Multi-mode early prediction method and prediction device for panthenic cancer based on cfDNA methylation and electronic equipment

Abstract

The method comprises the following steps: respectively collecting cfDNA (cell-free deoxyribonucleic acid) samples of i cancer people and healthy people, and extracting methylation data; combining the CpG sites based on the methylation data to obtain a plurality of methylation intervals, screening from the plurality of methylation intervals to obtain a plurality of candidate marker intervals, and further screening the plurality of candidate marker intervals; training a first prediction model by using the plurality of candidate marker intervals, extracting fragmentation features of chromosomes based on the reference genome to train a second prediction model, and extracting methylation entropy features of the chromosomes to train a third prediction model; a multi-mode panthenic cancer early-stage prediction model is trained through predicted values of cancer crowds of the first prediction model, the second prediction model and the third prediction model, and accurate combined early screening of lung cancer, intestinal cancer, gastric cancer, liver cancer, esophageal cancer, thyroid cancer and ovarian cancer can be provided.

Description

Technical Field

[0001] The present application belongs to the field of molecular biomedical technology, and specifically relates to a multimodal pan-cancer early prediction method, prediction device and electronic equipment based on cfDNA methylation. Background Art

[0002] Cancer is one of the leading causes of death among residents worldwide and a major public health issue that seriously endangers human health. A large number of relevant studies have shown that early detection and timely diagnosis can effectively improve the survival rate of patients with various cancers. In the field of cancer detection, compared with single cancer screening, multi-cancer screening is often more efficient. This is because people who meet the conditions for cancer screening usually have a potential risk of developing multiple cancers, and it is difficult for them to simply determine the possible site of disease based on their own condition before the examination, especially cancers occurring in the abdominal cavity or blood. cfDNA comes from DNA released after cells are damaged and ruptured, and circulates in the blood. The tumor genome usually carries characteristic gene mutation sites. By detecting mutation sites in cfDNA, especially methylation sites, cancer detection can be achieved.

[0003] In order to continuously improve the performance of cfDNA methylation in cancer detection, researchers have made great efforts in technological innovation and detection model construction. Although some achievements have been made in the early detection of cancer using methylation panels, the accuracy of joint detection of multiple cancers needs to be improved. Based on cfDNA sequencing data, a multimodal detection model is constructed by integrating multiple features such as methylation, fragmentation, and methylation entropy to pay more attention to sample data from the test population, which can effectively help improve the versatility and accuracy of multi-cancer detection. Therefore, this application designs a multimodal pan-cancer detection method based on methylation panel sequencing data through multi-dimensional analysis. Summary of the invention

[0004] In order to solve the problems existing in the prior art, the purpose of the present application is to provide a multimodal pan-cancer early prediction method, prediction device and electronic device. On the basis of cfDNA methylation difference analysis, fragmentation analysis of methylation sequencing data and detection analysis of methylation entropy are introduced, and a multimodal pan-cancer early prediction model is obtained by integrating multiple analysis methods. The model is used to analyze the sequencing samples of the test population to achieve high-accuracy joint detection of seven cancers, including lung cancer, intestinal cancer, gastric cancer, liver cancer, esophageal cancer, thyroid cancer, and ovarian cancer, providing a more accurate and feasible solution for pan-cancer early screening.

[0005] Specifically, this application involves the following aspects:

[0006] 1. A multimodal pan-cancer early prediction method, comprising: collecting multiple cfDNA samples from i types of cancer populations and healthy populations, respectively, and extracting methylation data of the multiple cfDNA samples; merging CpG sites based on the methylation data of the multiple cfDNA samples to obtain multiple methylation intervals, extracting differential methylation intervals between each cancer population and healthy population in the i types of cancer populations from the multiple methylation intervals, taking the differential methylation intervals that appear in no less than m types of cancer populations among the obtained multiple differential methylation intervals as candidate marker intervals, and screening the obtained multiple candidate marker intervals; training a first prediction model using the screened multiple candidate marker intervals, training a second prediction model based on the fragmentation features of chromosomes extracted from a reference genome, and training a third prediction model by extracting the methylation entropy features of chromosomes; training a multimodal pan-cancer early prediction model by using the cancer population prediction values ​​of the first prediction model, the second prediction model, and the third prediction model for pan-cancer early detection.

[0007] 2. According to the multimodal pan-cancer early prediction method of item 1, extracting the methylation data of multiple cfDNA samples separately includes: extracting multiple CpG sites of the multiple cfDNA samples and the methylation value of each CpG site in the multiple CpG sites; merging CpG sites based on the methylation data of multiple cfDNA samples includes: calculating the difference in methylation value of each CpG site between each cancer population and healthy population in i types of cancer populations respectively; merging corresponding CpG sites in response to the difference being not zero.

[0008] 3. According to the multimodal pan-cancer early prediction method of item 1, screening the obtained multiple candidate marker intervals includes: calculating the importance value of each candidate marker interval in the multiple candidate marker intervals, eliminating the corresponding candidate marker interval in response to the importance value being not greater than a first threshold, and retaining the corresponding candidate marker interval in response to the importance value being greater than the first threshold.

[0009] 4. A multimodal pan-cancer early prediction method according to item 1, wherein the second prediction model is trained based on the fragmented features of chromosomes extracted based on the reference genome, including: tiling the reference genome autosomes and evenly dividing them into multiple first intervals; recording the number of short-length fragments, medium-length fragments and long-length fragments in each of the multiple first intervals; sequentially merging the multiple first intervals to obtain multiple second intervals, calculating the coverage of each of the multiple second intervals, and performing dimensionality reduction on the obtained multiple coverages; and constructing a second prediction model through the multiple coverages after dimensionality reduction.

[0010] 5. According to the multimodal pan-cancer early prediction method of item 4, wherein the coverage is the short-length fragments, medium-length fragments and long-length fragments contained in each of the multiple second intervals; and the second prediction model is constructed using elastic network regression through the multiple coverages after dimensionality reduction.

[0011] 6. According to the multimodal pan-cancer early prediction method of item 1, extracting the methylation entropy features of chromosomes to train the third prediction model includes: extracting multiple insertion fragments from the methylation data of multiple cfDNA samples to obtain methylation pattern entropy values ​​of the multiple insertion fragments; calculating the methylation pattern entropy value of each chromosome respectively; and constructing the third prediction model using logistic regression through the methylation pattern entropy values ​​of all chromosomes.

[0012] 7. According to the multimodal pan-cancer early prediction method of item 6, wherein the extraction of multiple insert fragments from the methylation data of multiple cfDNA samples includes: extracting the insert fragments of the cfDNA samples read from left to right and read from right to left in the CpG mode, both of which can be compared with the reference genome; the formula for obtaining the methylation pattern entropy value of the multiple insert fragments is as follows:

[0013]

[0014] Where BiEn(s) is the methylation pattern entropy value of the inserted fragment, n represents the number of all CpG sites in the inserted fragment, k is a set of values ​​from 0 to n-2, and p is the probability value of k under a given value.

[0015] 8. According to the multimodal pan-cancer early prediction method of item 6, calculating the methylation pattern entropy value of each chromosome separately includes: calculating the average of the methylation pattern entropy values ​​of all inserted fragments on each chromosome separately as the methylation pattern entropy value of each chromosome.

[0016] 9. A multimodal pan-cancer early prediction device, comprising: a data collection unit, for collecting multiple cfDNA samples from i types of cancer populations and healthy populations, respectively, and extracting methylation data of the multiple cfDNA samples respectively; a marker screening unit, for merging CpG sites based on the methylation data of the multiple cfDNA samples to obtain multiple methylation intervals, extracting differential methylation intervals between each cancer population and the healthy population in the i types of cancer populations from the multiple methylation intervals, taking the differential methylation intervals that appear in no less than m types of cancer populations among the obtained multiple differential methylation intervals as candidate marker intervals, and screening the obtained multiple candidate marker intervals; a model construction unit, for training a first prediction model using the screened multiple candidate marker intervals, for training a second prediction model based on the fragmentation features of chromosomes extracted from a reference genome, and for training a third prediction model by extracting the methylation entropy features of chromosomes; a model integration unit, for training a multimodal pan-cancer early prediction model through the cancer population prediction values ​​of the first prediction model, the second prediction model, and the third prediction model for pan-cancer early detection.

[0017] 10. An electronic device, comprising: a processor; and a memory, wherein computer program instructions are stored in the memory, and when the computer program instructions are executed by the processor, the processor executes the multimodal pan-cancer early prediction method as described in any one of items 1-8.

[0018] The multimodal pan-cancer early prediction method of the present application integrates information from different modalities, such as the methylation data and chromosome fragment data mentioned in the present application. The model can capture data features in more dimensions, which can not only help the model better cope with noise and interference during prediction and enhance the generalization ability of the model, but also reveal hidden associations between different modalities to further improve the prediction accuracy of the model on the basis of a single modality, help doctors assess the risk of cancer in test subjects, and provide them with personalized diagnosis and treatment plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 The figure illustrates a flow chart of a multimodal pan-cancer early prediction method according to an embodiment of the present application.

[0020] Figure 2 The figure shows a flow chart of constructing three prediction models using three modality features in a multimodal pan-cancer early prediction method according to an embodiment of the present application.

[0021] Figure 3 The figure shows a schematic diagram of the ROC curve of the multimodal pan-cancer early prediction method on the test sample set according to an embodiment of the present application.

[0022] Figure 4 The figure shows a block diagram of a multi-modal pan-cancer early prediction device according to an embodiment of the present application.

[0023] Figure 5 A block diagram of an electronic device according to an embodiment of the present application is illustrated. DETAILED DESCRIPTION

[0024] Methylation data

[0025] The methylation level of DNA or RNA molecules is measured and analyzed by various technical means. Methylation data are particularly important in cancer research, as they can reveal the epigenetic mechanisms behind changes in gene expression and help understand the occurrence and development of cancer.

[0026] Methylation value

[0027] An indicator used to measure DNA methylation levels, commonly used in biostatistics and genomics research. The methylation value reflects the proportion of methylated cytosine at a specific CpG site, ranging from 0 to 1, where 0 indicates complete unmethylation and 1 indicates complete methylation. In practical applications, the methylation value can be used to compare methylation differences between different samples and different cell types, and is of great significance in disease diagnosis and gene expression regulation research.

[0028] The present application is further described below in conjunction with examples. It should be understood that the examples are only used to further describe and illustrate the present application and are not used to limit the present application.

[0029] Unless otherwise defined, the technical and scientific terms in this specification have the same meaning as those generally understood by those skilled in the art. Although similar or identical methods and materials as described herein may be used in experiments or practical applications, materials and methods are described herein below. In the event of a conflict, the present specification including the definitions therein shall prevail. In addition, materials, methods and examples are provided for illustration only and are not restrictive. The present application is further described below in conjunction with specific embodiments, but is not intended to limit the scope of the present application.

[0030] Application Overview

[0031] As mentioned above, in the field of cancer detection, due to the heterogeneity of cfDNA abnormalities in different cancer types, subtypes, stages and etiologies, the technology of using cfDNA sequencing data for genetic characterization of pan-cancer features will have a certain degree of false negatives and false positives. Due to the inconsistent detection standards of tumor tissue and ctDNA, the consistency data given by existing different studies vary greatly, and it is difficult to provide favorable information for cancer feature extraction, making it difficult to advance improved experiments on effective biomarker combinations for specific cancer types, thus making cfDNA testing, such as cfDNA methylation testing, limited in predicting individual cancer. In addition, in the early stages of some cancers or in the stage of minimal residual lesions, the content of ctDNA in plasma is extremely low, which greatly increases the cost and difficulty of detection.

[0032] Patent CN114045345A proposes a genome cancer information detection system that uses enzyme transfer to perform whole-genome methylation sequencing on plasma sample cfDNA, analyzes the methylation density, fragment length distribution, fragment 5' end motif and / or chromosome stability of the genome, and simultaneously achieves early detection and screening of multiple cancers. However, this monitoring system ignores capturing subtle changes in DNA or RNA molecules on the genome, including changes in fragment length, deletion or increase of specific sequences, etc. These fragmentation features often contain overall changes in the entire genome or transcriptome, and their analysis does not rely on specific cancer markers or genes. This makes it difficult for this monitoring system to provide strong support for pan-cancer detection.

[0033] Patent CN116356021A provides GutSeer, a multi-cancer early screening and localization technology for five digestive system cancers with a high mortality rate, and proves that a relatively small second-generation sequencing panel can be used to utilize multiple dimensional features including methylation, copy number changes, and terminal motifs to achieve more accurate cancer detection.

[0034] Methylation entropy is the entropy value generated during the DNA methylation process, that is, the degree of confusion or uncertainty of the DNA methylation state. Since methylation entropy is an indicator to measure the complexity and stability of DNA methylation patterns, it can be used as a specific indicator to reflect the distribution and variation of methylation sites on cfDNA molecules, which makes the methylation entropy feature helpful to help the prediction model understand the differences in methylation patterns between target cancer patients, thereby learning the heterogeneity between these cancers. Therefore, relative to the detection method constructed by methylation panel sequencing features of CN116356021A, the present application proposes to construct a multimodal detection model based on the methylation rate features of cfDNA and the genomic fragmentation features and the methylation entropy features, so that the prediction method, device and electronic device of the present application are adapted to the detection of cancer types not limited to the digestive system, but cover more cancers in multiple systems, and can achieve a wider range of high-accuracy joint detection compared to the above-mentioned method or system.

[0035] Specifically, the present application provides a multimodal pan-cancer early prediction method, prediction device and electronic device, which respectively collect cfDNA samples from i types of cancer populations and healthy populations, and extract methylation data of the cfDNA samples respectively; based on the methylation data of the cfDNA samples, the CpG sites in the cfDNA samples are merged to obtain multiple methylation intervals, and multiple differential methylation intervals between each cancer population and the healthy population in the i types of cancer populations are screened from the multiple methylation intervals, and the differential methylation intervals that appear in no less than m types of cancer populations in the multiple differential methylation intervals are extracted as multiple candidate marker intervals, and the multiple candidate marker intervals are screened; a first prediction model is trained using the multiple candidate marker intervals, and at the same time, a second prediction model is trained by extracting fragmentation features on a reference genome, and a third prediction model is trained by extracting methylation entropy features of chromosomes; a multimodal pan-cancer early prediction model is trained by the cancer population prediction values ​​of the first prediction model, the second prediction model and the third prediction model for pan-cancer early detection.

[0036] By screening multiple differential methylation intervals between i cancer populations and healthy populations, and further extracting differential methylation intervals that appear simultaneously in cfDNA samples of cancer populations belonging to multiple types of cancer as multiple candidate marker intervals in these differential methylation intervals, these candidate marker intervals can better reflect the overall genetic characterization characteristics of cancer in i cancer populations, and provide a basic predictive model for the joint detection of i cancers; by extracting the fragmentation characteristics of the reference genome, we can capture subtle changes in genomic DNA fragments, which are closely related to the occurrence and development of multiple cancers, thereby improving the sensitivity and accuracy of detecting each cancer, and at the same time providing rich biological information to reveal the heterogeneity between i cancers, and providing a predictive model for cancer differentiation between i cancers.

[0037] In addition, entropy reflects the degree of disorder of a system. The more disordered a system is, the greater its entropy is; and the more ordered a system is, the smaller its entropy is. For the methylation pattern entropy value, it refers to the overall state of the methylation pattern of the CpG site on the inserted fragment, reflecting the distribution and variability of the methylation state of the CpG site, and can be used to evaluate the epigenetic heterogeneity in the cell population where the cfDNA is located. Therefore, the methylation pattern entropy features are extracted to comprehensively evaluate the complexity and stability of the DNA methylation pattern and accurately identify methylation abnormalities associated with i types of cancer. The prediction model constructed using the methylation entropy features can help the prediction model based on the methylation panel data features and the genomic fragment features to further improve the accuracy of the overall detection and typing of the above cancers, thereby providing a highly feasible pan-cancer early screening method.

[0038] After introducing the basic principles of the present application, various non-limiting embodiments of the present application will be described in detail with reference to the accompanying drawings.

[0039] Exemplary Methods

[0040] Figure 1 The figure illustrates a multimodal pan-cancer early prediction method according to an embodiment of the present application.

[0041] like Figure 1 As shown, the multimodal pan-cancer early prediction method according to the embodiment of the present application includes the following steps.

[0042] S110, collect cfDNA samples of i kinds of cancer populations and healthy populations, respectively, and extract methylation data of cfDNA samples. In the present application, methylation data refers to the data obtained by conventional analysis of fastq data obtained using methylation panel, preferably a processable bam file. According to the multimodal pan-cancer early prediction method of the embodiment of the present application, up to 7 cancer populations can be included, namely lung cancer, intestinal cancer, gastric cancer, liver cancer, esophageal cancer, thyroid cancer, and ovarian cancer. Based on these 7 cancer populations, all samples of the training set in the cancer samples obtained by sequencing the methylation panel of its cfDNA sample are candidate groups, and all healthy samples are control groups, and methylation intervals are screened. It can be understood that methylation data can be data obtained by further screening the methylation sequencing data obtained by the methylation panel, such as methylation sequencing data, data quality preprocessing and evaluation (fastp software), genome alignment (Bismark software), or methylation sequencing data after removing duplicate data brought by sample / experimental technology. The reference genome used for sequencing is the human genome. In the art, there are multiple versions of human genome gene sequencing versions. The commonly used version is hg19. Technicians in this field can select the appropriate version.

[0043] In this way, the methylation data samples of the 7 cancer populations obtained can be the location information of all CpG sites and the methylation values ​​of these CpG sites obtained from all cancer population samples and healthy population samples through methylation panel sequencing; the CpG sites can be CpG sites on the cancer high-correlation target interval obtained according to existing public research, public detection or sequencing products, such as sites on the interval used by Borcheng's seven cancer NGS methylation detection kit, or CpG sites on the entire genome. These CpG sites can be merged according to certain rules to form a methylation interval containing multiple CpG sites, such as an interval with a high methylation level or an interval with a low methylation level, and these methylation intervals may therefore have differentiated methylation status performance in cancer patients and healthy people, thereby becoming favorable methylation markers.

[0044] S120, based on the methylation data of the cfDNA sample, CpG sites are merged to obtain multiple methylation intervals, and multiple differential methylation intervals between each cancer population and the healthy population in the i cancer populations are screened from the multiple methylation intervals, and the differential methylation intervals that appear in no less than m cancer populations are extracted from the multiple differential methylation intervals as multiple candidate marker intervals, and the multiple candidate marker intervals are screened. As can be seen from the above, the position information and methylation values ​​of all CpG sites of the two populations are obtained through step S110, and the expression difference of each CpG site in the two populations can be obtained by calculating the difference of the methylation values ​​of the cancer population and the healthy population at each CpG site; wherein, the difference of the methylation values ​​can be the difference between the average methylation value of each CpG site in a cancer population and the average methylation value of the healthy population.

[0045] In this way, differentiated CpG sites are obtained, and differentiated methylation intervals can be constructed based on this. A feasible method for merging differentiated CpG sites to obtain methylation intervals is to merge the corresponding CpG sites in response to each CpG site whose methylation value difference is not zero, that is, to merge these adjacent differentiated CpG sites in sequence. It should be noted that the methylation interval obtained in this way may have the problem of being too long. Since the length of human cfDNA is generally around 167bp, it is necessary to further set the maximum length of the methylation interval so that it does not significantly exceed the length of cfDNA. For example, the maximum length of the methylation interval can be set to 200bp, so as to ensure that each methylation interval is close to 167bp of cfDNA, so that the methylation level of the CpG site in the calculated interval is closer to the actual level, ensuring data quality. In addition, since the fewer the number of CpG sites in a given interval, the higher the false positive probability of the methylation value of the CpG site obtained by the existing technology, the number of CpG sites in each methylation interval can be additionally set to at least 3, and the CpG sites in the methylation interval can be screened by any feasible CpG site outlier processing or missing value processing method, so as to further improve the robustness of the obtained methylation interval.

[0046] In some embodiments, screening from multiple methylation intervals to obtain multiple differential methylation intervals between each cancer population and a healthy population in i cancer populations includes the following screening method: specifying that the detection sensitivity of the methylation interval detection for each cancer in i cancer populations and a healthy population under a given detection specificity is not less than a given threshold. When the specificity and sensitivity of a given detection are maintained at a certain level, for example, in a preferred embodiment, the sensitivity is guaranteed to be no less than 70%, so that more than 70% of the methylation values ​​of the methylation intervals from the cancer population have population differences, and the specificity is guaranteed to be no less than 80%, so that more than 80% of the methylation values ​​of the methylation intervals of the healthy population control group have population differences, so that differential methylation intervals with higher sensitivity and specificity can be screened, and the influence of outlier intervals in the methylation intervals on the differences is reduced, thereby ensuring the stability of the differences.

[0047] In addition, the following screening method can be further included: specifying that the detection AUC of the corresponding cancer population and the healthy population based on the methylation interval detection of each cancer in the i types of cancer is not less than a given threshold; and the difference between the average value of the methylation values ​​of all methylation sites in the methylation interval of each cancer population and the healthy population is not less than a given threshold, for example, in a preferred embodiment, it is ensured that the AUC area of ​​the cancer population detected by the methylation interval is not less than 0.7, and the absolute value of the difference between the average value of the methylation value of the methylation site in the methylation interval in the cancer population and the healthy population is not less than 0.02. In this way, it is possible to avoid the generation of methylation direction interference on the basis of determining that the differential methylation interval obtained by screening has a good population classification performance, that is, the methylation level of each CpG site in these differential methylation intervals of the two populations is different, and the methylation direction of the difference is not specified, so as to ensure that some differential methylation intervals will not be screened out because the total difference of the CpG sites is too small or zero.

[0048] In this way, multiple differential methylation intervals with high population classification performance are obtained from multiple methylation intervals through one or more of the aforementioned multiple screening methods. Since these differential methylation intervals are used for pan-cancer detection of i different types of cancer, in order to improve the detection speed and the generalization performance of the constructed prediction model, it is also necessary to screen and obtain candidate marker intervals with wider population adaptability relative to a single cancer population. Specifically, in some favorable embodiments, differential methylation intervals that appear in no less than m types of cancer populations from multiple differential methylation intervals can be extracted as multiple candidate marker intervals, and the value of m can be 2, 3, 4, 5, 6 or 7; for example, differential methylation intervals that appear in cfDNA samples of at least 2 cancer populations are selected as candidate marker intervals, so that the candidate marker intervals can show good classification performance for at least two cancer types.

[0049] Considering that the candidate marker intervals obtained by merging the CpG sites based on a larger range of targeted intervals or the whole genome may still have a high data dimension even after being screened by one or more of the aforementioned screening methods, which places too much burden on the prediction model to learn classification features and optimize parameters, the multimodal pan-cancer early prediction method according to the embodiment of the present application also includes screening the candidate marker intervals, that is, interval dimensionality reduction. According to some feasible implementation schemes, the screening method may include: calculating the importance value of each candidate marker interval in a plurality of candidate marker intervals, deleting the corresponding candidate marker interval from the plurality of candidate marker intervals in response to the importance value being not greater than a first threshold, and retaining the corresponding candidate marker interval from the plurality of candidate marker intervals in response to the importance value being greater than the first threshold.

[0050] Here, the importance value is an indicator that can reflect the contribution of each candidate marker interval to the prediction result of the prediction model. The larger the importance value, the greater the influence of the candidate marker interval on the prediction result of the model. The importance value can be obtained in a variety of ways, such as by inputting multiple candidate marker intervals into a linear regression model, a Lasso model, a random forest, a gradient boosting tree, etc. to screen intervals with high importance values. According to a feasible implementation scheme, multiple candidate marker intervals are input into a random forest model and the importance value of each candidate marker interval is calculated respectively. The multiple candidate marker intervals are sorted according to the importance value and the first l candidate marker intervals whose importance value is greater than the first threshold are selected, and all candidate marker intervals starting from the l+1th are deleted; the value of l is preferably 45, so as to ensure that the classification performance and construction cost of the simple binary classification prediction model can be optimized.

[0051] Step S130, using multiple candidate marker intervals to train the first prediction model, extracting chromosome fragmentation features based on the reference genome to train the second prediction model, and extracting chromosome methylation entropy features to train the third prediction model. Here, step S130 can be further divided into three prediction steps of the first prediction model, the second prediction model, and the third prediction model.

[0052] Step S1301, train a first prediction model using multiple candidate marker intervals, and obtain a prediction value for the cancer population based on the first prediction model. Here, since multiple candidate marker intervals have been obtained after multiple screenings and dimensionality reduction, a concise and efficient binary classification model can be selected, and the binary classification model can be trained using the candidate marker intervals to obtain the first prediction model. Therefore, in an advantageous embodiment, a logistic regression algorithm is selected to construct a binary classification model for cancer populations and healthy populations. The computational complexity of logistic regression is relatively low, and it is suitable for processing large-scale sequencing data. In addition, it can cooperate with the optimization algorithm to quickly iterate the classification model parameters, which is suitable for real-time analysis and processing of cancer detection.

[0053] In this way, the first prediction model is obtained through the logistic regression algorithm. By providing the model with a fully connected layer or any possible classifier, it can output the probability value of each individual in the cancer population for each of the i types of cancer, which can be used as part of the cancer population prediction value for the final training of the multimodal pan-cancer early prediction model.

[0054] Step S1302, based on the reference genome, extract the chromosome fragmentation features to train the second prediction model, and obtain the cancer population prediction value based on the second prediction model. Specifically, the reference genome autosomes selected for methylation panel sequencing are tiled to divide into adjacent, non-overlapping first intervals, for example, multiple 100kb intervals. In particular, the length of ctDNA fragments in plasma samples will be shorter than the length of normal cfDNA fragments. Specifically, the distribution of ctDNA in cancer populations and cfDNA in healthy populations is different in the three types of fragments of 90-150bp, 180-220bp and 250-320bp.

[0055] Therefore, in each first interval, the short length fragment is set as short fragment, whose length can be set between 100-150bp, which is included in the first type of differential fragments; the medium length fragment is middle fragment, whose length is set between 150-260bp, including the second type of differential fragments; and the long length fragment is long fragment, whose length is set between 260-320bp, which is included in the third type of differential fragments; in this way, more differential fragments are obtained through the length type, so that the fragmentation feature has higher specificity. In addition, the above three fragments can also be integrated to obtain the overall length fragment, that is, nfrags fragment, whose interval length is between 100-320bp. Then, the number of short, middle, long, and nfrags in each first interval of each interval is counted to obtain the fragmentation feature.

[0056] Then, the first intervals are merged in sequence to obtain multiple second intervals of a certain length, for example, multiple second intervals of 1MB, and finally 2608 non-overlapping second intervals are obtained, and the coverage is defined for each second interval: j represents the jth second interval in the multiple second intervals, and the value of j can be exemplified as 1, 2, ..., 2608, then the various types of fragments short, middle, long and nfrags in the jth second interval and their corresponding numbers can be set as fragment features and corresponding feature values, for example, 10432 fragment features and corresponding numbers of fragments are obtained through 2608 second intervals. In this way, the elastic network regression algorithm can be used to construct a fragmented binary classification model of cancer population and healthy population as the second prediction model.

[0057] It can be understood that after obtaining the fragment features, the PCA method is used to reduce the dimension of the massive fragment features, and the fragment feature matrix is ​​constructed with the principal component that can explain 95% of the differences, and the second prediction model is constructed by the elastic network regression algorithm. The purpose of selecting the elastic network regression is that the number of fragment features in the second interval may be much higher than the number of the second interval, and the elastic network regression can effectively select the fragment features to avoid overfitting of the second prediction model.

[0058] In this way, the second prediction model can show better results in dealing with the high collinearity problem between chromosome segments than the first prediction model. This is because there are usually complex interactions and associations between chromosome segments, and the first prediction model constructed using candidate marker intervals does not take this into account; in addition, elastic network regression is more suitable for building sparse models to reduce the complexity and amount of calculation of the model. Therefore, the second prediction model can show higher specificity in classification and has a faster fitting speed. The specificity of the first prediction model can be effectively improved by constructing the second prediction model and integrating it with the first prediction model.

[0059] Step S1303, extract the methylation entropy features of the chromosome to train the third prediction model, and obtain the prediction value of the cancer population based on the third prediction model. Specifically, extract multiple inserts in the methylation data of the cfDNA sample in CpG mode, and obtain the methylation pattern entropy values ​​of the multiple inserts; calculate the methylation pattern entropy value of each chromosome as the methylation entropy feature. First, extract the inserts in the methylation panel sequencing data of the cfDNA sample, and filter out unqualified inserts according to certain standards.

[0060] It should be noted that extracting insert fragments refers to extracting fragments in which both the reads read from left to right and the reads read from right to left during the sequencing process can be compared with the reference genome (e.g., the same fragment on the genome), which can be exemplified by a fragment of about 300 bp in length that has been interrupted, or a fragment of about 170 bp in length for free DNA. In some advantageous embodiments, filtering unqualified insert fragments may include: removing insert fragments containing less than 3 or more than 32 CpG sites and / or removing insert fragments with missing CpG sites in the middle, etc., thereby eliminating erroneous fragments or fragments with too low methylation levels.

[0061] In particular, the CpG pattern refers to the pattern for obtaining the methylation status of the entire CpG site on the inserted fragment. Using the CpG pattern, the chromosome position, the search position of the first CpG site and the last CpG site, the CpG pattern diagram (the methylation status of the entire CpG site on the inserted fragment, that is, the pattern diagram of the methylation value size) and the frequency of occurrence of the CpG pattern can be statistically obtained. Then, the methylation pattern entropy value of each inserted fragment is calculated separately in units of chromosomes. The calculation method example is shown in the following formula:

[0062]

[0063] Wherein, BiEn(s) represents the methylation pattern entropy value of the inserted fragment, n represents the number of all CpG sites in the inserted fragment, k is a set of values ​​from 0 to n-2, and p is the probability value of k at a given value. In this way, the calculated methylation pattern entropy value can determine the disorder or randomness of the methylation pattern of the inserted fragment. The inserted fragments screened by the multimodal pan-cancer early prediction method according to the embodiment of the present application are the collective effect of the entropy of cfDNA fragments from different sources, while the methylation pattern entropy in the prior art is calculated based on the site, which is not like the inserted fragments extracted in the present application, which is the methylation pattern entropy value of the cfDNA shed from a certain tumor cell, but the entropy of several continuous CpG sites of DNA with multiple sources as a whole, so the methylation pattern entropy value of the present application will not cover up the CpG site information of the target tumor cell cfDNA.

[0064] In this way, the methylation pattern entropy value of each chromosome or specified interval can be further calculated using the following formula:

[0065] RE=1 / N*(e1+e2+e3+…+eN)

[0066] (Formula 2)

[0067] Among them, RE is a certain region, such as a certain chromosome fragment methylation entropy value, N represents the total number of inserted fragments in the region, and e represents the methylation entropy value of each inserted fragment in the region. In this way, the mean of the methylation pattern entropy values ​​of all inserted fragments of each chromosome can be used as the methylation pattern entropy value of the chromosome. Finally, with the methylation entropy value as the methylation entropy feature, a classification model of cancer population and healthy population is constructed using a logistic regression algorithm as the third prediction model.

[0068] In another embodiment, multiple target intervals can also be selected and the mean of the methylation pattern entropy values ​​of all inserted fragments in each target interval can be used as the methylation pattern entropy value of each target interval. The target interval can also be a cancer-high correlation target interval obtained based on existing public research, public detection or sequencing products, such as the high-throughput methylation sequencing target interval used in Borcheng's seven-cancer NGS methylation detection kit. In this way, the calculation cost of the methylation pattern entropy value of the inserted fragment can be reduced, data redundancy can be prevented, and the fitting speed of the third prediction model can be accelerated; it can be understood that researchers in this field can select appropriate target intervals according to actual research or production conditions and use the multimodal pan-cancer early prediction method of the implementation scheme of the fundamental application to calculate the methylation pattern entropy value of the inserted fragment.

[0069] Step S140, training a multimodal pan-cancer early prediction model through the cancer population prediction values ​​of the first prediction model, the second prediction model, and the third prediction model for pan-cancer early detection. For the case where the sample size of this application is small and the number of features is large, the basic model constructed by multiple samples, that is, the cancer population prediction values ​​of the first prediction model-the third prediction model are combined to form a cancer population prediction feature set, and then the cancer population prediction feature set is passed as input to the multimodal pan-cancer early prediction model for training, and the final prediction result is output. In this way, it can help the basic models of different modalities to be integrated, so that the multimodal pan-cancer early prediction model can adapt to different types of data and classification problems; in addition, since the sample prediction conditions of the three basic models can be fully considered during training, the overfitting risk that may exist in a single model is also reduced, and the model complementarity is used to provide a more stable classification result and improve robustness.

[0070] In an advantageous embodiment, the logistic regression algorithm can be used to construct a binary multimodal pan-cancer early prediction model, and the cancers of the i types of cancer populations can be predicted respectively. It should be noted that after the prediction value of the cancer population is obtained through steps S110-S130, that is, the prediction label of the cancer population by at least one prediction model, a prediction model of three or more classification items can be obtained through any feasible algorithm training as a multimodal pan-cancer early prediction model, or the prediction value of the cancer population can be used to fine-tune the large language model to obtain a multi-classification result, and this application does not limit this.

[0071] The present application provides a multimodal pan-cancer early prediction method based on cfDNA methylation, comprising a first prediction model constructed using the methylation level of the methylation interval detected by a methylation panel, a fragmented second prediction model constructed using genomic fragment features, and a third prediction model constructed using chromosome methylation entropy. By integrating the three modal models, the present application obtains a multimodal pan-cancer early prediction model, which can show better pan-cancer classification accuracy than any single modality prediction model in the detection of multiple cancer types, while also maintaining high sensitivity and specificity, and is suitable for early non-invasive screening of multiple tumors.

[0072] Exemplary Devices

[0073] Figure 4 FIG. 1 is a block diagram of a multimodal pan-cancer early prediction device according to an embodiment of the present application. Figure 4 As shown, the multimodal pan-cancer early prediction device 200 according to an embodiment of the present application includes:

[0074] A data collection unit 210 collects multiple cfDNA samples from i types of cancer populations and healthy populations, and extracts methylation data of the multiple cfDNA samples respectively; a marker screening unit 220 merges CpG sites based on the methylation data of the multiple cfDNA samples to obtain multiple methylation intervals, extracts differential methylation intervals between each cancer population and healthy population in the i types of cancer populations from the multiple methylation intervals, uses the differential methylation intervals that appear in no less than m types of cancer populations as candidate marker intervals, and screens the multiple candidate marker intervals; a model construction unit 230 trains a first prediction model using the screened multiple candidate marker intervals, trains a second prediction model based on the fragmentation features of chromosomes extracted from the reference genome, and trains a third prediction model by extracting the methylation entropy features of chromosomes; and a model integration unit 240 trains a multimodal pan-cancer early prediction model through the cancer population prediction values ​​of the first prediction model, the second prediction model, and the third prediction model for pan-cancer early detection.

[0075] Here, those skilled in the art can understand that the specific functions and operations of the various units and modules in the multimodal pan-cancer early prediction device 200 have been described in detail above. Figure 1 to Figure 2 The multimodal pan-cancer early prediction method has been introduced in detail in the description of the multimodal pan-cancer early prediction method, and therefore, its repeated description will be omitted.

[0076] As described above, the multimodal pan-cancer early prediction device 200 according to the embodiment of the present application can be implemented in various terminal devices, such as a server for training any prediction model and a multimodal pan-cancer early prediction model. In an example, the multimodal pan-cancer early prediction device 200 according to the embodiment of the present application can be integrated into the terminal device as a software module and / or a hardware module. For example, the multimodal pan-cancer early prediction device 200 can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the multimodal pan-cancer early prediction device 200 can also be one of the many hardware modules of the terminal device.

[0077] Alternatively, in another example, the multimodal pan-cancer early prediction apparatus 200 and the terminal device may also be separate devices, and the multimodal pan-cancer early prediction apparatus 200 may be connected to the terminal device through a wired and / or wireless network, and transmit interactive information according to an agreed data format.

[0078] Exemplary Electronic Devices

[0079] Below, reference Figure 5 To describe an electronic device according to an embodiment of the present application.

[0080] Figure 5 A block diagram of an electronic device according to an embodiment of the present application is illustrated.

[0081] like Figure 5 As shown, the electronic device 10 includes one or more processors 11 and a memory 12 .

[0082] The processor 13 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

[0083] The memory 12 may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on a computer-readable storage medium, and the processor 11 may run the program instructions to implement the multimodal pan-cancer early prediction method of each embodiment of the present application described above and / or other desired functions. Various contents such as cfDNA samples, candidate marker intervals, fragmentation features, methylation entropy features, etc. may also be stored in the computer-readable storage medium.

[0084] In one example, the electronic device 10 may further include: an input device 13 and an output device 14, and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0085] The input device 13 may include, for example, a keyboard, a mouse, etc.

[0086] The output device 14 can output various information to the outside, including the trained multi-modal pan-cancer early prediction model, etc. The output device 14 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, etc.

[0087] Of course, to simplify, Figure 5 Only some of the components related to the present application in the electronic device 10 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application situations, the electronic device 10 may also include any other appropriate components.

[0088] Example

[0089] This application provides a general and / or specific description of the materials and test methods used in the experiment. The raw materials or instruments used without indicating the manufacturer are all conventional raw materials or instruments that can be obtained through commercial purchase.

[0090] Example 1: Construction of the first prediction model

[0091] For each of the seven types of cancer samples provided by Bolcheng, including lung cancer, intestinal cancer, gastric cancer, liver cancer, esophageal cancer, thyroid cancer, and ovarian cancer, cfDNA samples were collected and divided into training set samples and test set samples. The candidate marker intervals were screened for the training set samples according to the following method:

[0092] The human genome hg19 was selected as the reference genome, and the cfDNA samples were subjected to methylation sequencing. The methylation intervals were obtained using the intervals used by the seven cancer NGS methylation detection kit of Boercheng. For the high methylation intervals, the thresholds for sensitivity (Sensitivity), the difference in methylation values ​​of CpG sites in the methylation interval (DELTA), and the significance p-value of CpG sites were set to Sensitivity ≥ 0.6, DELTA ≥ 0.02, and p-value < 0.01, respectively.

[0093] For low methylation intervals, the thresholds of sensitivity, delta, and p-value were set to sensitivity ≥ 0.55, delta ≤ -0.02, and p-value < 0.01, respectively. Then, the top 200 differentially methylated intervals were selected by sorting according to the absolute value of delta. All intervals with less than 200 were selected. Then, differentially methylated intervals that appeared in no less than 2 cancer types were selected as candidate marker intervals. Finally, 152 high methylation intervals and 313 low methylation intervals were obtained in the candidate marker intervals. Finally, the importance values ​​of the candidate marker intervals were calculated using the random forest algorithm, and 45 candidate marker intervals with importance values ​​greater than 0.0055 were selected as features for post-selection modeling.

[0094] The first prediction model of binary classification was constructed using the logistic regression algorithm. Classification prediction was performed on the test set samples, and the prediction performance of the model was evaluated. The evaluation results showed that in the test set, the AUC was 0.97, the sensitivity was 85.7%, and the specificity was 96.6%.

[0095] Example 2 Construction of the Second Prediction Model

[0096] The reference genome autosomes were tiled into adjacent, non-overlapping first intervals of 100kb in length. Short fragments were short, with a length between 100-150bp; middle fragments were middle, with a length between 150-260bp; long fragments were long, with a length between 260-320bp; nfrags were 100-320bp in length, and the number of each type of fragments, short, middle, long, and nfrags, was counted. The 100kb first intervals were merged into second intervals of 1MB in length, and 2608 non-overlapping second intervals were obtained. j represents the jth second interval, and its value on the reference genome autosomes is 1, 2, ..., 2608. The number of various fragments (short, middle, long, nfrags) in the jth second interval is the coverage of the second interval. All coverages were used as fragment features, and a total of 10432 features were obtained.

[0097] Then, the PCA method was used to reduce the dimension of the fragment features, and the fragment feature matrix was constructed with principal components that could explain 95% of the differences. The elastic network regression was used to construct a fragmented binary classification discriminant model, i.e., the second prediction model. The test set samples were classified and predicted, and the model prediction performance was evaluated. The evaluation results showed that the AUC in the test set was 0.95, the sensitivity was 79.1%, and the specificity was 98.3%.

[0098] Example 3: Construction of the third prediction model

[0099] The methylation pattern entropy value of the inserted fragments in each chromosome was calculated based on each chromosome. Specifically, all inserted fragments were extracted and unqualified inserted fragments were filtered out. The methylation pattern entropy value of each inserted fragment was used as the methylation entropy feature, and a binary classification discriminant model based on methylation entropy was constructed using a logistic regression algorithm, i.e., the third prediction model. The test set samples were classified and predicted, and the model prediction performance was evaluated. The evaluation results showed that the AUC in the test set was 0.85, the sensitivity was 65.9%, and the specificity was 84.5%.

[0100] Example 4 Construction of a multimodal pan-cancer early prediction model

[0101] Using cfDNA samples as input, the first prediction model, the second prediction model, and the third prediction model were used to obtain the positive prediction values ​​of the seven cancer populations in Example 1. All positive prediction values ​​were used as features, and a multimodal two-class discrimination model, i.e., a multimodal pan-cancer early prediction model, was constructed using a logistic regression algorithm. The test set samples were classified and predicted, and the model prediction performance was evaluated. The evaluation results showed that the multimodal model had good cancer and non-cancer discrimination capabilities, and the ROC curve of the test set was as follows: Figure 3 As shown, the AUC in the test set was 0.98, the sensitivity was 89.0%, and the specificity was 98.3%.

[0102] Combining Example 1-Example 4 and Figure 3 It can be seen that after the three basic models are integrated to obtain the multimodal pan-cancer early prediction model, its prediction sensitivity for cancer samples is improved compared with the basic models, while the prediction specificity for healthy samples is also maintained at the highest level compared with the basic models. Therefore, the multimodal pan-cancer early prediction model of the present application can be trained with small sample data and accurately distinguish plasma samples from 7 types of cancer populations and healthy populations.

[0103] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.

[0104] In this application, words such as "including", "comprising", "having", etc. are open words, meaning "including but not limited to", and can be used interchangeably therewith. The words "or" and "and" used herein mean the words "and / or" and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein means the phrase "such as but not limited to", and can be used interchangeably therewith.

[0105] It should also be noted that in the method, system and device of the present application, each step or each module can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present application.

[0106] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A multimodal pan-cancer early prediction method, comprising: Collecting multiple cfDNA samples from i types of cancer populations and healthy populations, respectively, and extracting methylation data of the multiple cfDNA samples; Based on the methylation data of the multiple cfDNA samples, CpG sites are merged to obtain multiple methylation intervals, and differential methylation intervals between each cancer population and a healthy population in i cancer populations are respectively extracted from the multiple methylation intervals, and differential methylation intervals that appear in no less than m cancer populations among the obtained multiple differential methylation intervals are used as candidate marker intervals, and the obtained multiple candidate marker intervals are screened; The first prediction model is trained using the screened multiple candidate marker intervals, the second prediction model is trained by extracting chromosome fragmentation features based on the reference genome, and the third prediction model is trained by extracting chromosome methylation entropy features; A multimodal pan-cancer early prediction model is trained using the cancer population prediction values ​​of the first prediction model, the second prediction model, and the third prediction model for pan-cancer early detection.

2. The multimodal pan-cancer early prediction method according to claim 1, wherein: The extracting methylation data of the multiple cfDNA samples respectively comprises: Extracting a plurality of CpG sites of the plurality of cfDNA samples and a methylation value of each CpG site in the plurality of CpG sites; The merging of CpG sites based on the methylation data of the multiple cfDNA samples comprises: Calculate the difference in methylation value of each CpG site between each cancer population and the healthy population in the i-type cancer population; In response to the difference being not zero, corresponding CpG sites are merged.

3. The multimodal pan-cancer early prediction method according to claim 1, wherein: The screening of the obtained multiple candidate marker intervals comprises: An importance value of each candidate marker interval among the multiple candidate marker intervals is calculated, and in response to the importance value being not greater than a first threshold, the corresponding candidate marker interval is eliminated, and in response to the importance value being greater than the first threshold, the corresponding candidate marker interval is retained.

4. The multimodal pan-cancer early prediction method according to claim 1, wherein: The second prediction model is trained based on the fragmentation features of the chromosome extracted from the reference genome, comprising: The reference genome autosomes are tiled and evenly divided into multiple first intervals; Recording the number of short-length segments, medium-length segments, and long-length segments of each first interval in the plurality of first intervals; sequentially merging the plurality of first intervals to obtain a plurality of second intervals, calculating the coverage of each of the plurality of second intervals, and performing dimensionality reduction on the obtained plurality of coverages; and The second prediction model is constructed using the multiple coverages after dimensionality reduction.

5. The multimodal pan-cancer early prediction method according to claim 4, wherein: The coverage is the short-length fragments, medium-length fragments and long-length fragments contained in each second interval of the multiple second intervals; The second prediction model is constructed by using elastic network regression through the multiple coverages after dimensionality reduction.

6. The multimodal pan-cancer early prediction method according to claim 1, wherein: The extracting of the methylation entropy feature of the chromosome to train the third prediction model comprises: Extracting multiple inserted fragments from the methylation data of the multiple cfDNA samples to obtain methylation pattern entropy values ​​of the multiple inserted fragments; Calculate the methylation pattern entropy value of each chromosome separately; and The third prediction model was constructed using logistic regression based on the methylation pattern entropy values ​​of all chromosomes.

7. The multimodal pan-cancer early prediction method according to claim 6, wherein: The extracting of multiple inserted fragments from the methylation data of the multiple cfDNA samples comprises: The reads extracted from the cfDNA sample in CpG mode, read from left to right and read from right to left, can both be compared with the insert fragments on the reference genome; The formula for obtaining the methylation pattern entropy values ​​of the multiple inserted fragments is as follows: Where BiEn(s) is the methylation pattern entropy value of the inserted fragment, n represents the number of all CpG sites in the inserted fragment, k is a set of values ​​from 0 to n-2, and p is the probability value of k under a given value.

8. The multimodal pan-cancer early prediction method according to claim 6, wherein: The step of calculating the methylation pattern entropy value of each chromosome comprises: The mean of the methylation pattern entropy values ​​of all inserted fragments on each chromosome was calculated as the methylation pattern entropy value of each chromosome.

9. A multimodal pan-cancer early prediction device, comprising: A data collection unit, collecting a plurality of cfDNA samples from i types of cancer populations and healthy populations, respectively, and extracting methylation data of the plurality of cfDNA samples; a marker screening unit, merging CpG sites based on the methylation data of the multiple cfDNA samples to obtain multiple methylation intervals, extracting differential methylation intervals between each cancer population and a healthy population in i types of cancer populations from the multiple methylation intervals, taking differential methylation intervals that appear in no less than m types of cancer populations among the multiple obtained differential methylation intervals as candidate marker intervals, and screening the multiple obtained candidate marker intervals; A model building unit is used to train a first prediction model using the screened multiple candidate marker intervals, to train a second prediction model based on the fragmentation characteristics of the chromosome extracted from the reference genome, and to train a third prediction model by extracting the methylation entropy characteristics of the chromosome; The model integration unit trains a multimodal pan-cancer early prediction model through the cancer population prediction values ​​of the first prediction model, the second prediction model, and the third prediction model for pan-cancer early detection.

10. An electronic device comprising: processor; as well as A memory, wherein computer program instructions are stored in the memory, and when the computer program instructions are executed by the processor, the processor executes the multimodal pan-cancer early prediction method according to any one of claims 1 to 8.

Images