reliability of each reaction well according to the present invention.

[0021] Figure 4 The diagram shown is a structural diagram of a quantitative optimization analysis system for nucleic acid modification sites according to the present invention. Detailed Implementation

[0022] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious modifications will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0023] It is understood that the term "a" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0024] Example 1: Please combine Figure 1 This invention discloses a quantitative optimization analysis method for nucleic acid modification sites, the method comprising the following steps: S1. Obtain real-time fluorescence detection data and amplification configuration characteristics of each qPCR reaction well. Construct detection system combination bonds based on the amplification configuration characteristics and perform feature grouping to obtain a fluorescence equivalence class set. Perform median absolute deviation trimming on the fluorescence equivalence class set to obtain an initial reference object set.

[0025] qPCR refers to Quantitative Real-time Polymerase Chain Reaction. (Reaction), a reaction well refers to a unit set in a real-time quantitative PCR reaction plate for independent amplification reactions. Each reaction well corresponds to a nucleic acid sample to be tested, a template-free control sample, or an unmodified control sample, and has a unique well location identifier. Real-time fluorescence detection data refers to the amplification curve and melting curve collected during and after the PCR cycle reaction. The amplification curve is the fluorescence value curve formed by collecting fluorescence values ​​once during the amplification process of each PCR cycle reaction, while the melting curve is the temperature-fluorescence value curve and its derivative with respect to temperature obtained by collecting fluorescence values ​​point by point during the melting stage after amplification at a fixed step size and post-processing. The amplification configuration characteristics refer to the characteristic set of detection configuration attributes corresponding to each reaction well, including target identifier, probe fluorescence channel identifier, modified motif flanking sequence type, and well location usage marker, used to characterize the detection object, fluorescence channel, modification type, and experimental purpose of the corresponding reaction well. The well location usage marker is derived from the experimental well location layout table and is used to distinguish between sample wells, template-free control wells (NTC), unmodified control wells (WT, such as known 0% methylated gDNA), and internal control housekeeping gene wells.

[0026] The experimental well layout table refers to a pre-established correspondence between each reaction well and the object being tested, for example: The well site usage markers include at least the test sample partition and the template-free control partition (NTC); the target identifiers include the target modification site identifier, the wild-type site identifier, or the housekeeping gene site identifier; the probe fluorescence channel identifiers include the FAM channel, the HEX / VIC channel, or the Cy5 channel; the modification motif flanking sequence type refers to the adjacent base sequence characteristics within the context of the test nucleic acid modification site. For a candidate modification site C (pG), when the central C is a methylation candidate site, the upstream flanking sequence is a GC-enriched region, and the downstream flanking sequence is a CG continuation region. The upstream flanking sequence and the downstream flanking sequence together constitute the modification motif flanking sequence type, which is used to reflect the influence of the local sequence environment on amplification efficiency, probe binding characteristics, and melting behavior.

[0027] In detail, the process of obtaining real-time fluorescence detection data and amplification configuration characteristics for each qPCR reaction well includes: According to the preset experimental well layout table, add the pre-configured nucleic acid samples to be tested, template-free control samples, unmodified control samples, and internal control samples into the corresponding reaction wells; Simultaneous PCR cycle amplification was performed on each reaction well, and the real-time fluorescence signal of each reaction well was collected during the cycle amplification to obtain the amplification curve; Temperature data and fluorescence signals of each reaction well were collected after the cycle amplification was completed, and melting curves were constructed. The amplification curves and melting curves were then combined to form real-time fluorescence detection data. The target identifier, probe fluorescence channel identifier, modified motif flanking sequence type, and well purpose marker corresponding to each reaction well are extracted from the experimental well layout table and the preset amplification system configuration file, and then compiled into amplification configuration features.

[0028] The nucleic acid sample to be tested refers to a nucleic acid sample containing the target modification site and whose modification abundance is to be measured; the template-free control sample refers to a reaction system that does not contain any nucleic acid template and is used to monitor background contamination in reagents and the environment; the unmodified control sample refers to a nucleic acid sample that is known to not contain the target modification and is used to establish a negative baseline reference; the internal control sample refers to a housekeeping gene or a reference gene with a known constant modification level and is used for normalization correction of sample loading amount and amplification efficiency; after adding the above samples to the corresponding reaction wells according to the layout table, the real-time fluorescence PCR detector is used to perform synchronous cyclic amplification of each reaction well; the amplification system configuration file is a detection system configuration file that is pre-saved when the PCR reaction is established and is used to record parameters such as primer number, probe number, probe fluorescence channel, modification site, amplification program and detection mode corresponding to each reaction well.

[0029] Specifically, the process of constructing detection system combination keys and performing feature grouping based on the amplification configuration features includes: The target identifier, probe fluorescence channel identifier, and modified motif flanking sequence type in the amplification configuration features are concatenated into a detection system combination bond; The real-time fluorescence detection data corresponding to each reaction well are grouped according to the combination key of the detection system to obtain the fluorescence equivalence class set.

[0030] Among them, each fluorescence equivalence class in the fluorescence equivalence class set is a set of real-time fluorescence detection data that satisfies the combinatorial bonds of a specific detection system.

[0031] In detail, the process of performing median absolute deviation clipping on the fluorescent equivalence class set includes: One by one, the fluorescence equivalence classes in the fluorescence equivalence class set are selected as target equivalence classes. Real-time fluorescence detection data corresponding to well positions marked as template-free control wells and unmodified control wells are selected from the target equivalence classes as control fluorescence detection data. The fluorescence intensity values ​​corresponding to the control fluorescence detection data in the early stage of cyclic amplification are compiled into a control fluorescence value set; The median and median absolute deviation of the control fluorescence value set were calculated and compiled into a statistical summary; Based on the statistical summary, outlier values ​​in the control fluorescence value set are removed to obtain the clean control fluorescence value set; The truncated mean value of the fluorescence intensity values ​​in the clean control fluorescence value set is calculated to obtain the local baseline fluorescence value of the target equivalence class; Calculate the reference dispersion corresponding to the clean control fluorescence value set, encapsulate the local baseline fluorescence value and the reference dispersion as an initial reference object, and aggregate the initial reference objects of all target equivalence classes into an initial reference object set.

[0032] The fluorescence equivalence class set refers to the set of fluorescence detection data after deterministic grouping according to the detection system combinatorial bonds. The reaction wells within each fluorescence equivalence class maintain consistency in target type, fluorescence channel, and flanking sequence environment, thus possessing consistent detection background, fluorescence response mechanism, and amplification kinetics characteristics, serving as the basic analytical unit for subsequent statistical modeling. The early stage of cyclic amplification refers to the initial cyclic stage before PCR cyclic amplification enters the exponential growth phase, such as cycles 3 to 15, specifically determined based on the pre-experimental signal-to-noise ratio. During this period, the amplification signal of the positive template is not significantly higher than the background level, and the fluorescence values ​​of the control wells and unmodified control wells mainly reflect the system's optical background and physical baseline level. Median Absolute Deviation (MAD) is a robust statistic used to measure the degree of data dispersion. It involves first calculating the median of the fluorescence intensity values, then calculating the absolute deviation (absolute value of the difference) of each fluorescence intensity value from the median, and using the median of all absolute deviations as the median absolute deviation. Outlier clipping can be performed according to the following criteria: ; in, This refers to the first [number] in the control fluorescence value set. Each fluorescence intensity value, This refers to the median of fluorescence intensity values. It is a constant between 2.5 and 3.0. This refers to the median absolute deviation. The scaling factor corresponding to the median absolute deviation can be 1.4826. When the corresponding fluorescence intensity value meets the above conditions, it is retained in the clean control fluorescence value set; otherwise, it is truncated and discarded. The truncated mean refers to the mean of the remaining fluorescence intensity values ​​after removing 5% of the extreme values ​​(2.5% from each end) of the fluorescence intensity values ​​in the clean control fluorescence value set. The reference dispersion can be determined by the standard deviation, interquartile range, or median absolute deviation, for example, it can be the interquartile range divided by 1.349.

[0033] By acquiring real-time fluorescence detection data and amplification configuration characteristics, and using detection system combination keys to complete feature grouping of different detection systems, mutual interference between different targets, different modification types, and different fluorescence channels can be avoided, as well as baseline crosstalk and statistical confusion between different detection systems. By screening early cycle data of control wells within equivalence classes and using median absolute deviation for robust outlier pruning, occasional noise, contaminated wells, and abnormal background signals can be effectively removed, improving the stability of local background estimation. By constructing an initial reference object set containing local baseline fluorescence values ​​and reference dispersion, a stable and reliable reference basis is provided for subsequent dual thermodynamic boundary construction, soft membership weight calculation, and modification rate estimation, thereby improving the accuracy, robustness, and cross-batch consistency of nucleic acid modification site quantitative optimization analysis results.

[0034] S2. Construct a dual thermodynamic boundary set corresponding to the fluorescence equivalence class set based on the initial reference object set, and perform soft membership weight calculation and modification rate estimation on the real-time fluorescence detection data to obtain an initial quantitative result set.

[0035] The dual thermodynamic boundary in the dual thermodynamic boundary set corresponds to the fluorescence equivalence class in the fluorescence equivalence class set. It is used to represent the range of thermal stability changes of the amplified products of methylated and unmethylated templates in the corresponding fluorescence equivalence class during the melting process, and to describe the continuous thermodynamic response range corresponding to different modification ratios.

[0036] For details, please refer to Figure 2 The process of constructing the dual thermodynamic boundary set corresponding to the fluorescence equivalence class set based on the initial reference object set includes: S21. Select each fluorescence equivalence class in the fluorescence equivalence class set as the target equivalence class, and take the initial reference object corresponding to the target equivalence class in the initial reference object set as the target initial object. S22. Extract the melting curve set from the real-time fluorescence detection data of the target equivalence class, and extract the local baseline fluorescence value and the corresponding reference dispersion from the target initial object; S23. The melting curve set is superimposed and averaged to calculate the reference melting spectrum. The peak of the reference melting spectrum is located according to the local baseline fluorescence value and the reference dispersion to obtain the boundary of the characteristic peak of the methylated probe and the boundary of the characteristic peak of the unmethylated probe. S24. The temperature range between the boundary of the methylated probe characteristic peak and the boundary of the unmethylated probe characteristic peak is taken as the melting rain region. The boundary of the methylated probe characteristic peak, the boundary of the unmethylated probe characteristic peak, and the melting rain region are encapsulated as a dual thermodynamic boundary, and all dual thermodynamic boundaries are gathered into a dual thermodynamic boundary set.

[0037] The reference melting spectrum is a line spectrum obtained by averaging the melting curves in the melting curve group and calculating the temperature-fluorescence derivative. Peak localization refers to using the local baseline fluorescence value as the background zero reference and the reference scattering as the detection scale, and determining the main peak in the reference melting spectrum that satisfies the following formula as the probe characteristic peak: in, The fluorescence value is the reference melting spectrum. The temperature in the reference melting spectrum, As the background zero reference, This is the preset scale factor, which defaults to 3. To determine the measurement scale, the left side of the main peak was moved back to... The corresponding horizontal axis temperature is taken as the starting temperature of the peak, and the right side of the main peak is reduced to... The corresponding horizontal axis temperature is used as the termination temperature of the peak. The melting curves of the reaction wells marked as unmodified control wells in the target equivalence class are superimposed and averaged to locate the main peak position as the boundary of the unmethylated probe characteristic peak. The melting curves of the positive control reaction wells with known high methylation levels are superimposed and averaged to locate the main peak position as the boundary of the methylated probe characteristic peak. Methylated probe channels refer to oligonucleotide probe channels that specifically bind to the methylated cytosine sequence retained after bisulfite conversion, while unmethylated probe channels refer to oligonucleotide probe channels that specifically bind to the uracil sequence converted from unmethylated cytosine after bisulfite conversion.

[0038] Specifically, the process of calculating soft membership weights and estimating the modification rate of the real-time fluorescence detection data includes: The reaction wells marked as sample wells in the amplification configuration features are used as target reaction wells, and the real-time fluorescence detection data corresponding to the target reaction wells are used as target fluorescence detection data. The dual thermodynamic boundary corresponding to the target fluorescence detection data is taken as the target dual boundary; The target temperature-fluorescence derivative spectrum corresponding to the melting curve is extracted from the target fluorescence detection data. The soft membership weight is calculated at each temperature point on the target temperature-fluorescence derivative spectrum based on the target double boundary. The modification rate is estimated based on the soft membership weight to obtain the initial modification rate. The initial modification rate of the target reaction wells is encapsulated into an initial quantitative result, and the initial quantitative results of all target reaction wells are aggregated into an initial quantitative result set.

[0039] The target temperature-fluorescence derivative spectrum refers to the line spectrum formed by calculating the temperature-fluorescence derivative of the melting curve. Soft membership weights are calculated based on this line spectrum. For example, temperature points on the target temperature-fluorescence derivative spectrum are extracted point-by-point as target temperature points. When the target temperature point is less than or equal to the termination temperature of the unmethylated probe characteristic peak boundary of the target double boundary, the soft membership weight is assigned a value of 0. When the target temperature point is greater than or equal to the starting temperature of the methylated probe characteristic peak boundary of the target double boundary, the soft membership weight is assigned a value of 1. When the target temperature point is within the melting rain zone of the target double boundary, the corresponding soft membership weight is calculated according to the following formula: in, For soft membership weights, It is the temperature distance from the target temperature point to the boundary of the characteristic peak of the methylation probe. This refers to the temperature distance from the target temperature point to the boundary of the characteristic peak of the unmethylated probe. Modification rate estimation involves integrating and summing the soft membership weights of the methylated state along the temperature axis for all temperature points to obtain the total membership degree of the methylated state; integrating and summing the soft membership weights of the unmethylated state along the temperature axis for all temperature points to obtain the total membership degree of the unmethylated state; dividing the total membership degree of the methylated state by the sum of the total membership degrees of the methylated and unmethylated states yields the initial modification rate of the reaction well; and compiling the initial modification rates of all reaction wells into an initial quantitative result set.

[0040] By constructing dual thermodynamic boundaries corresponding to each fluorescence equivalence class based on the initial reference object set and analyzing them in conjunction with the thermodynamic characteristics of the melting curve, the amplification process and melting information can be fully utilized to achieve continuous determination of modified and unmodified templates. Through the soft membership weight calculation method of the melting rain zone, the boundary misjudgment caused by the traditional hard threshold method can be avoided, and the stability of the determination when the modification ratio changes continuously can be improved, thereby enhancing the robustness and accuracy of quantitative optimization of nucleic acid modification sites.

[0041] S3. Calculate the amplification confidence level of each reaction well based on the real-time fluorescence detection data, and use the amplification confidence level to label the initial quantitative result set to obtain the labeled quantitative result set.

[0042] Among them, amplification reliability refers to a quantitative indicator that characterizes the authenticity and reliability of each reaction well during the PCR amplification process. The lower the amplification reliability, the greater the possibility that the data of the corresponding reaction well has quality defects.

[0043] For details, please refer to Figure 3 The process of calculating the amplification confidence of each reaction well based on the real-time fluorescence detection data includes: S31. The reaction wells marked as sample wells in the amplification configuration features are designated as target reaction wells, the real-time fluorescence detection data corresponding to the target reaction wells are designated as target fluorescence detection data, and the fluorescence equivalence class corresponding to the target fluorescence detection data in the fluorescence equivalence class set is designated as the target equivalence class. S32. Extract the exponential phase rise slope from the amplification curve of the target fluorescence detection data, and perform slope anomaly analysis on the exponential phase rise slope according to the target equivalence class to obtain the exponential phase slope anomaly degree. S33. Extract the half-width of the main peak from the melting curve of the target fluorescence detection data, and perform threshold anomaly analysis on the half-width of the main peak according to the target equivalence class to obtain the peak width anomaly degree. S34. Select the Ct values ​​corresponding to the sample complex well group to which the target reaction well belongs in the target equivalence class, collect them into a target Ct value group, and calculate the Ct value dispersion corresponding to the target Ct value group. S35. Based on the exponential period slope anomaly, peak width anomaly, and Ct value dispersion, a comprehensive reliability evaluation is performed to obtain the amplification reliability.

[0044] The extraction of the exponential phase rise slope includes: performing baseline subtraction on the amplification curve, identifying the point where the fluorescence value first exceeds the baseline plus three times the standard deviation threshold as the exponential phase start cycle, identifying the zero-crossing point of the second derivative of the amplification curve as the exponential phase end cycle, performing linear regression on the cycle number-fluorescence value within the interval between the start and end cycles, and using the obtained regression coefficient as the exponential phase rise slope; slope anomaly analysis includes: extracting the exponential phase rise slope of all amplification curves in the target equivalence class, calculating the median of all exponential phase rise slopes in the target equivalence class, obtaining the median rise slope, calculating the deviation between the exponential phase rise slope of the target fluorescence detection data and the median rise slope, and using the ratio of this deviation to a preset rise slope deviation threshold as the exponential phase slope anomaly degree, where the rise slope deviation threshold can be the sum of the median of all exponential phase rise slopes in the target equivalence class and twice the median absolute deviation; extracting the main peak half-width includes: smoothing and filtering the temperature-fluorescence derivative spectrum of the melting curve, locating the main peak temperature, and along the temperature axis... The spectrum extends to the left and right directions respectively, reaching a position where the derivative spectrum height drops to half the peak value, and the temperature difference between the two positions is taken as the half-width at half maximum (WHM) of the main peak. Threshold anomaly analysis includes: extracting the WHM of the main peak from all melting curves in the target equivalence class, calculating the median WHM of all main peaks in the target equivalence class, obtaining the median WHM, calculating the deviation between the WHM of the main peak and the median WHM of the target fluorescence detection data, and using the ratio of this deviation to a preset WHM deviation threshold as the peak width anomaly. The WHM deviation threshold can be... The median of the half-peak width of all main peaks in the equivalence class is the sum of twice the median absolute deviation; the sample duplicate well group refers to the combination of reaction wells that are technically repeated with the target reaction wells using the same target and the same processing steps; the Ct value dispersion refers to the standard deviation of the target Ct value group; comprehensive reliability evaluation refers to establishing the reciprocal function of the weighted sum of the exponential period slope anomaly, peak width anomaly, and Ct value dispersion, such that the greater the exponential period slope anomaly, the greater the peak width anomaly, and the greater the Ct value dispersion, the lower the corresponding amplification reliability.

[0045] Specifically, the process of using the amplified confidence level to label the initial quantitative result set includes: Each initial quantitative result in the initial quantitative result set is selected as the target initial quantitative result, and the initial modification rate of the target initial quantitative result is taken as the target initial modification rate. The amplification confidence level of the reaction well corresponding to the initial quantitative result of the target is taken as the target confidence level; The target initial modification rate is labeled using the target credibility and then jointly encapsulated to obtain labeled quantitative results. All labeled quantitative results are then aggregated into a labeled quantitative result set.

[0046] By extracting the exponential rise slope, half-peak width of the main peak, and Ct dispersion of the nucleic acid sample to be tested, a joint reliability analysis can be performed by combining the stability of PCR amplification, the specificity of amplification products, and the reproducibility of the technique. This allows for a comprehensive evaluation of the reliability of the detection results of each reaction well, effectively identifying and labeling abnormal or low-quality data, and improving the stability and accuracy of the quantitative analysis results of nucleic acid modification sites.

[0047] S4. Update the initial reference object set using the labeled quantitative result set to obtain the standard reference object set.

[0048] In detail, the process of updating the initial reference object set using the labeled quantitative result set includes: One by one, the fluorescence equivalence classes in the fluorescence equivalence class set are selected as target equivalence classes. The labeled quantitative results corresponding to the target equivalence classes in the labeled quantitative results set are collected as target labeled results into a target labeled result group. The initial reference objects corresponding to the target equivalence classes are extracted from the initial reference object set as the current reference objects. Traverse the target annotation results in the target annotation result group, and determine one by one whether the amplification confidence of the target annotation result is less than a preset confidence threshold, or whether the initial modification rate of the target annotation result is greater than a preset modification rate threshold. If so, the real-time fluorescence detection data corresponding to the target annotation result will be removed from the target equivalence class as abnormal fluorescence detection data; If not, return to the step of selecting the target annotation results in the target annotation result group one by one; Until all target standard results in the target annotation result group have been traversed, the target equivalence class after removing all abnormal fluorescence detection data will be taken as the target standard equivalence class; The local baseline fluorescence value and reference scatter of the target standard equivalence class are recalculated, the current reference object is updated, a standard reference object is obtained, and all standard reference objects are aggregated into a standard reference object set.

[0049] The confidence threshold is adaptively determined based on the amplification confidence distribution of all labeled quantitative results within the target equivalence class. It can be the difference between the median of all amplification confidences within the target equivalence class and 1.5 times the median absolute deviation. The modification rate threshold is adaptively determined based on the initial modification rate distribution of all labeled quantitative results within the target equivalence class. It can be the sum of the median of all initial modification rates within the target equivalence class and 1.5 times the median absolute deviation. The method for recalculating the local baseline fluorescence value and reference scattering is the same as that in S1, and will not be repeated here.

[0050] By updating the initial reference set using the labeled quantitative results set, amplification data from positive reaction wells with low confidence or high modification rates can be removed from the calculation of the reference set. This allows the local baseline fluorescence values ​​in the reference set to gradually approach the true fluorescence values ​​of the unmodified state, reducing the contamination effect of positive signals on the negative reference system caused by sample cross-contamination or incomplete chemical transformation, and ensuring the accuracy and consistency of the quantitative results.

[0051] S5. Based on the standard reference object set, recalculate the soft membership weights, estimate the modification rate, and perform iterative convergence analysis to obtain a quantitative result set of modification sites.

[0052] In detail, recalculating soft membership weights and estimating modification rates based on the standard reference object set refers to updating the initial reference object set using the standard reference object set and iteratively executing steps S2 to S4. The iterative convergence analysis refers to recording the initial quantitative result set of the current batch iteration and the initial quantitative result set of the previous batch iteration, and recording the iteration number at each iteration. When any of the following conditions are met, the iteration stops, and the labeled quantitative result set of the last batch is used as the quantitative result set of the modification sites: First, the root mean square error of the initial modification rate corresponding to the two initial quantitative result sets is calculated. When the root mean square error is less than the preset convergence threshold, the iteration stops. Second, the iteration stops when the number of iterations reaches the preset iteration number threshold.

[0053] By iteratively calculating soft membership weights and estimating modification rates, we can ensure that the fluorescence value of the reference object gradually approaches the true unmodified state during multiple iterations. By combining root mean square error and iteration number threshold for convergence analysis, we can ensure that the output quantitative results set of modified sites comes from a well-purified and stable reference object, thereby improving the quantitative accuracy and batch reproducibility of low-abundance modified samples in complex backgrounds.

[0054] Example 2: Please refer to Figure 4 The present invention discloses a quantitative optimization analysis system for nucleic acid modification sites, the system comprising an inter-class trimming module 101, an initial quantification module 102, a confidence weighting module 103, an object update module 104, and a result iteration module 105; The inter-class trimming module 101 acquires real-time fluorescence detection data and amplification configuration features of each reaction well in qPCR, constructs detection system combination bonds based on the amplification configuration features and performs feature grouping to obtain a fluorescence equivalence class set, and trims the fluorescence equivalence class set by median absolute deviation to obtain an initial reference object set. The initial quantification module 102 constructs a dual thermodynamic boundary set corresponding to the fluorescence equivalence class set based on the initial reference object set, and performs soft membership weight calculation and modification rate estimation on the real-time fluorescence detection data to obtain the initial quantification result set. The confidence weighting module 103 calculates the amplification confidence of each reaction well based on the real-time fluorescence detection data, and uses the amplification confidence to label the initial quantitative result set to obtain a labeled quantitative result set. The object update module 104 updates the initial reference object set using the labeled quantitative result set to obtain a standard reference object set; The result iteration module 105 recalculates the soft membership weights, estimates the modification rate, and performs iterative convergence analysis based on the standard reference object set to obtain a quantitative result set of modification sites.

[0055] The processes described above with reference to the flowcharts in the embodiments disclosed in this invention can be implemented as computer software programs. The embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the functions defined in the methods of this invention. It should be noted that the computer-readable medium described above in this invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more conductor segments, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical cables, RF, etc., or any suitable combination thereof.

[0056] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation that may be implemented in systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0057] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been shown and explained in the embodiments. Without departing from the stated principles, the implementation of the present invention may have any variations or modifications.

Claims

1. A quantitative optimization analysis method for nucleic acid modification sites, characterized in that, The method includes: Real-time fluorescence detection data and amplification configuration characteristics of each reaction well in qPCR are obtained. Based on the amplification configuration characteristics, detection system combination bonds are constructed and feature groups are performed to obtain fluorescence equivalence class sets. The fluorescence equivalence class sets are then truncated by median absolute deviation to obtain an initial reference object set. Based on the initial reference object set, a dual thermodynamic boundary set corresponding to the fluorescence equivalence class set is constructed, and soft membership weights and modification rates are calculated and estimated on the real-time fluorescence detection data to obtain an initial quantitative result set. The amplification confidence level of each reaction well is calculated based on the real-time fluorescence detection data. The amplification confidence level is then used to label the initial quantitative result set to obtain a labeled quantitative result set. The initial reference object set is updated using the labeled quantitative result set to obtain a standard reference object set; Based on the aforementioned standard reference object set, the soft membership weights were recalculated, the modification rate was estimated, and iterative convergence analysis was performed to obtain a quantitative result set of modification sites.

2. The quantitative optimization analysis method for nucleic acid modification sites according to claim 1, characterized in that, The process of obtaining real-time fluorescence detection data and amplification configuration characteristics for each qPCR reaction well includes: According to the preset experimental well layout table, add the pre-configured nucleic acid samples to be tested, template-free control samples, unmodified control samples, and internal control samples into the corresponding reaction wells; Simultaneous PCR cycle amplification was performed on each reaction well, and the real-time fluorescence signal of each reaction well was collected during the cycle amplification to obtain the amplification curve; Temperature data and fluorescence signals of each reaction well were collected after the cycle amplification was completed, and melting curves were constructed. The amplification curves and melting curves were then combined to form real-time fluorescence detection data. The target identifier, probe fluorescence channel identifier, modified motif flanking sequence type, and well purpose marker corresponding to each reaction well are extracted from the experimental well layout table and the preset amplification system configuration file, and then compiled into amplification configuration features.

3. The quantitative optimization analysis method for nucleic acid modification sites according to claim 2, characterized in that, The process of constructing detection system combination keys and performing feature grouping based on the amplification configuration features includes: The target identifier, probe fluorescence channel identifier, and modified motif flanking sequence type in the amplification configuration features are concatenated into a detection system combination bond; The real-time fluorescence detection data corresponding to each reaction well are grouped according to the combination key of the detection system to obtain the fluorescence equivalence class set.

4. The quantitative optimization analysis method for nucleic acid modification sites according to claim 1, characterized in that, The process of performing median absolute bias clipping on the fluorescence equivalence class set includes: One by one, the fluorescence equivalence classes in the fluorescence equivalence class set are selected as target equivalence classes. Real-time fluorescence detection data corresponding to well positions marked as template-free control wells and unmodified control wells are selected from the target equivalence classes as control fluorescence detection data. The fluorescence intensity values ​​corresponding to the control fluorescence detection data in the early stage of cyclic amplification are compiled into a control fluorescence value set; The median and median absolute deviation of the control fluorescence value set were calculated and compiled into a statistical summary; Based on the statistical summary, outlier values ​​in the control fluorescence value set are removed to obtain the clean control fluorescence value set; The truncated mean value of the fluorescence intensity values ​​in the clean control fluorescence value set is calculated to obtain the local baseline fluorescence value of the target equivalence class; Calculate the reference dispersion corresponding to the clean control fluorescence value set, encapsulate the local baseline fluorescence value and the reference dispersion as an initial reference object, and aggregate the initial reference objects of all target equivalence classes into an initial reference object set.

5. The quantitative optimization analysis method for nucleic acid modification sites according to claim 4, characterized in that, The process of constructing the dual thermodynamic boundary set corresponding to the fluorescence equivalence class set based on the initial reference object set includes: Each fluorescence equivalence class in the fluorescence equivalence class set is selected as the target equivalence class, and the initial reference object corresponding to the target equivalence class in the initial reference object set is taken as the target initial object. The melting curve set is extracted from the real-time fluorescence detection data of the target equivalence class, and the local baseline fluorescence value and the corresponding reference scattering are extracted from the target initial object. The melting curves are superimposed and averaged to calculate the baseline melting spectrum. The peaks of the baseline melting spectrum are located based on the local baseline fluorescence value and the reference dispersion to obtain the boundary between the characteristic peaks of the methylated probe and the characteristic peaks of the unmethylated probe. The temperature range between the boundary of the methylated probe characteristic peak and the boundary of the unmethylated probe characteristic peak is defined as the melting rain region. The boundary of the methylated probe characteristic peak, the boundary of the unmethylated probe characteristic peak, and the melting rain region are encapsulated as a dual thermodynamic boundary, and all dual thermodynamic boundaries are aggregated into a dual thermodynamic boundary set.

6. The quantitative optimization analysis method for nucleic acid modification sites according to claim 5, characterized in that, The process of calculating soft membership weights and estimating modification rate for the real-time fluorescence detection data includes: The reaction wells marked as sample wells in the amplification configuration features are used as target reaction wells, and the real-time fluorescence detection data corresponding to the target reaction wells are used as target fluorescence detection data. The dual thermodynamic boundary corresponding to the target fluorescence detection data is taken as the target dual boundary; The target temperature-fluorescence derivative spectrum corresponding to the melting curve is extracted from the target fluorescence detection data. The soft membership weight is calculated at each temperature point on the target temperature-fluorescence derivative spectrum based on the target double boundary. The modification rate is estimated based on the soft membership weight to obtain the initial modification rate. The initial modification rate of the target reaction wells is encapsulated into an initial quantitative result, and the initial quantitative results of all target reaction wells are aggregated into an initial quantitative result set.

7. The quantitative optimization analysis method for nucleic acid modification sites according to claim 6, characterized in that, The process of calculating the amplification confidence of each reaction well based on the real-time fluorescence detection data includes: The reaction wells marked as sample wells in the amplification configuration features are designated as target reaction wells, the real-time fluorescence detection data corresponding to the target reaction wells are designated as target fluorescence detection data, and the fluorescence equivalence class corresponding to the target fluorescence detection data in the fluorescence equivalence class set is designated as the target equivalence class. The exponential phase rise slope is extracted from the amplification curve of the target fluorescence detection data, and the slope anomaly analysis is performed on the exponential phase rise slope according to the target equivalence class to obtain the exponential phase slope anomaly degree. The half-peak width of the main peak is extracted from the melting curve of the target fluorescence detection data, and the half-peak width of the main peak is subjected to threshold anomaly analysis based on the target equivalence class to obtain the peak width anomaly degree. Select the Ct values ​​corresponding to the sample complex group to which the target reaction well belongs in the target equivalence class, aggregate them into a target Ct value group, and calculate the Ct value dispersion corresponding to the target Ct value group; The amplification credibility is obtained by comprehensively evaluating the exponential slope anomaly, peak width anomaly, and Ct value dispersion.

8. The quantitative optimization analysis method for nucleic acid modification sites according to claim 7, characterized in that, The process of using the amplified confidence level to label the initial quantitative result set includes: Each initial quantitative result in the initial quantitative result set is selected as the target initial quantitative result, and the initial modification rate of the target initial quantitative result is taken as the target initial modification rate. The amplification confidence level of the reaction well corresponding to the initial quantitative result of the target is taken as the target confidence level; The target initial modification rate is labeled using the target credibility and then jointly encapsulated to obtain labeled quantitative results. All labeled quantitative results are then aggregated into a labeled quantitative result set.

9. The quantitative optimization analysis method for nucleic acid modification sites according to claim 8, characterized in that, The process of updating the initial reference object set using the labeled quantitative result set includes: One by one, the fluorescence equivalence classes in the fluorescence equivalence class set are selected as target equivalence classes. The labeled quantitative results corresponding to the target equivalence classes in the labeled quantitative results set are collected as target labeled results into a target labeled result group. The initial reference objects corresponding to the target equivalence classes are extracted from the initial reference object set as the current reference objects. Traverse the target annotation results in the target annotation result group, and determine one by one whether the amplification confidence of the target annotation result is less than a preset confidence threshold, or whether the initial modification rate of the target annotation result is greater than a preset modification rate threshold. If so, the real-time fluorescence detection data corresponding to the target annotation result will be removed from the target equivalence class as abnormal fluorescence detection data; If not, return to the step of selecting the target annotation results in the target annotation result group one by one; Until all target standard results in the target annotation result group have been traversed, the target equivalence class after removing all abnormal fluorescence detection data will be taken as the target standard equivalence class; The local baseline fluorescence value and reference scatter of the target standard equivalence class are recalculated, the current reference object is updated, a standard reference object is obtained, and all standard reference objects are aggregated into a standard reference object set.

10. A quantitative optimization analysis system for nucleic acid modification sites, used to perform the method as described in any one of claims 1 to 9, characterized in that, The system includes an inter-class trimming module, an initial quantitative module, a credibility weighting module, an object update module, and a result iteration module; The inter-class trimming module acquires real-time fluorescence detection data and amplification configuration features of each qPCR reaction well, constructs detection system combination bonds based on the amplification configuration features and performs feature grouping to obtain a fluorescence equivalence class set, and trims the fluorescence equivalence class set by median absolute deviation to obtain an initial reference object set; The initial quantification module constructs a dual thermodynamic boundary set corresponding to the fluorescence equivalence class set based on the initial reference object set, and performs soft membership weight calculation and modification rate estimation on the real-time fluorescence detection data to obtain the initial quantification result set. The confidence weighting module calculates the amplification confidence of each reaction well based on the real-time fluorescence detection data, and uses the amplification confidence to label the initial quantitative result set to obtain a labeled quantitative result set. The object update module updates the initial reference object set using the labeled quantitative result set to obtain a standard reference object set; The result iteration module recalculates the soft membership weights, estimates the modification rate, and performs iterative convergence analysis based on the standard reference object set to obtain a quantitative result set of modification sites.

Images