A method and system for amplicon internal reference normalization

Abstract

The application discloses a method and system for amplicon internal reference normalization, and belongs to the technical field of biological information analysis. The method for amplicon normalization comprises the steps of obtaining an internal reference normalization baseline and obtaining an internal reference correction coefficient, and further performing internal reference normalization according to the internal reference normalization baseline and the internal reference correction coefficient. The method can improve the detection sensitivity and has very important popularization and application value.

Description

Technical Field

[0001] This invention belongs to the field of bioinformatics analysis technology, specifically, it relates to a method for normalizing amplicon internal controls. Background Technology

[0002] Amplicon sequencing is a method based on PCR amplification of target genomic regions for high-depth sequencing, enabling efficient analysis of mutations in the target genome. Based on amplicon sequencing, we can perform not only point mutation analysis but also copy number / quantitative analysis. For copy number / quantitative analysis, we need to normalize different samples to ensure analysis at a consistent scale. Similar to qPCR, we need to introduce an internal reference gene as a baseline for quantitative analysis to normalize the samples.

[0003] Currently, there are several main methods. One approach involves using a large number of amplicon reads, covering a certain amount of internal reference genes, and using the sum of all amplicon reads as the correction baseline. Another approach is to select a suitable sum of internal references as the correction baseline. However, different amplicon reads have varying efficiencies, leading to different stability levels. Furthermore, the expression abundance of internal references also varies, resulting in significant differences in the amount of data obtained after capture sequencing, which in turn affects stability. If the amount of internal reference data (number of reads) varies greatly, being either too low or too high, such excessively high internal references, if unstable, will have a significant impact on the overall internal reference sum. Conversely, internal references with excessively low data, even if very stable, will contribute little to the overall stable system. Summary of the Invention

[0004] To solve at least one of the above-mentioned technical problems, the present invention adopts the following technical solution:

[0005] The first aspect of this invention provides a method for amplicon internal reference normalization, including the step of obtaining a normalized baseline for the internal reference, specifically including:

[0006] S11, obtain sequencing data from m samples, of which j are intrinsic parameters, and obtain the number R of reads for intrinsic parameter i in sample n. ni And obtain the representative number of reads of intrinsic parameter i in all m samples, denoted as MR. i , where i = 1, ..., j, n = 1, 2, ..., m;

[0007] S12, Calculate the initial normalized value of the intrinsic parameter i in sample n: k ni =R ni / MR i Next, calculate the initial normalized values ​​of all intrinsic parameters in sample n: Next, calculate the number of reads of intrinsic parameter i after initial normalization in sample n as R' ni =R ni / K1n ;

[0008] S13, calculate the coefficient of variation (CV) of the number of reads after initial normalization of the intrinsic parameter i across all m samples. i If CV i If the sample exceeds the prediction threshold, remove the abnormal sample.

[0009] S14, repeat steps S11-S13 for the remaining samples until CV is achieved. i Without exceeding a preset threshold, obtain the representative number of reads for intrinsic parameter i in the remaining samples, denoted as BR. i , which serves as the normalized baseline for internal references.

[0010] In some embodiments of the present invention, the representative number refers to the median, mean, mode, quarter quantile, or third quantile. Preferably, the representative number is the median.

[0011] In some embodiments of the present invention, the preset threshold is a representative number of CV values ​​obtained using population data. Further, the representative number refers to the median, mean, mode, 1 / 4 quantile, or 3 / 4 quantile; preferably, the representative number is the 3 / 4 quantile.

[0012] In some embodiments of the present invention, in step S13, removing abnormal samples means that if one or more samples are removed, the CV... i If the value is not greater than the prediction threshold, then the sample or samples mentioned above are considered abnormal samples.

[0013] In some embodiments of the present invention, the method further includes the step of obtaining an internal reference correction coefficient, specifically including:

[0014] S21, Calculate the initial normalization coefficient of the intrinsic parameter i in sample n: k' ni =R ni / BR i And calculate the initial correction coefficients for all intrinsic parameters in sample n.

[0015] S22, the correction coefficient of the intrinsic parameter i is selected from one of 0.5, 0.6, 0.7, 0.8, 0.9 and 1. The final correction coefficient of the intrinsic parameter is determined by exhaustive search:

[0016] S221, For each different intrinsic parameter, a correction coefficient is randomly selected to obtain the combination of intrinsic parameter correction coefficients, and the normalized values ​​of all intrinsic parameters in sample n are calculated. The number of all intrinsic parameter reads in sample n is calculated using K2. n Normalization involves comparing the normalized number of intrinsic parameter reads with the normalized baseline BR. iPerform error analysis to obtain the root mean square error;

[0017] S222, perform the same operation as step S221 on all combinations of internal parameter correction coefficients to obtain different root mean square errors;

[0018] S223, Select the set of internal parameter correction coefficients with the smallest mean square error as the optimal combination of internal parameter correction coefficients: q′1, ..., q′ i , where q′ i is the optimal correction coefficient for intrinsic parameter i.

[0019] In some embodiments of the present invention, the method further includes a step of normalization using the combination of the intrinsic parameter normalization baseline and the optimal intrinsic parameter correction coefficients as a correction library:

[0020] S31, Calculate the initial normalization coefficient of the intrinsic parameter i in the sample: k′ i =R i / BR i And calculate the initial correction coefficients for all intrinsic parameters in the sample. Among them, R i Let i be the number of reads of intrinsic parameter i in the sample, i = 1, ..., j, where j is the number of intrinsic parameters in the sample;

[0021] S32, using the optimal combination of intrinsic parameter correction coefficients, calculate the normalized values ​​of all intrinsic parameters in the sample.

[0022] S33, normalization is performed using the K2 value.

[0023] A second aspect of the present invention provides a system for amplicon internal reference normalization, comprising the following modules:

[0024] The data input module is used to accept sequencing data input;

[0025] The correction library module is used to store the normalized baseline of the intrinsic parameters and the optimal combination of intrinsic parameter correction coefficients;

[0026] The normalization module, connected to both the data input module and the correction library module, is used to perform normalization using the combination of the intrinsic parameter normalization baseline and the optimal intrinsic parameter correction coefficients as the correction library.

[0027] (1) Calculate the initial normalization coefficient of the intrinsic parameter i in the sample: k′ i =R i / BR i And calculate the initial correction coefficients for all intrinsic parameters in the sample. Among them, R i BR represents the number of reads of intrinsic parameter i in the sample. iLet i be the normalized baseline of the intrinsic parameter i, where i = 1, ..., j, and j is the number of intrinsic parameters in the sample.

[0028] (2) Using the optimal combination of intrinsic parameter correction coefficients, calculate the normalized values ​​of all intrinsic parameters in the sample. Where, q′ i This is the optimal correction coefficient for the internal reference;

[0029] S33, normalization is performed using the K2 value;

[0030] The output module, connected to the normalization module, is used to output the normalized data.

[0031] In some embodiments of the present invention, the correction library module is connected to the data input module for obtaining the internal reference normalized baseline using sequencing data from multiple samples:

[0032] (1) Obtain sequencing data from m samples, of which j are intrinsic parameters, and obtain the number R of reads of intrinsic parameter i in sample n. ni And obtain the representative number of reads of intrinsic parameter i in all m samples, denoted as MR. i , where i = 1, ..., j, n = 1, 2, ..., m;

[0033] (2) Calculate the initial normalized value of the intrinsic parameter i in sample n: k ni =R ni / MR i Next, calculate the initial normalized values ​​of all intrinsic parameters in sample n: Next, calculate the number of reads of intrinsic parameter i after initial normalization in sample n as R' ni =R ni / K1 n ;

[0034] (3) Calculate the coefficient of variation (CV) of the number of reads after initial normalization of the intrinsic parameter i in all m samples. i If CV i If the sample exceeds the prediction threshold, remove the abnormal sample.

[0035] (4) Repeat steps S11-S13 for the remaining samples until CV is achieved. i Without exceeding a preset threshold, obtain the representative number of reads for intrinsic parameter i in the remaining samples, which is the BR. i .

[0036] In some embodiments of the present invention, the representative number refers to the median, mean, mode, quarter quantile, or third quantile. Preferably, the representative number is the median.

[0037] In some embodiments of the present invention, the preset threshold is a representative number of CV values ​​obtained using population data. Further, the representative number refers to the median, mean, mode, 1 / 4 quantile, or 3 / 4 quantile; preferably, the representative number is the 3 / 4 quantile.

[0038] In some embodiments of the present invention, the correction library module is further configured to obtain the optimal combination of internal reference correction coefficients using sequencing data from multiple samples:

[0039] (1) Calculate the initial normalization coefficient of the intrinsic parameter i in the sample n: k' ni =R ni / BR i And calculate the initial correction coefficients for all intrinsic parameters in sample n.

[0040] (2) The correction coefficient of the intrinsic parameter i is selected from one of 0.5, 0.6, 0.7, 0.8, 0.9 and 1. The final combination of correction coefficients of the intrinsic parameter is determined by exhaustive search:

[0041] (2-1) Randomly select a correction coefficient for each different intrinsic parameter to obtain the combination of intrinsic parameter correction coefficients, and calculate the normalized value of all intrinsic parameters in sample n. The number of all intrinsic parameter reads in sample n is calculated using K2. n Normalization involves comparing the normalized number of intrinsic parameter reads with the normalized baseline BR. i Perform error analysis to obtain the root mean square error;

[0042] (2-2) Perform the same operation as step S221 on all combinations of internal parameter correction coefficients to obtain different root mean square errors;

[0043] (2-3) Select the set of internal parameter correction coefficients with the smallest mean square error as the optimal internal parameter correction coefficient combination: q′1, ... q′ i .

[0044] A third aspect of the present invention provides a computer device, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of any of the methods described in the first aspect of the present invention.

[0045] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of any of the methods described in the first aspect of the present invention.

[0046] Beneficial effects of the present invention

[0047] Compared with the prior art, the effective effects of the present invention include:

[0048] The normalization method of this invention can reduce internal reference error to a certain extent and significantly improve the accuracy of amplicon quantification, which has great value for promotion and application. Attached Figure Description

[0049] Figure 1 A schematic diagram of an amplicon internal reference normalization system according to Embodiment 2 of the present invention is shown.

[0050] Figure 2 A schematic diagram of another amplicon internal reference normalization system of Embodiment 2 of the present invention is shown. Detailed Implementation

[0051] Unless otherwise stated, implied from the context, or as is customary in the art, all parts and percentages in this application are based on weight, and all testing and characterization methods used are concurrent with the filing date of this application. Where applicable, any patent, patent application, or disclosure relating to this application is incorporated herein by reference in its entirety, and its equivalent patent families are also incorporated herein by reference, particularly the definitions disclosed in these documents concerning synthetic techniques, product and processing design, polymers, comonomers, initiators, or catalysts in the art. If any definition of a specific term disclosed in the prior art is inconsistent with any definition provided in this application, the definition provided in this application shall prevail.

[0052] The numerical ranges in this application are approximate values ​​and therefore may include values ​​outside the range unless otherwise stated. A numerical range includes all values ​​from the lower limit to the upper limit, increasing by one unit, provided there is an interval of at least two units between any lower and any higher value. For example, if the stated composition, physical, or other property (such as molecular weight, melt index, etc.) is 100 to 1000, it means that all individual values, such as 100, 101, 102, etc., are explicitly listed, as well as all subranges, such as 100 to 166, 155 to 170, 198 to 200, etc. For ranges containing values ​​less than 1 or fractions greater than 1 (e.g., 1.1, 1.5, etc.), one unit is appropriately considered as 0.0001, 0.001, 0.01, or 0.1. For ranges containing single digits less than 10 (e.g., 1 to 5), one unit is generally considered as 0.1. These are merely specific examples of what is intended to be expressed, and all possible combinations of values ​​between the listed minimum and maximum values ​​are considered to be clearly stated in this application.

[0053] The terms “comprising,” “including,” “having,” and their derivatives do not exclude the presence of any other components, steps, or processes, regardless of whether such other components, steps, or processes are disclosed in this application. To eliminate any doubt, unless expressly stated otherwise, all compositions using the terms “comprising,” “including,” or “having” in this application may contain any additional additives, excipients, or compounds. Conversely, except for those necessary for operational performance, the term “substantially constitutes…” excludes any other components, steps, or processes described below with respect to that term. The term “consisting of…” does not include any components, steps, or processes not specifically described or listed. Unless expressly stated otherwise, the term “or” refers to the individual members listed or any combination thereof.

[0054] To make the technical problems solved by the present invention, the technical solutions and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments.

[0055] Example

[0056] The following examples are used to illustrate preferred embodiments of the invention. Those skilled in the art will understand that the techniques disclosed in the examples represent techniques discovered by the inventors that can be used to implement the invention, and therefore can be considered preferred embodiments for implementing the invention. However, those skilled in the art should understand from this specification that many modifications can be made to the specific embodiments disclosed herein, still yielding the same or similar results, without departing from the spirit or scope of the invention.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains, and all materials publicly cited herein and referenced by them are incorporated herein by reference.

[0058] Those skilled in the art will recognize, or can learn through routine experimentation, many equivalents of the specific embodiments of the invention described herein. These equivalents will be included in the claims.

[0059] Unless otherwise specified, the experimental methods used in the following examples are conventional methods. Unless otherwise specified, the instruments and equipment used in the following examples are all conventional laboratory instruments and equipment; unless otherwise specified, the experimental materials used in the following examples were all purchased from conventional biochemical reagent stores.

[0060] Example 1: Construction and Application of the Correction Library

[0061] 1. Construction of the normalized baseline for internal references

[0062] This embodiment first constructs a central baseline for the expression pattern of an internal parameter based on a certain number of samples, namely the normalized baseline.

[0063] For a single sequencing run, assume there are m samples and j internal parameters.

[0064] (1) The number of reads of the intrinsic parameter i (i = 1, ..., j) in the sample n (n = 1, 2, ..., m) is R. ni The median number of reads for intrinsic parameter i across all m samples is MR. i .

[0065] (2) Calculate the initial normalized value of the intrinsic parameter i in sample n: k ni =R ni / MR i Next, calculate the initial normalized values ​​of all intrinsic parameters in sample n: This normalizes all 9046 samples; then, the number of reads with the intrinsic parameter i after initial normalization in sample n is calculated as R'. ni =R ni / K1 n ;

[0066] (3) Calculate the number of reads (R′) of the intrinsic parameter i after initial normalization in all m samples. 1i 、R' 2i ... R' mi Coefficient of variation (CV) i If CV i If the sample exceeds the prediction threshold, remove the abnormal sample.

[0067] (4) Repeat steps S11-S13 for the remaining samples until CV is achieved. i Without exceeding a preset threshold, obtain the representative number BR of the number of reads of intrinsic parameter i in the remaining samples. i , which serves as the normalized baseline for internal references.

[0068] 2. Determination of the correction coefficient

[0069] (1) The normalized baseline BR of the intrinsic parameter i calculated above i Calculate the initial normalization coefficient of intrinsic parameter i in sample n: k' ni =R ni / BR i And calculate the initial correction coefficients for all intrinsic parameters in sample n.

[0070] (2) The correction coefficient q of the internal reference i i ∈[0.5,0.6,0.7,0.8,0.9,1]. The final combination of correction coefficients for the intrinsic parameters is determined using an exhaustive method: First, a correction coefficient is randomly selected for each different intrinsic parameter, and then the normalized values ​​of all intrinsic parameters in sample n are calculated. The number of all intrinsic parameter reads in sample n is calculated using K2. n Perform normalization analysis, and compare the normalized number of internal reference reads with the baseline BR. i Perform error analysis to obtain the root mean square error; select the combination of internal parameter correction coefficients with the smallest root mean square error as the optimal combination of correction coefficients for different internal parameters.

[0071] 3. Application

[0072] The above-obtained normalized baseline BR i The optimal intrinsic parameter correction coefficients are combined to form the final correction library.

[0073] During amplicon data analysis, the baseline BR in the correction library is used. i Calculate the normalization coefficient k' i And K1', combined with the intrinsic parameter correction coefficient q i Calculate the final normalized value K2 for the sample, and finally normalize the sample data.

[0074] Example 2: Amplicon Internal Reference Normalization System

[0075] Using the correction library obtained in Implementation 1, this embodiment constructs an amplicon internal reference normalization system, such as... Figure 1 As shown, it includes:

[0076] The data input module is used to accept sequencing data input;

[0077] The correction library module stores the normalized baseline and optimal correction coefficients of the intrinsic parameters obtained in Example 1.

[0078] The normalization module, connected to both the data input module and the correction library module, is used to perform normalization using the combination of the intrinsic parameter normalization baseline and the optimal intrinsic parameter correction coefficients as the correction library.

[0079] (1) Calculate the initial normalization coefficient of the intrinsic parameter i in the sample: k′ i =R i / BR i And calculate the initial correction coefficients for all intrinsic parameters in the sample. Among them, R i BR represents the number of reads of intrinsic parameter i in the sample. i Let i be the normalized baseline of the intrinsic parameter i, where i = 1, ..., j, and j is the number of intrinsic parameters in the sample.

[0080] (2) Using the optimal combination of intrinsic parameter correction coefficients, calculate the normalized values ​​of all intrinsic parameters in the sample. Where, q′ i This is the optimal correction coefficient for the internal reference;

[0081] S33, normalization is performed using the K2 value;

[0082] The output module, connected to the normalization module, is used to output the normalized data.

[0083] The correction library module can also be connected to the data input module, such as... Figure 2 As shown. In this way, new correction libraries can be created for different combinations of intrinsic parameters, or the correction libraries can be updated.

[0084] Example 3 Sequencing Data Analysis

[0085] (1) The situation where the internal reference is stable

[0086] Assume the product being tested contains three internal parameters a, b, and c, and one amplicon d to be tested. Under the minimum total data volume A set for the product, the number of reads for internal parameters a, b, and c are 1000, 2000, and 30000, respectively, and the number of reads for amplicon d is 10000.

[0087] Ideally (with absolute stability among the intrinsic parameters), when the total data volume is B (1.5 times A), the number of reads for intrinsic parameters a, b, and c should theoretically be 1500, 3000, and 45000 respectively, and the number of reads for amplicon d should theoretically be 15000. In this case, the amplitudes (normalization coefficients) of the three intrinsic parameters from the total data volume A to B are: a: 1500 / 1000 = 1.5, b: 3000 / 2000 = 1.5, c: 45000 / 30000 = 1.5.

[0088] If the method of Example 1 is used, the sample normalization coefficient is: (1.5 + 1.5 + 1.5) / 3 = 1.5 (the internal reference correction coefficients for a, b, and c are all 1). After data B is normalized based on a normalization coefficient of 1.5 (divided by 1.5), the quantification of all amplicon data under data B is converted to the scale of data A: a = 1500 / 1.5 = 1000; b = 3000 / 1.5 = 2000; c = 45000 / 1.5 = 30000. Similarly, after other amplicon data are converted, the quantification of other amplicon reads can be compared. After normalization, d is 15000 / 1.5 = 10000. If the total amount of intrinsic data is used for normalization, the sample normalization coefficient is (1500+3000+45000) / (1000+2000+30000)=1.5. Then the data results of the total data normalization are consistent with those of the normalization method in Example 1.

[0089] (2) Cases where internal references are unstable

[0090] Assuming the total data volume C (equivalent to B+c*0.1) is 20% higher than the intrinsic parameter c (experimental error), then the number of reads for a, b, and c are 1500, 3000, and 54000 respectively, and d is 15000. Theoretically, after normalization, they should be equivalent to A.

[0091] If the method of Example 1 is used:

[0092] The sample normalization coefficient is ((1500 / 1000)+(3000 / 2000)+(54000 / 30000)) / 3=1.6. The normalization coefficient is 1.6, and the normalization coefficient error is 1.6 / 1.5=1.067. d=15000 / 1.6=9375, and the error of d is 1-(9375 / 1000)=0.065. If d is triploid (15000*1.5), then the normalized value is 14063.

[0093] If we use the total amount of intrinsic parameter data for normalization:

[0094] The normalization coefficient is (1500+3000+54000) / (1000+2000+30000)=1.773. The normalization coefficient error is 1.773 / 1.5=1.182. d=15000 / 1.773=8460, and the error of d is 1-(8460 / 10000)=0.154. If d is triploid (15000*1.5=22500), then the normalized value is 12690.

[0095] At this point, it can be seen that the expression level (reads) of d should have been 15000. Due to the systematic error introduced by the internal reference c, the expression level of d after internal reference normalization was deviated. After normalization by the method in Example 1, it was 14063, while the commonly used method of normalizing total internal reference data was 12690. For d, in the case of triploid CNV, d should be 15000, and the judgment value for detecting CNV should be between 10000 and 15000. However, 12690 is already in the middle, making it difficult to distinguish whether the result is due to a low 15000 or a high 10000, thus making it difficult to determine whether it is CNV positive. 14063, on the other hand, makes it easier to accurately determine whether the result is due to a low 15000 rather than a high 10000.

[0096] Therefore, while the error increase from the two sample normalization methods may seem small, they can improve detection sensitivity. Furthermore, for the same sample and the same amplicon, repeating the process several times could result in a difference of up to 100% in the actual amplicon quantification.

[0097] Furthermore, assuming the correction coefficients for the intrinsic parameters a, b, and c are 1, 1, and 0.5 respectively, then for the correction of data volume C, the sample normalization coefficient using the method in Example 1 is:

[0098] K1=((1500 / 1000)+((3000 / 2000)+(54000 / 30000)) / 3=1.6,

[0099] Recalculate:

[0100] K2=(((1500 / 1000)+(1.6-(1500 / 1000))×(1-1))+((3000 / 2000)+(1.6-(3000 / 2000))×(1-1))+((54000 / 30000)+(1.6-(54000 / 30000))×(1-0.5)))=1.567.

[0101] After normalization, the values ​​of a, b, c, and d are 1500 / 1.567, 3000 / 1.567, 54000 / 1.567, and 15000 / 1.567, respectively, which are 957.2, 1914, 34461, and 9572. If d is triploid (15000*1.5=22500), then the normalized value of the included intrinsic parameter correction coefficient is 14358 (error 1-14358 / 15000=0.0428), which is closer to the simulated 15000 than the 14063 (error 1-14063 / 15000=0.062467) without intrinsic parameter correction coefficient normalization. This is far more accurate than the quantitative value of 12690 (error 1-12690 / 15000=0.154) after normalization using the total intrinsic parameter data method. For the total intrinsic parameter data method, even the assumed CNV case cannot be detected and distinguished, while the method in Example 1 makes more sensitive detection possible.

[0102] Example 4: Analysis of Actual Sequencing Data

[0103] This embodiment provides normalized sequencing data for 5 internal references. The data before normalization are shown in Table 1:

[0104] Table 1. Expression levels of internal references before normalization.

[0105]

[0106] Normalization was performed using the method in Example 2, and the normalized data are shown in Table 2:

[0107] Table 2. Normalized expression levels of internal references.

[0108]

[0109]

[0110] All documents mentioned in this invention are incorporated herein by reference as if each document were individually incorporated by reference. Furthermore, it should be understood that after reading the foregoing teachings of this invention, those skilled in the art can make various alterations or modifications to this invention, and these equivalent forms also fall within the scope defined by the appended claims.

Claims

1. A method for normalizing amplicon internal controls, characterized in that, This includes the steps for obtaining the normalized baseline of the internal references, specifically: S11, obtained m Sequencing data for one sample, including intrinsic parameters of [missing information]. j One, obtain internal reference i In the sample n The number of reads and obtain internal reference i In all m The representative number of reads in a sample is denoted as . ,in i =1、…… j , n =1, 2, ..., m ; S12, Calculate internal parameters i In the sample n Initial normalized values ​​in: Next, calculate the sample n Initial normalized values ​​for all intrinsic parameters: Then calculate the internal parameters. i In the sample n The initial normalized number of reads is ; S13, Calculate internal parameters i In all m Coefficient of variation of the number of reads after initial normalization in a sample ,like If the sample exceeds the prediction threshold, remove the abnormal sample. S14, repeat steps S11-S13 for the remaining samples until... Obtain internal parameters if the threshold is not exceeded. i The representative number of the number of reads in the remaining samples is denoted as... As an internal reference normalization baseline, The process further includes the step of obtaining the internal reference correction coefficient, specifically including: S21, Calculate internal parameters i In the sample n Initial normalization coefficients in: and calculate the sample n Initial correction coefficients for all internal parameters ; S22, Internal Reference i Correction coefficient q i Select one of 0.5, 0.6, 0.7, 0.8, 0.9, and 1, and use an exhaustive method to determine the final correction coefficient of the internal reference: S221, For different intrinsic parameters, randomly select a correction coefficient to obtain the combination of intrinsic parameter correction coefficients, and calculate the sample. n Normalized values ​​of all intrinsic parameters For the sample n The number of all internal parameter reads is used Normalization involves comparing the normalized number of intrinsic parameter reads with the normalized baseline of the intrinsic parameter. Perform error analysis to obtain the root mean square error; S222, perform the same operation as step S221 on all combinations of internal parameter correction coefficients to obtain different root mean square errors; S223, Select the set of internal parameter correction coefficients with the smallest mean square error as the optimal combination of internal parameter correction coefficients: ... ,in, For internal reference i The optimal correction coefficient, The method further includes the step of normalizing using the internal parameter normalized baseline and the optimal internal parameter correction coefficient combination as a correction library: S31, Calculate internal parameters i Initial normalization coefficients in the sample: And calculate the initial correction coefficients for all intrinsic parameters in the sample. ,in, For internal reference i The number of reads in the sample. i =1、…… j , j This represents the number of internal parameters in the sample. S32, using the optimal combination of intrinsic parameter correction coefficients, calculate the normalized values ​​of all intrinsic parameters in the sample. ; S33, normalization is performed using the K2 value.

2. The method for normalizing amplicon internal controls according to claim 1, characterized in that, The preset threshold is obtained using group data. CV The representative number of the value.

3. The method for normalizing amplicon internal controls according to claim 2, characterized in that, The representative numbers refer to the median, mean, mode, 1 / 4 quantile, or 3 / 4 quantile.

4. A system for amplicon internal reference normalization, comprising the following modules: The data input module is used to accept sequencing data input; The correction library module is used to store the normalized baseline of the intrinsic parameters and the optimal combination of intrinsic parameter correction coefficients; The normalization module, connected to both the data input module and the correction library module, is used to perform normalization using the combination of the intrinsic parameter normalization baseline and the optimal intrinsic parameter correction coefficients as the correction library. (1) Calculate internal parameters i Initial normalization coefficients in the sample: And calculate the initial correction coefficients for all intrinsic parameters in the sample. ,in, For internal reference i The number of reads in the sample. For internal reference i The normalized baseline, i =1、…… j , j This represents the number of internal parameters in the sample. (2) Using the optimal combination of intrinsic parameter correction coefficients, calculate the normalized values ​​of all intrinsic parameters in the sample. ,in, This is the optimal correction coefficient for the internal reference; S33, normalization is performed using the K2 value; The output module, connected to the normalization module, is used to output the normalized data. The correction library module is connected to the data input module and is used to obtain the internal reference normalized baseline using sequencing data from multiple samples. (1) Obtain m Sequencing data for one sample, including intrinsic parameters of [missing information]. j One, obtain internal reference i In the sample n The number of reads and obtain internal reference i In all m The representative number of reads in a sample is denoted as . ,in i =1、…… j , n =1, 2, ..., m ; (2) Calculate internal parameters i In the sample n Initial normalized values ​​in: Next, calculate the sample n Initial normalized values ​​for all intrinsic parameters: Then calculate the internal parameters. i In the sample n The initial normalized number of reads is ; (3) Calculate internal parameters i In all m Coefficient of variation of the number of reads after initial normalization in a sample ,like If the sample exceeds the prediction threshold, remove the abnormal sample. (4) Repeat steps S11-S13 for the remaining samples until... Obtain internal parameters if the threshold is not exceeded. i The representative number of reads in the remaining samples is the... , The correction library module is further used to obtain the optimal combination of internal reference correction coefficients using sequencing data from multiple samples: (1) Calculate internal parameters i In the sample n Initial normalization coefficients in: and calculate the sample n Initial correction coefficients for all internal parameters ; (2) Internal reference i Correction coefficient q i Select from one of 0.5, 0.6, 0.7, 0.8, 0.9, and 1, and use an exhaustive method to determine the final combination of correction coefficients for the internal references: (2-1) Randomly select a correction coefficient for each different intrinsic parameter to obtain the combination of intrinsic parameter correction coefficients, and calculate the sample. n Normalized values ​​of all intrinsic parameters For the sample n The number of all internal parameter reads is used Normalization involves comparing the normalized number of intrinsic parameter reads with the normalized baseline of the intrinsic parameter. Perform error analysis to obtain the root mean square error; (2-2) Perform the same operation as step S221 on all combinations of internal parameter correction coefficients to obtain different root mean square errors; (2-3) Select the combination of internal parameter correction coefficients with the smallest mean square error as the optimal combination of internal parameter correction coefficients: ... .

5. A computer device, characterized in that, include: Memory, used to store computer programs; A processor for implementing the steps of the method as described in any one of claims 1 to 3 when executing the computer program.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1-3.

Images