Method and device for detecting tumor molecular subtypes based on low-depth whole genome sequencing and application

Abstract

The invention provides a method and device for detecting tumor molecular typing based on low-depth whole genome sequencing and application. The method comprises the following steps: independently dividing low-depth whole genome sequencing data into an unfixed bin group and a fixed bin group; calculating the ratio of the average copy number to the mutation region of each group by using the copy numbers of the unfixed bin group and the fixed bin group; establishing a scoring model according to the ratio of the average copy number to the mutation region, and judging the copy number variation type of each structure level of the chromosome so as to obtain a corresponding molecular subtype; wherein the unfixed bin group comprises a first target area belonging to an exon and an anti-target area except the exon; the fixed bin group comprises a second target area with the length of 1M and a third target area with the length of 10K which are obtained by continuous division. The copy number variation types of the chromosomes at different hierarchical structure positions are obtained, so that relatively accurate molecular typing detection of each hierarchical size of the tumor chromosomes is realized.

Description

Technical Field

[0001] The present invention relates to the field of gene detection technology, and in particular to a method, device and application for detecting tumor molecular typing based on low-depth whole genome sequencing. Background Art

[0002] Brain tumors include gliomas, meningiomas, pituitary tumors, ependymomas, medulloblastomas, central nervous system lymphomas, hereditary brain tumors, etc. With the release of the latest version of the WHO CNS5 guidelines in 2021, molecular diagnosis has become increasingly important in the field of brain tumors. The detection can obtain the subtype of the tumor to determine the patient's prognosis and guide the patient's clinical treatment and medication plan. The complex molecular typing requirements of various types of cancer include methylation, gene mutations, and mutations in different sizes of chromosome regions. Common detection schemes for second-generation sequencing (targeted sequencing or whole exome sequencing) have the defect of insufficient detection range.

[0003] Currently available detection methods are limited to copy number detection of partial levels of chromosome regions, or limited to the molecular typing of a specific cancer type. To complete copy number detection of multiple cancer types and full-region chromosomes, additional high time and financial costs are required. In addition, since most clinical samples are FFPE samples, it is difficult to extract nucleic acids from samples, and some experimental-based detection schemes may be difficult to implement; and a variety of combination schemes require too large a sample size, which objectively limits the feasibility of detection. Using a large number of samples for machine learning modeling to identify molecular typing, through the process of bioinformatics and statistical modeling, will be affected by the sample data situation and parameter adjustment, has a certain degree of subjectivity, and requires sufficient prior data accumulation, and it is difficult to expand to brain tumors.

[0004] Therefore, how to provide a method for detecting multi-level chromosome copy number variations in different tumor subtypes is very important for realizing tumor molecular typing detection. Summary of the invention

[0005] The main purpose of the present invention is to provide a method, device and application for detecting tumor molecular typing based on low-depth whole genome sequencing, so as to solve the problem of poor accuracy in detecting copy number variation of multi-level chromosome structure in the prior art.

[0006] In order to achieve the above-mentioned purpose, according to the first aspect of the present invention, a method for detecting tumor molecular typing based on low-depth whole genome sequencing is provided, the method comprising: S1, independently dividing the low-depth whole genome sequencing data into a non-fixed bin group and a fixed bin group; S2, calculating the average copy number and mutation region ratio of each group using the copy number of the non-fixed bin group and the fixed bin group; S3, establishing a scoring model according to the average copy number and mutation region ratio, judging the copy number variation type of each structural level of the chromosome, and then deriving the corresponding molecular typing; wherein, the non-fixed bin group includes a first target region belonging to the exon and an anti-target region outside the exon; the fixed bin group includes a second target region with a length of 1M and a third target region with a length of 10K obtained by continuous division; each structural level of the chromosome includes: the whole chromosome, chromosome arm, chromosome region and chromosome band.

[0007] Furthermore, before the division, the inaccessible part of the second-generation sequencing in the low-depth whole genome sequencing data is removed; preferably, the low-depth whole genome sequencing data is a BAM file obtained by comparing with the human reference genome after quality control.

[0008] Furthermore, the copy numbers of the non-fixed bin group and the fixed bin group are calculated respectively using CNVkit software; preferably, before calculating the average copy number and mutation region ratio of each group, the method also includes: annotating the position of the region in the chromosome in the non-fixed bin group and the fixed bin group whose copy number is calculated using CNVkit software; judging the copy number variation type of each region in the non-fixed bin group and the fixed bin group according to the copy number variation threshold; the copy number variation types include: loss, loss risk Loss_risk, amplification Gain and amplification risk Gain_risk.

[0009] Furthermore, calculating the average copy number and mutation region ratio of each group includes: calculating the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of each region whose copy number is calculated by CNVkit software in the non-fixed bin group at the chromosome whole and chromosome arm levels; calculating the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of the region whose copy number is calculated by CNVkit software in the second target region and the third target region in the fixed bin group respectively; average copy number = (the sum of the length of the region whose copy number is calculated by CNVkit software in the object * the corresponding region copy number) / the sum of the lengths of all regions in the object for which the copy numbers are calculated; the first mutation region ratio=the sum of the lengths of the Loss regions in the object / the sum of the lengths of all regions in the object for which the copy numbers are calculated; the second mutation region ratio=the sum of the lengths of the Loss_risk regions in the object / the sum of the lengths of all regions in the object for which the copy numbers are calculated; the third mutation region ratio=the sum of the lengths of the Gain regions in the object / the sum of the lengths of all regions in the object for which the copy numbers are calculated; the fourth mutation region ratio=the sum of the lengths of the Gain_risk regions in the object / the sum of the lengths of all regions in the object for which the copy numbers are calculated; wherein the object is: the entire chromosome, a chromosome arm, the second target region or the third target region.

[0010] Further, S3 includes: using the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio at the chromosome level to establish a first scoring model to determine the copy number variation type of the chromosome as a whole; using the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio at the chromosome arm level to establish a second scoring model to determine the copy number variation type of the chromosome arm; based on the copy number variation type of the chromosome as a whole and the copy number variation type of the chromosome arm, using the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of the second target region, establish a third scoring model to further determine the copy number variation type of the chromosome region; based on the copy number variation type of the chromosome as a whole and the copy number variation type of the chromosome arm, using the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of the third target region, establish a fourth scoring model to further determine the copy number variation type of the chromosome band.

[0011] Further, the copy number variation type of the whole chromosome and the chromosome arm is judged according to the following judgment criteria: when the average copy number of the judgment object is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the judgment object is ≥ 85%, the copy number variation type of the judgment object is Loss; when the average copy number of the judgment object is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the judgment object and the second mutation region ratio of the judgment object is ≥ 85%, the copy number variation type of the judgment object is Loss_risk; when the average copy number of the judgment object is ≥ the baseline average copy number * 1.2 and the third mutation region ratio of the judgment object is ≥ 85%, the copy number variation type of the judgment object is Gain; when the average copy number of the judgment object is ≥ the baseline average copy number * 1.1 and the third mutation region ratio of the judgment object and the fourth mutation region ratio of the judgment object are ≥ 85%, the copy number variation type of the judgment object is Gain_risk; the judgment object is the whole chromosome or the chromosome arm.

[0012] Furthermore, the copy number variation type of the chromosome region is determined according to the following judgment criteria: when the copy number variation type of the chromosome arm in the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the second target region is ≥ 85%, then the copy number variation type of the chromosome region is deletion Loss; when the copy number variation type of the chromosome arm in the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the second target region and the second mutation region ratio of the second target region is ≥ 85%, then the copy number variation type of the chromosome region is It is the deletion risk Loss_risk; when the copy number variation type of the chromosome arm in the chromosome region is not deletion Loss, when the average copy number of the second target region is ≥ the baseline average copy number*1.2 and the proportion of the third mutation region in the second target region is ≥85%, then the copy number variation type of the chromosome region is amplification Gain; when the copy number variation type of the chromosome arm in the chromosome region is not deletion Loss, when the average copy number of the second target region is ≥ the baseline average copy number*1.1 and the proportion of the third mutation region in the second target region and the proportion of the fourth mutation region in the second target region are ≥85%, then the copy number variation type of the chromosome region is amplification risk Gain_risk.

[0013] Further, the copy number variation type of the chromosome band is determined according to the following judgment criteria: when the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the third target region is ≥ 85%, then the copy number variation type of the chromosome band is deletion Loss; when the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the third target region and the second mutation region ratio of the third target region is ≥ 85%, then the copy number variation type of the chromosome band is It is the deletion risk Loss_risk; when the copy number variation type of the chromosome arm of the chromosome band is not deletion Loss, when the average copy number of the third target region is ≥ the baseline average copy number * 1.2 and the proportion of the third mutation region of the third target region is ≥ 85%, then the copy number variation type of the chromosome band is amplification Gain; when the copy number variation type of the chromosome arm of the chromosome band is not deletion Loss, when the average copy number of the third target region is ≥ the baseline average copy number * 1.1 and the proportion of the third mutation region of the third target region and the proportion of the fourth mutation region of the third target region are ≥ 85%, then the copy number variation type of the chromosome band is amplification risk Gain_risk.

[0014] In order to achieve the above-mentioned purpose, according to the second aspect of the present invention, there is provided a device for detecting molecular typing of tumors based on low-depth whole genome sequencing, the device comprising: a bin division module, configured to independently divide low-depth whole genome sequencing data into non-fixed bin groups and fixed bin groups; a calculation module, configured to calculate the average copy number and mutation region ratio of each group using the copy number of the non-fixed bin group and the fixed bin group; a judgment module, configured to establish a scoring model based on the average copy number and mutation region ratio, judge the copy number variation type of each structural level of the chromosome, and then obtain the corresponding molecular typing; wherein, the non-fixed bin group includes a first target region belonging to the exon and an anti-target region outside the exon; the fixed bin group includes a second target region with a length of 1M and a third target region with a length of 10K obtained by continuous division; each structural level of the chromosome includes: the whole chromosome, chromosome arm, chromosome region and chromosome band.

[0015] Furthermore, in the bin partitioning module, before partitioning, the inaccessible part of the second-generation sequencing in the low-depth whole genome sequencing data is removed; preferably, the low-depth whole genome sequencing data is a BAM file obtained by comparing with the human reference genome after quality control.

[0016] Furthermore, the calculation module includes: a CNVkit software calculation unit, which is configured to use the CNVkit software to respectively calculate the copy numbers of the non-fixed bin group and the fixed bin group; preferably, before calculating the average copy number and mutation region ratio of each group, the calculation module also includes: an annotation unit, which is configured to annotate the position of the region in the chromosome in the non-fixed bin group and the fixed bin group whose copy number is calculated using the CNVkit software; a threshold judgment unit, which is configured to judge the copy number variation type of each region in the non-fixed bin group and the fixed bin group based on the threshold of the copy number variation; the copy number variation types include: loss, loss risk Loss_risk, amplification Gain and amplification risk Gain_risk.

[0017] Furthermore, the calculation module also includes: a first calculation unit, which is configured to calculate the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of each region whose copy number is calculated by CNVkit software in the non-fixed bin group at the chromosome whole level and the chromosome arm level respectively; a second calculation unit, which is configured to calculate the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of the region whose copy number is calculated by CNVkit software in the second target region and the third target region in the fixed bin group respectively; average copy number=(length of the region whose copy number is calculated by CNVkit software in the object*corresponding region copy number) The first mutation region ratio=the sum of the lengths of the Loss regions in the object / the sum of the lengths of the regions in the object for which the copy numbers are calculated; the second mutation region ratio=the sum of the lengths of the Loss_risk regions in the object / the sum of the lengths of the regions in the object for which the copy numbers are calculated; the third mutation region ratio=the sum of the lengths of the Gain regions in the object / the sum of the lengths of the regions in the object for which the copy numbers are calculated; the fourth mutation region ratio=the sum of the lengths of the Gain_risk regions in the object / the sum of the lengths of the regions in the object for which the copy numbers are calculated; wherein the object is: the entire chromosome, the chromosome arm, the second target region or the third target region.

[0018] Furthermore, the judgment module includes: a chromosome overall copy number variation type judgment unit, which is configured to use the average copy number at the chromosome overall level, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio to establish a first scoring model to judge the chromosome overall copy number variation type; a chromosome arm copy number variation type judgment unit, which is configured to use the average copy number at the chromosome arm level, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio to establish a second scoring model to judge the chromosome arm copy number variation type; a chromosome region copy number variation type judgment unit, which is configured to use the average copy number at the chromosome arm level, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio to establish a second scoring model to judge the chromosome arm copy number variation type The copy number variation type of the entire chromosome and the copy number variation type of the chromosome arm are determined by using the average copy number of the second target region, the proportion of the first mutation region, the proportion of the second mutation region, the proportion of the third mutation region and the proportion of the fourth mutation region to establish a third scoring model, and then determine the copy number variation type of the chromosome region; the chromosome band copy number variation type judgment unit is configured to be based on the copy number variation type of the entire chromosome and the copy number variation type of the chromosome arm, and the average copy number of the third target region, the proportion of the first mutation region, the proportion of the second mutation region, the proportion of the third mutation region and the proportion of the fourth mutation region to establish a fourth scoring model, and then determine the copy number variation type of the chromosome band.

[0019] Further, the judgment criteria of the copy number variation type in the chromosome whole copy number variation type judgment unit and the chromosome arm copy number variation type judgment unit include: when the average copy number of the judgment object is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the judgment object is ≥ 85%, the copy number variation type of the judgment object is loss; when the average copy number of the judgment object is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the judgment object and the second mutation region ratio of the judgment object is ≥ 85%, the copy number variation type of the judgment object is loss risk Loss_risk; when the average copy number of the judgment object is ≥ the baseline average copy number * 1.2 and the third mutation region ratio of the judgment object is ≥ 85%, the copy number variation type of the judgment object is amplification Gain; when the average copy number of the judgment object is ≥ the baseline average copy number * 1.1 and the third mutation region ratio of the judgment object and the fourth mutation region ratio of the judgment object are ≥ 85%, the copy number variation type of the judgment object is amplification risk Gain_risk; the judgment object is the whole chromosome or the chromosome arm.

[0020] Furthermore, the judgment criteria of the copy number variation type of the chromosome region in the chromosome region copy number variation type judgment unit include: when the copy number variation type of the chromosome arm of the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the second target region is ≥ 85%, then the copy number variation type of the chromosome region is deletion Loss; when the copy number variation type of the chromosome arm of the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the second target region and the second mutation region ratio of the second target region is ≥ 85%, then the copy number variation type of the chromosome region is deletion Loss. The copy number variation type is deletion risk Loss_risk; when the copy number variation type of the chromosome arm in the chromosome region is not deletion Loss, when the average copy number of the second target region is ≥ the baseline average copy number*1.2 and the proportion of the third mutation region in the second target region is ≥85%, then the copy number variation type of the chromosome region is amplification Gain; when the copy number variation type of the chromosome arm in the chromosome region is not deletion Loss, when the average copy number of the second target region is ≥ the baseline average copy number*1.1 and the proportion of the third mutation region in the second target region and the proportion of the fourth mutation region in the second target region are ≥85%, then the copy number variation type of the chromosome region is amplification risk Gain_risk.

[0021] Furthermore, the judgment criteria of the copy number variation type of the chromosome band in the chromosome band copy number variation type judgment unit include: when the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the third target region is ≥ 85%, then the copy number variation type of the chromosome band is deletion Loss; when the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the third target region and the second mutation region ratio of the third target region is ≥ 85%, then the copy number variation type of the chromosome band is deletion Loss. The copy number variation type is deletion risk Loss_risk; when the copy number variation type of the chromosome arm of the chromosome band is not deletion Loss, when the average copy number of the third target region ≥ baseline average copy number * 1.2 and the proportion of the third mutation region of the third target region ≥ 85%, then the copy number variation type of the chromosome band is amplification Gain; when the copy number variation type of the chromosome arm of the chromosome band is not deletion Loss, when the average copy number of the third target region ≥ baseline average copy number * 1.1 and the proportion of the third mutation region of the third target region and the proportion of the fourth mutation region of the third target region ≥ 85%, then the copy number variation type of the chromosome band is amplification risk Gain_risk.

[0022] In order to achieve the above-mentioned purpose, according to the third aspect of the present invention, a computer-readable storage medium is provided, which includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the above-mentioned method for detecting tumor molecular typing based on low-depth whole genome sequencing.

[0023] In order to achieve the above-mentioned purpose, according to the fourth aspect of the present invention, a processor is provided, which is used to run a program, wherein the program runs the above-mentioned method for detecting tumor molecular typing based on low-depth whole genome sequencing.

[0024] By applying the technical solution of the present invention, the sWGS sequencing data is divided into different bin combinations according to different standards for subsequent copy number analysis, including being divided into non-fixed bin groups based on whether they are exons and being continuously divided into fixed bin groups based on 1M and 10K lengths. The copy number data of different bin groups are judged to determine the copy number variation types of chromosomes at different hierarchical structural positions, and then the corresponding molecular typing is obtained, thereby achieving more accurate molecular typing detection of tumor chromosomes at various levels, which is further suitable for the detection of molecular typing of different cancer subspecies. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0026] Figure 1 A flow chart of the detection method of the present application is shown;

[0027] Figure 2 The figure shows the detection result of the whole chromosomes of 12 samples using sWGS in Example 1;

[0028] Figure 3 The result diagram of sWGS detection of sample 1 in Example 1 is shown;

[0029] Figure 4 The result diagram of Panel+Marker SNP detection of sample 1 in Example 1 is shown, wherein Chromosome: chromosome; Allele Frequency: genotype frequency; Chromosomal Location: chromosome location; type: type; abnormal: abnormal; normal: normal;

[0030] Figure 5 The result diagram of Panel+Marker SNP detection of sample 1 in Example 1 is shown, wherein CopyNumber (log2 ratio): copy number (log2 ratio); gain: amplification; deletion: deletion; cnvType: cnv type;

[0031] Figure 6 The result of gene chip detection of sample 1 in Example 1 is shown, wherein 1-22 and X correspond to the chromosome name, and the copy number of the large fragment of the whole chromosome is displayed on the left side of each chromosome, red represents amplification, blue represents deletion, and white represents neutrality. The darker the color, the more serious the copy number variation. The gene chip detection will set a threshold to determine whether amplification occurs. In the figure, A represents chr7 amplification (dark red), B represents chr12 amplification (light red), and C represents chr18 amplification (light red);

[0032] Figure 7 The result diagram of sWGS detection of sample 11 in Example 1 is shown;

[0033] Figure 8The result diagram of gene chip detection of sample 11 in Example 1 is shown; wherein, the mark (1-22, X and Y) in the white box in the lower left corner of the nuclear view corresponds to the chromosome name, and the copy number variation occurring in the chromosome segment is displayed on the right side of each chromosome, red represents deletion, blue represents amplification, and no annotation represents neutrality. In the figure, A represents 1p neutrality, B represents chr7 neutrality, C represents chr10 neutrality, D represents a small fragment deletion on chr19, E represents a small fragment amplification on chr19, and F represents 19q neutrality. DETAILED DESCRIPTION

[0034] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0035] As mentioned in the background technology, the existing methods for detecting tumor molecular typing are limited to a certain specific cancer or partial hierarchical position of chromosomes, and it is impossible to detect more accurate molecular typing results of chromosomes in the whole region, and the cost is relatively high. The existing method exists by traditional fluorescence in situ hybridization (FISH) to verify the copy number variation of the whole chromosome and the arm region; by combining a gene chip with a certain bioinformatics method, the copy number variation of the chromosome region and the band region is detected; by professionally customized targeted capture gene package (Panel) combined with labeled single nucleotide polymorphism sites (SNP) to detect the molecular typing of specific cancers. Therefore, in order to obtain accurate molecular typing of chromosomes in the whole region and expand the scope of application of cancer, the present application intends to provide a detection method for molecular typing based on low-depth whole genome sequencing results.

[0036] In the first typical embodiment of the present application, a method for detecting tumor molecular typing based on low-depth whole genome sequencing is provided, the method comprising: S1, independently dividing low-depth whole genome sequencing data into unfixed bin groups and fixed bin groups; S2, using the copy numbers of unfixed bin groups and fixed bin groups to calculate the average copy number and mutation region ratio of each group; S3, establishing a scoring model based on the average copy number and mutation region ratio, judging the copy number variation type of each structural level of the chromosome, and then deriving the corresponding molecular typing; wherein the unfixed bin group includes the first target region belonging to the exon and the anti-target region outside the exon; the fixed bin group includes the second target region with a length of 1M and the third target region with a length of 10K obtained by continuous division; each structural level of the chromosome includes: the whole chromosome, chromosome arm, chromosome region and chromosome band. wherein the sequencing depth of the tumor sample is 1-5×.

[0037] In this application, low-depth whole-genome sequencing results are divided into bins according to different standards to obtain a non-fixed bin group divided by whether it is an exon or not and a fixed bin group divided continuously by lengths of 1M and 10K. The two different bin groupings are used to calculate the copy number, and the average copy number and the proportion of mutation regions are obtained based on the copy number calculation, which serve as the basis for judging the copy number variation type of each chromosome level (whole, arm, region, band), and a scoring model is established to determine the copy number variation type of the entire chromosome, and further obtain accurate tumor molecular typing results.

[0038] Among them, copy number variation at different chromosome levels is one of the criteria for molecular typing of brain tumors. Based on the WHO guidelines, 353 brain tumor-related variations were screened (covering molecular markers related to diagnosis, grading, and prognosis of different cancer types), including chromosome level (such as chr7_AMP, chr10_DEL), broken arm and long arm level (such as 1p_DEL, 19q_DEL), gene or above level (such as MET amplification, CDK4 amplification, MYC amplification, etc.). The copy number variation at different chromosome levels can be combined with key gene mutations (Snp+Indel), methylation, etc. as diagnostic criteria for molecular typing of brain tumors, and each standard is independent and parallel to each other.

[0039] In order to further improve the efficiency of bin division and obtain more valid data for data analysis, the low-depth whole genome sequencing data needs to be cleaned and filtered. In a preferred embodiment, before binning, the inaccessible part of the second-generation sequencing in the low-depth whole genome sequencing data is removed; preferably, the low-depth whole genome sequencing data is a BAM file obtained by comparing with the human reference genome after quality control.

[0040] Among them, due to the short read length of the second-generation sequencing, the specificity of many non-uniquely matched regions is not very good. The variations in these regions, whether point mutations or CNV / SV, are not very reliable. Even in the uniquely matched regions, there will be preferences in the GC abnormal regions during the library construction phase. The Github website summarizes the following characteristics of the above-mentioned second-generation sequencing inaccessible parts: 1. Repeated sequence regions; 2. GC abnormal (high GC and low GC content) regions; 3. Telomere and centrosome regions; 4. Pseudogenes / highly homologous genes; 5. Low matching score regions.

[0041] There is a column on the Encode data website that can evaluate the uniqueness of the alignment of sequences of various lengths on the genome; there is also an open source project about this on github (https: / / github.com / Boyle-Lab / Blacklist). The manual of the open source software CNVkit provides access files for accessible regions with high sequencing quality, and recommends using this software in preliminary analysis to remove detection regions with low quality in second-generation sequencing to obtain more accurate analysis results (https: / / cnvkit.readthedocs.io / en / stable / pipeline.html). The above database can be used to remove the inaccessible parts of the second-generation sequencing.

[0042] Any method capable of calculating copy number data is applicable to the present application. In a preferred embodiment, the copy numbers of the non-fixed bin group and the fixed bin group are calculated using CNVkit software.

[0043] Among them, this application completed the open source software testing (CNVkit, Control-Free, Sequeza) during the testing phase, among which CNVkit: 17 direct citations in Pubmed within 5 years, 700 total citations; Control-Free: 63 direct citations in Pubmed within 5 years, 471 total citations; Sequeza: 12 direct citations in Pubmed within 5 years, 320 total citations. CNVkit can support low-depth WGS data and is more accurate in detecting large fragment copy numbers at the chromosome region level; Control-Free has low depth and single sample, and the accuracy of the results is poor in special area detection when calculating allele frequencies; Sequeza has low depth and single sample, and it is difficult to obtain valid data when calculating allele frequencies, and the deviation is large. Finally, CNVkit was selected as the solution that meets our detection range expectations.

[0044] By using the existing threshold standard for judging copy number variation, the copy number variation type of the region in each bin group where the copy number is detected is judged, so as to facilitate the subsequent judgment of the copy number variation type at each level of the whole chromosome. In a preferred embodiment, before calculating the average copy number and mutation region ratio of each group, the method also includes: annotating the position of the region in the chromosome where the copy number is calculated by CNVkit software in the unfixed bin group and the fixed bin group; judging the copy number variation type of each region in the unfixed bin group and the fixed bin group according to the threshold of copy number variation; the copy number variation types include: loss, loss risk Loss_risk, amplification Gain and amplification risk Gain_risk.

[0045] For chromosomes at different levels, the copy numbers of different bin groups are used for calculation to obtain relevant data that can determine the types of chromosome copy number variation at different levels. In a preferred embodiment, calculating the average copy number and mutation region ratio of each group includes: respectively calculating the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of each region whose copy number is calculated by CNVkit software in the non-fixed bin group at the chromosome whole and chromosome arm levels; respectively calculating the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of the region whose copy number is calculated by CNVkit software in the second target region and the third target region in the fixed bin group;

[0046] Average copy number = (the sum of the length of the region in the object where the copy number is calculated by CNVkit software * the corresponding region copy number) / the sum of the lengths of all regions in the object where the copy number is calculated;

[0047] The first mutation region ratio = the sum of the lengths of the Loss regions in the object / the sum of the lengths of all regions with calculated copy numbers in the object;

[0048] The second mutation region ratio = the sum of the lengths of the Loss_risk regions in the object / the sum of the lengths of all regions with calculated copy numbers in the object;

[0049] The third mutation region ratio = the sum of the Gain region lengths in the object / the sum of the region lengths of all calculated copy numbers in the object;

[0050] Fourth, mutation region ratio = the sum of the lengths of the Gain_risk regions in the object / the sum of the lengths of all regions with calculated copy numbers in the object;

[0051] The object is: the entire chromosome, a chromosome arm, a second target region or a third target region.

[0052] The copy number-related data of different chromosome levels (whole, arm) or different target regions (second and third target regions) are used to judge the copy number variation type, and a related scoring model is established to obtain the copy number variation type of the chromosome in the entire region, and further obtain the tumor molecular typing. In a preferred embodiment, S3 includes: using the average copy number at the chromosome overall level, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio to establish a first scoring model to judge the copy number variation type of the chromosome as a whole; using the average copy number at the chromosome arm level, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio to establish a second scoring model to judge the copy number variation type of the chromosome arm.

[0053] Based on the copy number variation type of the entire chromosome and the copy number variation type of the chromosome arm, a third scoring model is established using the average copy number of the second target region, the proportion of the first mutation region, the proportion of the second mutation region, the proportion of the third mutation region and the proportion of the fourth mutation region, and then the copy number variation type of the chromosome region is determined; based on the copy number variation type of the entire chromosome and the copy number variation type of the chromosome arm, a fourth scoring model is established using the average copy number of the third target region, the proportion of the first mutation region, the proportion of the second mutation region, the proportion of the third mutation region and the proportion of the fourth mutation region, and then the copy number variation type of the chromosome band is determined.

[0054] In a preferred embodiment, the copy number variation type of the whole chromosome and the chromosome arm is determined according to the following judgment criteria: when the average copy number of the object to be judged is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the object to be judged is ≥ 85%, the copy number variation type of the object to be judged is loss; when the average copy number of the object to be judged is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the object to be judged and the second mutation region ratio of the object to be judged is ≥ 85%, the copy number variation type of the object to be judged is loss risk_risk; when the average copy number of the object to be judged is ≥ the baseline average copy number * 1.2 and the third mutation region ratio of the object to be judged is ≥ 85%, the copy number variation type of the object to be judged is gain; when the average copy number of the object to be judged is ≥ the baseline average copy number * 1.1 and the third mutation region ratio of the object to be judged is ≥ 85% of the fourth mutation region ratio of the object to be judged, the copy number variation type of the object to be judged is gain risk_risk; the object to be judged is the whole chromosome or the chromosome arm.

[0055] The baseline average copy number is calculated based on the baseline established according to the official manual of the CNVkit software based on the WGS data of normal people. 0.8 times is obtained by referring to the positive and negative result sets of the test samples. The ratio of 85% is determined because the proportion of the largest long arm of all chromosomes in the human reference genome does not exceed 85% of the overall chromosome, and is further used in the judgment of chromosomes at all levels.

[0056] In order to further accurately judge the copy number variation type of the chromosome region, the strong positive results of the chromosome arms obtained above are used for filtering, that is, if the copy number variation type of the chromosome arm is Gain, then the copy number variation type of the chromosome region containing the chromosome arm is unlikely to be a Loss.

[0057] In a preferred embodiment, the copy number variation type of the chromosome region is determined according to the following judgment criteria: when the copy number variation type of the chromosome arm of the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the second target region is ≥ 85%, then the copy number variation type of the chromosome region is deletion Loss; when the copy number variation type of the chromosome arm of the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the second target region and the second mutation region ratio of the second target region is ≥ 85%, then the copy number variation type of the chromosome region is deletion Loss. The aberrant type is deletion risk Loss_risk; when the copy number variation type of the chromosome arm in the chromosome region is not deletion Loss, when the average copy number of the second target region is ≥ the baseline average copy number * 1.2 and the proportion of the third mutation region in the second target region is ≥ 85%, then the copy number variation type of the chromosome region is amplification Gain; when the copy number variation type of the chromosome arm in the chromosome region is not deletion Loss, when the average copy number of the second target region is ≥ the baseline average copy number * 1.1 and the proportion of the third mutation region in the second target region and the proportion of the fourth mutation region in the second target region are ≥ 85%, then the copy number variation type of the chromosome region is amplification risk Gain_risk.

[0058] In order to further accurately judge the type of chromosome band copy number variation, the strong positive results of the chromosome arms obtained above are used for filtering, that is, if the copy number variation type of the chromosome arm is Gain, then the copy number variation type of the chromosome band containing the chromosome arm is unlikely to be a Loss.

[0059] In a preferred embodiment, the copy number variation type of the chromosome band is determined according to the following judgment criteria: when the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the third target region is ≥ 85%, then the copy number variation type of the chromosome band is deletion Loss; when the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the third target region and the second mutation region ratio of the third target region is ≥ 85%, then the copy number variation type of the chromosome band is deletion Loss. The aberrant type is deletion risk Loss_risk; when the copy number variation type of the chromosome arm of the chromosome band is not deletion Loss, when the average copy number of the third target region is ≥ the baseline average copy number * 1.2 and the proportion of the third mutation region of the third target region is ≥ 85%, then the copy number variation type of the chromosome band is amplification Gain; when the copy number variation type of the chromosome arm of the chromosome band is not deletion Loss, when the average copy number of the third target region is ≥ the baseline average copy number * 1.1 and the proportion of the third mutation region of the third target region and the proportion of the fourth mutation region of the third target region are ≥ 85%, then the copy number variation type of the chromosome band is amplification risk Gain_risk.

[0060] In a second typical embodiment of the present application, a device for detecting tumor molecular typing based on low-depth whole genome sequencing is provided, and the device includes: a bin division module, which is configured to independently divide the low-depth whole genome sequencing data into a non-fixed bin group and a fixed bin group; a calculation module, which is configured to calculate the average copy number and mutation region ratio of each group using the copy number of the non-fixed bin group and the fixed bin group; a judgment module, which is configured to establish a scoring model based on the average copy number and mutation region ratio, judge the copy number variation type of each structural level of the chromosome, and then derive the corresponding molecular typing; wherein the non-fixed bin group includes a first target region belonging to the exon and an anti-target region outside the exon; the fixed bin group includes a second target region with a length of 1M and a third target region with a length of 10K obtained by continuous division; each structural level of the chromosome includes: the whole chromosome, chromosome arm, chromosome region and chromosome band.

[0061] The device divides the low-depth whole genome sequencing results into bins with different standards through the bin division module, and obtains a non-fixed bin group divided by whether it is an exon and a fixed bin group divided continuously by length 1M and 10K. The calculation module uses the two different bin groupings to calculate the copy number, and obtains the average copy number and the proportion of mutation areas based on the copy number calculation. The judgment module uses this as the basis for judging the copy number variation type of each chromosome level (whole, arm, region, band), establishes a scoring model, determines the copy number variation type of the entire chromosome, and further obtains accurate tumor molecular typing results.

[0062] In order to further improve the efficiency of bin division and obtain more valid data for data analysis, it is necessary to clean and filter the low-depth whole genome sequencing data. In a preferred embodiment, in the bin division module, the second-generation sequencing inaccessible part of the low-depth whole genome sequencing data is removed before division; preferably, the low-depth whole genome sequencing data is a BAM file obtained by comparing with the human reference genome after quality control.

[0063] Any method capable of calculating copy number data is applicable to the present application. In a preferred embodiment, the calculation module includes: a CNVkit software calculation unit, which is configured to use the CNVkit software to calculate the copy numbers of the non-fixed bin group and the fixed bin group respectively.

[0064] By using the existing threshold standard for judging copy number variation, the copy number variation type of the region in each bin group where the copy number is detected is judged, so as to facilitate the subsequent judgment of the copy number variation type at each level of the whole chromosome. In a preferred embodiment, before calculating the average copy number and mutation region ratio of each group, the calculation module also includes: an annotation unit, which is configured to annotate the position of the region in the chromosome in the non-fixed bin group and the fixed bin group whose copy number is calculated by CNVkit software; a threshold judgment unit, which is configured to judge the copy number variation type of each region in the non-fixed bin group and the fixed bin group according to the threshold of copy number variation; the copy number variation types include: loss, loss risk Loss_risk, amplification Gain and amplification risk Gain_risk.

[0065] For chromosomes at different levels, the copy numbers of different bin groups are used for calculation to obtain relevant data that can determine the types of chromosome copy number variation at different levels. In a preferred embodiment, the calculation module also includes: a first calculation unit, which is configured to respectively calculate the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of each region whose copy number is calculated by the CNVkit software in the non-fixed bin group at the chromosome as a whole and chromosome arm levels; a second calculation unit, which is configured to respectively calculate the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of the region whose copy number is calculated by the CNVkit software in the second target region and the third target region in the fixed bin group;

[0066] Wherein, average copy number = (the sum of the length of the region in the object where the copy number is calculated by CNVkit software * the corresponding region copy number) / the sum of the lengths of all regions in the object where the copy number is calculated;

[0067] The first mutation region ratio = the sum of the lengths of the Loss regions in the object / the sum of the lengths of all regions with calculated copy numbers in the object;

[0068] The second mutation region ratio = the sum of the lengths of the Loss_risk regions in the object / the sum of the lengths of all regions with calculated copy numbers in the object;

[0069] The third mutation region ratio = the sum of the Gain region lengths in the object / the sum of the region lengths of all calculated copy numbers in the object;

[0070] Fourth, mutation region ratio = the sum of the lengths of the Gain_risk regions in the object / the sum of the lengths of all regions with calculated copy numbers in the object;

[0071] The object is: the entire chromosome, a chromosome arm, a second target region or a third target region.

[0072] The copy number related data of different chromosome levels (whole, arm) or different target regions (second and third target regions) are used to judge the copy number variation type, and a related scoring model is established to obtain the copy number variation type of the chromosome in the whole region, and further obtain the tumor molecular typing. In a preferred embodiment, the judgment module includes: a chromosome whole copy number variation type judgment unit, which is configured to use the average copy number at the chromosome whole level, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio to establish a first scoring model to judge the copy number variation type of the chromosome as a whole;

[0073] The chromosome arm copy number variation type judgment unit is configured to establish a second scoring model using the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio at the chromosome arm level to judge the copy number variation type of the chromosome arm;

[0074] The chromosome region copy number variation type judgment unit is configured to establish a third scoring model based on the copy number variation type of the entire chromosome and the copy number variation type of the chromosome arm, using the average copy number of the second target region, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio, so as to judge the copy number variation type of the chromosome region;

[0075] The chromosome band copy number variation type judgment unit is configured to establish a fourth scoring model based on the copy number variation type of the chromosome as a whole and the copy number variation type of the chromosome arm, using the average copy number of the third target region, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio, to further judge the copy number variation type of the chromosome band.

[0076] In a preferred embodiment, the judgment criteria of the copy number variation type in the chromosome whole copy number variation type judgment unit and the chromosome arm copy number variation type judgment unit include: when the average copy number of the judgment object is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the judgment object is ≥ 85%, the copy number variation type of the judgment object is loss; when the average copy number of the judgment object is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the judgment object and the second mutation region ratio of the judgment object is ≥ 85%, the copy number variation type of the judgment object is loss risk Loss_risk; when the average copy number of the judgment object is ≥ the baseline average copy number * 1.2 and the third mutation region ratio of the judgment object is ≥ 85%, the copy number variation type of the judgment object is amplification Gain; when the average copy number of the judgment object is ≥ the baseline average copy number * 1.1 and the third mutation region ratio of the judgment object and the fourth mutation region ratio of the judgment object are ≥ 85%, the copy number variation type of the judgment object is amplification risk Gain_risk; the judgment object is the whole chromosome or the chromosome arm.

[0077] In a preferred embodiment, the judgment criteria of the copy number variation type of the chromosome region in the chromosome region copy number variation type judgment unit include: when the copy number variation type of the chromosome arm in the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the second target region is ≥ 85%, then the copy number variation type of the chromosome region is deletion Loss; when the copy number variation type of the chromosome arm in the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the second target region and the second mutation region ratio of the second target region is ≥ 85%, then the chromosome The copy number variation type of the chromosome region is deletion risk Loss_risk; when the copy number variation type of the chromosome arm in the chromosome region is not deletion Loss, when the average copy number of the second target region is ≥ the baseline average copy number * 1.2 and the proportion of the third mutation region in the second target region is ≥ 85%, then the copy number variation type of the chromosome region is amplification Gain; when the copy number variation type of the chromosome arm in the chromosome region is not deletion Loss, when the average copy number of the second target region is ≥ the baseline average copy number * 1.1 and the proportion of the third mutation region in the second target region and the proportion of the fourth mutation region in the second target region are ≥ 85%, then the copy number variation type of the chromosome region is amplification risk Gain_risk.

[0078] In a preferred embodiment, the judgment criteria of the copy number variation type of the chromosome band in the chromosome band copy number variation type judgment unit include: when the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ the baseline average copy number * 0.8 and the first mutation region ratio of the third target region is ≥ 85%, then the copy number variation type of the chromosome band is deletion Loss; when the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ the baseline average copy number * 0.9 and the sum of the first mutation region ratio of the third target region and the second mutation region ratio of the third target region is ≥ 85%, then the chromosome The copy number variation type of the band is deletion risk Loss_risk; when the copy number variation type of the chromosome arm of the chromosome band is not deletion Loss, when the average copy number of the third target region is ≥ baseline average copy number * 1.2 and the proportion of the third mutation region of the third target region is ≥ 85%, then the copy number variation type of the chromosome band is amplification Gain; when the copy number variation type of the chromosome arm of the chromosome band is not deletion Loss, when the average copy number of the third target region is ≥ baseline average copy number * 1.1 and the proportion of the third mutation region of the third target region and the proportion of the fourth mutation region of the third target region are ≥ 85%, then the copy number variation type of the chromosome band is amplification risk Gain_risk.

[0079] In a third typical embodiment of the present application, a computer-readable storage medium is provided, which includes a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute the above-mentioned method for detecting tumor molecular typing based on low-depth whole genome sequencing.

[0080] In a fourth typical embodiment of the present application, a processor is provided, which is used to run a program, wherein the program runs the above-mentioned method for detecting tumor molecular typing based on low-depth whole genome sequencing.

[0081] The present application is further described in detail below in conjunction with specific embodiments. These embodiments should not be construed as limiting the scope of protection claimed in the present application.

[0082] Example 1

[0083] This application detection method: (such as Figure 1 (shown)

[0084] 1) Use fastp software to clean and filter the sWGS data, align the cleaned sequencing data to the human reference genome, and obtain a BAM format file. The cleaning and filtering uses the control sWGS data or negative baseline data of the test sample itself.

[0085] 2) The inaccessible part of the second generation sequencing in the whole genome in the BAM file obtained in 1) is removed, and then the first target region and the anti-target region outside the exon are divided according to the whole exon to obtain the BED format file of this series.

[0086] 3) The inaccessible part of the second generation sequencing in the whole genome in the BAM file obtained in 1) is removed, and then the second target region and the third target region are continuously split according to fixed lengths of 1M and 10K to obtain the series of BED format files.

[0087] 4) Use CNVkit software in pairing mode to calculate the copy number of each region in each series of BED files.

[0088] 5) Based on the calculation results of the software, annotate the location of the region where the copy number is calculated in the reference genome, obtain its specific location on the chromosome arm, region, and band, and determine the copy number qualitative results (loss, loss risk Loss_risk, gain, and gain risk Gain_risk) corresponding to the chromosome arm, region, and band according to the threshold (cutoff value) for determining the copy number variation.

[0089] 6) Calculate the average copy number and mutation region ratio of each region in the non-fixed bin group at the chromosome and chromosome arm level, and the average copy number and mutation region ratio of each region in the fixed bin group. The calculation method is:

[0090] Taking the whole chromosome / chromosome arm / second target region / third target region as the object, the average copy number and mutation region ratio of each region are calculated independently:

[0091] Average copy number = (the sum of the length of the region where the copy number is calculated by the software * the corresponding copy number) / the sum of the lengths of all regions where the copy number is calculated in the object;

[0092] The first mutation region ratio = the sum of the lengths of the Loss regions in the object / the sum of the lengths of all regions with calculated copy numbers in the object;

[0093] The second mutation region ratio = the sum of the lengths of the Loss_risk regions in the object / the sum of the lengths of all regions with calculated copy numbers in the object;

[0094] The third mutation region ratio = the sum of the Gain region lengths in the object / the sum of the region lengths of all calculated copy numbers in the object;

[0095] The fourth mutation region ratio=the sum of the lengths of the Gain_risk regions in the object / the sum of the lengths of all regions with calculated copy numbers in the object.

[0096] 7) Import the above calculation results into the scoring model, refer to the comparison of the average copy number, the proportion of mutation regions and the threshold matrix to determine the copy number variation type results of the corresponding regions at each level, and then obtain the corresponding molecular typing.

[0097] 1. Determination of copy number variation of the entire chromosome and its arms:

[0098] Clinically, molecular typing in this dimension objectively requires that only a single copy number variation occurs in the vast majority of regions, the average copy number will vary significantly around 2, the proportion of mutation regions is very high, and various parameters at the chromosome as a whole and chromosome arm levels are compared using non-fixed bin groups to establish a scoring model.

[0099] Loss of the entire chromosome or chromosome arm: average copy number ≤ baseline average copy number * 0.8 and the proportion of the first mutation region ≥ 85%;

[0100] The whole chromosome or chromosome arm has a loss risk Loss_risk: the average copy number is ≤ the baseline average copy number * 0.9 and the first mutation region ratio + the second mutation region ratio is ≥ 85%;

[0101] The whole chromosome or chromosome arm has amplification gain: the average copy number ≥ the baseline average copy number * 1.2 and the proportion of the third mutation region ≥ 85%;

[0102] There is a risk of amplification of the entire chromosome or chromosome arm Gain_risk: the average copy number ≥ the baseline average copy number * 1.1 and the proportion of the third mutation region + the proportion of the fourth mutation region ≥ 85%.

[0103] 2. Targeting copy number variation in chromosome regions

[0104] Combined with the copy number variation of the entire chromosome or chromosome arm, the copy number variation position in the chromosome region or band is further confirmed. The variation type of the region or band should be consistent with the chromosome as a whole and the arm.

[0105] Determination of the copy number variation type of the chromosome region:

[0106] Loss: When the chromosome arm is judged not to be a gain, the average copy number of the second target region is ≤ the average copy number of the baseline * 0.8 and the proportion of the first mutation region in the second target region is ≥ 85%;

[0107] Loss_risk: When the chromosome arm is judged not to be Gain, the average copy number of the second target region ≤ the average copy number of the baseline * 0.9 and the proportion of the first mutation region in the second target region + the proportion of the second mutation region in the second target region ≥ 85%;

[0108] Gain: When the chromosome arm is judged not to be Loss, the average copy number of the second target region ≥ the average copy number of the baseline * 1.2 and the proportion of the third mutation region of the second target region ≥ 85%;

[0109] Gain_risk: When the chromosome arm is judged not to be Loss, the average copy number of the second target region ≥ the average copy number of the baseline * 1.1 and the proportion of the third mutation region of the second target region + the proportion of the fourth mutation region of the second target region ≥ 85%.

[0110] Determination of the copy number variation type of chromosome band:

[0111] Loss: When the chromosome arm is judged not to be a gain, the average copy number of the third target region is ≤ the average copy number of the baseline * 0.8 and the proportion of the first mutation region in the third target region is ≥ 85%;

[0112] Loss_risk: when the chromosome arm is judged not to be Gain, the average copy number of the third target region ≤ the average copy number of the baseline * 0.9 and the proportion of the first mutation region in the third target region + the proportion of the second mutation region in the third target region ≥ 85%;

[0113] Gain: When the chromosome arm is judged not to be Loss, the average copy number of the third target region ≥ the average copy number of the baseline * 1.2 and the proportion of the third mutation region in the third target region ≥ 85%;

[0114] Gain_risk: When the chromosome arm is judged not to be Loss, the average copy number of the third target region ≥ the average copy number of the baseline * 1.1 and the proportion of the third mutation region of the third target region + the proportion of the fourth mutation region of the third target region ≥ 85%.

[0115] This embodiment uses a total of 12 tissue single samples with prior standard positive and negative from the hospital. The types of each sample are shown in Table 1. At the same time, multiple existing schemes are used for cross-validation to prove that the detection method of this application is feasible. The results are summarized in Table 2 below.

[0116] Among them, the existing schemes include Fish method, gene chip and Panel+Marker SNP. The specific methods are as follows:

[0117] Fish method: A technique that uses a known sequence, specific single-stranded nucleic acid as a probe, labeled with biotin or fluorescein, and uses the base complementary pairing rule at a certain temperature and ion concentration to enable DNA-DNA in situ hybridization, and uses fluorescence to display, and ultimately marks the DNA at its original position on cell slides and tissue sections (paraffin sections or frozen sections). Since the chromosomes / genes of normal people are diploid, the normal number of detection signals is two. When the number of signals is more than two or less than two (excluding the influence of the section), it can be inferred that the chromosome or gene has an abnormality (amplification or deletion).

[0118] Gene chip: also known as DNA microarray, is a chip that integrates a large number of known sequence probes on the same substrate. Several labeled target nucleotide sequences are hybridized with the probes at specific sites on the chip, and the genetic information in biological cells or tissues is analyzed by detecting the hybridization signals.

[0119] Panel+Marker SNP: Use the second-generation sequencing technology (usually using a hybrid capture platform) to perform high-depth sequencing of samples, and use probes to comprehensively capture SNP sites in all exons of sample-related genes and intron regions of some genes. After standardized library construction, use a second-generation sequencer for precise sequencing. For areas of chromosomes that may have copy number variations, more SNP sites to be tested are covered. Based on the actual copy number obtained from the corresponding SNP test of each chromosome, combined with the CNV threshold defined by the database, bioinformatics calculations are performed to evaluate the possible chromosomal heterozygous loss or homozygous loss of the target chromosome.

[0120] Table 1:

[0121] Sample_ID Sample Type Patient 01 Paraffin sections of brain glioma Patient 02 Paraffin sections of brain glioma Patient 03 Fresh glioma tissue Patient 04 Fresh glioma tissue Patient 05 Paraffin sections of brain glioma Patient 06 Paraffin sections of brain glioma Patient 07 Paraffin sections of brain malignant glioma Patient 08 Secondary glioma paraffin section Patient 09 Paraffin sections of brain glioma Patient 10 Paraffin sections of brain glioma Patient 11 Paraffin sections of brain glioma Patient 12 Paraffin sections of brain glioma

[0122] Table 2:

[0123]

[0124]

[0125] Among them, 1p_DEL is the deletion of the short arm of chromosome 1, 19q_DEL is the deletion of the long arm of chromosome 19, chr7_AMP is the overall amplification of chromosome 7, and chr10_DEL is the overall deletion of chromosome 10. The results of various detection schemes "-" mean that the corresponding test was not performed.

[0126] Figure 2The figure shows the result of the variation type of all chromosomes of all 12 samples sequenced by sWGS, which is drawn by the R package of the R language platform based on the calculation results of the scoring model. Sample 1 is a priori negative for 1p19q deletion and positive for Chr7 amplification. Figure 3 The sWGS sequencing results shown clearly show that the results are consistent with the prior knowledge. In addition, amplification of Chr12 and Chr18 chromosomes was detected in the samples.

[0127] For the same sample 1, Figure 4 and Figure 5 As shown, the Panel+Marker SNP test results showed that 1p19q deletion was positive, which was inconsistent with the priori. It may be that there will be inaccurate detection during odd-number amplification. Chr7 amplification was positive, which was consistent with the priori.

[0128] For the same sample 1, Figure 6 As shown, the gene chip test results showed that 1p19q deletion was negative, which was consistent with the priori results. In addition, the amplification of Chr7, Chr12, and Chr18 chromosomes was consistent with the sWGS results.

[0129] For sample 11, Figure 7 As shown in the sWGS sequencing results, it can be clearly seen that 1p19q deletion is negative, Chr7 amplification is negative, and Chr10 deletion is negative, which is consistent with the prior results. Figure 8 As shown, the gene chip test results showed that 1p, 19q, Chr7, and Chr10 were all neutral results, corresponding to negative. It can be concluded that 1p19q deletion was negative, Chr7 amplification was negative, and Chr10 deletion was negative, which was consistent with the prior results. However, 19p and 19q both detected additional small area deletions and amplifications, which had a certain interference with the accuracy.

[0130] From the above description, it can be seen that the above-mentioned embodiments of the present invention achieve the following technical effects: the sWGS sequencing data is divided into different block (bin) combinations according to different standards for subsequent copy number analysis, including being divided into non-fixed bin groups based on whether they are exons and being continuously divided into fixed bin groups based on 1M and 10K lengths, and the copy number data of different bin groups are judged to determine the copy number variation types of chromosomes at different hierarchical structural positions, and then the corresponding molecular typing is obtained, thereby achieving more accurate molecular typing detection of tumor chromosomes at various levels, which is further suitable for the detection of molecular typing of different cancer subspecies.

[0131] In addition, the detection method of the present application can be applied on all high-throughput sequencing platforms, supports single-sample detection (mixed negative samples can be used instead of controls), supports detection and verification of copy number variations in specific large regions, and can be extended to other projects that require chromosome-level copy number detection (molecular typing of other cancers, large-fragment deletions of genetic diseases, etc.).

[0132] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for detecting tumor molecular typing based on low-depth whole genome sequencing, characterized in that: The method comprises: S1, the low-depth whole-genome sequencing data are independently divided into non-fixed bin groups and fixed bin groups; S2, using the copy numbers of the non-fixed bin group and the fixed bin group to calculate the average copy number and mutation region ratio of each group; S3, establishing a scoring model based on the average copy number and the ratio of the mutation region, determining the copy number variation type of each structural level of the chromosome, and then obtaining the corresponding molecular typing; The non-fixed bin group includes a first target region belonging to the exon and an anti-target region outside the exon; the fixed bin group includes a second target region with a length of 1M and a third target region with a length of 10K obtained by continuous division; The various structural levels of the chromosome include: whole chromosome, chromosome arm, chromosome region and chromosome band.

2. The method according to claim 1, characterized in that Before performing the division, the inaccessible part of the second generation sequencing in the low-depth whole genome sequencing data is removed; Preferably, the low-depth whole genome sequencing data is a BAM file obtained by alignment with the human reference genome after quality control.

3. The method according to claim 1, characterized in that The copy numbers of the non-fixed bin group and the fixed bin group are calculated using CNVkit software respectively; Preferably, before calculating the average copy number and mutation region ratio of each group, the method further comprises: Annotating the positions of the regions in the chromosome in the non-fixed bin group and the fixed bin group where the copy numbers are calculated using the CNVkit software; Determining the copy number variation type of each of the regions in the non-fixed bin group and the fixed bin group according to the copy number variation threshold; The copy number variation types include: loss, loss risk Loss_risk, amplification Gain and amplification risk Gain_risk.

4. The method according to claim 3, characterized in that The calculation of the average copy number and mutation region ratio of each group includes: Under the calculation of the chromosome as a whole and chromosome arm level, the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of each region of the copy number calculated by the CNVkit software in the unfixed bin group are used; Respectively calculating the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio, and the fourth mutation region ratio of the second target region and the third target region in the fixed bin group in which the copy number is calculated by the CNVkit software; The average copy number = the sum of (the length of the region in the object where the copy number is calculated by the CNVkit software*the corresponding copy number of the region) / the sum of the lengths of all the regions in the object where the copy number is calculated; The first mutation region ratio=the sum of the lengths of the Loss regions in the object / the sum of the lengths of all the regions in the object whose copy numbers are calculated; The second mutation region ratio=the sum of the lengths of the Loss_risk regions in the object / the sum of the lengths of all the regions in the object whose copy numbers are calculated; The third mutation region ratio=the sum of the Gain region lengths in the object / the sum of the region lengths of all the calculated copy numbers in the object; The fourth mutation region ratio=the sum of the lengths of the Gain_risk regions in the object / the sum of the lengths of all the regions in the object where the copy numbers are calculated; Wherein, the object is: the entire chromosome, the chromosome arm, the second target region or the third target region.

5. The method according to claim 4, characterized in that The S3 includes: Using the average copy number at the chromosome overall level, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio, and the fourth mutation region ratio, a first scoring model is established to determine the copy number variation type of the chromosome as a whole; Establishing a second scoring model by using the average copy number at the chromosome arm level, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio, and the fourth mutation region ratio to determine the copy number variation type of the chromosome arm; Based on the copy number variation type of the entire chromosome and the copy number variation type of the chromosome arm, a third scoring model is established using the average copy number of the second target region, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio to further determine the copy number variation type of the chromosome region; Based on the copy number variation type of the chromosome as a whole and the copy number variation type of the chromosome arm, a fourth scoring model is established using the average copy number of the third target region, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio to further determine the copy number variation type of the chromosome band.

6. The method according to claim 5, characterized in that The copy number variation type of the entire chromosome and the chromosome arm is determined according to the following criteria: When the average copy number of the object to be judged is ≤ the baseline average copy number*0.8 and the first mutation region ratio of the object to be judged is ≥85%, the copy number variation type of the object to be judged is Loss; When the average copy number of the judgment object is ≤ the baseline average copy number*0.9 and the sum of the first mutation region ratio of the judgment object and the second mutation region ratio of the judgment object is ≥85%, the copy number variation type of the judgment object is loss risk Loss_risk; When the average copy number of the judgment object is ≥ the baseline average copy number*1.2 and the third mutation region ratio of the judgment object is ≥85%, the copy number variation type of the judgment object is amplification Gain; When the average copy number of the judgment object is ≥ the baseline average copy number*1.1 and the third mutation region ratio of the judgment object and the fourth mutation region ratio of the judgment object are ≥85%, the copy number variation type of the judgment object is amplification risk Gain_risk; The judgment object is the entire chromosome or the chromosome arm.

7. The method according to claim 5, characterized in that The copy number variation type of the chromosome region is determined according to the following criteria: When the copy number variation type of the chromosome arm in the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ baseline average copy number*0.8 and the first mutation region ratio of the second target region is ≥85%, then the copy number variation type of the chromosome region is deletion Loss; When the copy number variation type of the chromosome arm in the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ baseline average copy number*0.9 and the sum of the first mutation region ratio of the second target region and the second mutation region ratio of the second target region is ≥85%, then the copy number variation type of the chromosome region is deletion risk Loss_risk; When the copy number variation type of the chromosome arm in the chromosome region is not Loss, when the average copy number of the second target region is ≥ baseline average copy number*1.2 and the third mutation region ratio of the second target region is ≥85%, then the copy number variation type of the chromosome region is Gain; When the copy number variation type of the chromosome arm in the chromosome region is not loss, when the average copy number of the second target region is ≥ baseline average copy number*1.1 and the ratio of the third mutation region in the second target region to the fourth mutation region in the second target region is ≥85%, then the copy number variation type of the chromosome region is amplification risk Gain_risk.

8. The method according to claim 5, characterized in that The copy number variation type of the chromosome band is determined according to the following criteria: When the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ baseline average copy number*0.8 and the first mutation region ratio of the third target region is ≥85%, the copy number variation type of the chromosome band is deletion Loss; When the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ baseline average copy number*0.9 and the sum of the first mutation region ratio of the third target region and the second mutation region ratio of the third target region is ≥85%, then the copy number variation type of the chromosome band is deletion risk Loss_risk; When the copy number variation type of the chromosome arm in the chromosome band is not Loss, when the average copy number of the third target region is ≥ baseline average copy number*1.2 and the third mutation region ratio of the third target region is ≥85%, then the copy number variation type of the chromosome band is Gain; When the copy number variation type of the chromosome arm of the chromosome band is not loss, when the average copy number of the third target region is ≥ baseline average copy number*1.1 and the ratio of the third mutation region of the third target region to the ratio of the fourth mutation region of the third target region is ≥85%, then the copy number variation type of the chromosome band is amplification risk Gain_risk.

9. A device for detecting tumor molecular typing based on low-depth whole genome sequencing, characterized in that: The device comprises: The bin division module is configured to independently divide the low-depth whole genome sequencing data into non-fixed bin groups and fixed bin groups; A calculation module is configured to calculate the average copy number and mutation region ratio of each group using the copy numbers of the non-fixed bin group and the fixed bin group; A judgment module is configured to establish a scoring model based on the average copy number and the ratio of the mutation region, judge the copy number variation type of each structural level of the chromosome, and then obtain the corresponding molecular typing; The non-fixed bin group includes a first target region belonging to the exon and an anti-target region outside the exon; the fixed bin group includes a second target region with a length of 1M and a third target region with a length of 10K obtained by continuous division; The various structural levels of the chromosome include: whole chromosome, chromosome arm, chromosome region and chromosome band.

10. The device according to claim 9, characterized in that In the bin division module, before performing the division, the inaccessible part of the second generation sequencing in the low-depth whole genome sequencing data is removed; Preferably, the low-depth whole genome sequencing data is a BAM file obtained by alignment with the human reference genome after quality control.

11. The device according to claim 9, characterized in that The calculation module comprises: A CNVkit software calculation unit is configured to use the CNVkit software to respectively calculate the copy numbers of the non-fixed bin group and the fixed bin group; Preferably, before calculating the average copy number and mutation region ratio of each group, the calculation module further includes: An annotation unit is configured to annotate the positions of the regions in the chromosomes in the non-fixed bin group and the fixed bin group whose copy numbers are calculated using the CNVkit software; A threshold judgment unit is configured to judge the copy number variation type of each of the regions in the non-fixed bin group and the fixed bin group according to the copy number variation threshold; The copy number variation types include: loss, loss risk Loss_risk, amplification Gain and amplification risk Gain_risk.

12. The device according to claim 11, characterized in that The calculation module also includes: The first calculation unit is configured to calculate the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of each region of the copy number calculated by the CNVkit software in the non-fixed bin group at the chromosome whole level and the chromosome arm level respectively; A second calculation unit is configured to respectively calculate the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio of the second target region and the third target region in the fixed bin group, in the region where the copy number is calculated by the CNVkit software; The average copy number = the sum of (the length of the region in the object where the copy number is calculated by the CNVkit software*the corresponding copy number of the region) / the sum of the lengths of all the regions in the object where the copy number is calculated; The first mutation region ratio=the sum of the lengths of the Loss regions in the object / the sum of the lengths of all the regions in the object whose copy numbers are calculated; The second mutation region ratio=the sum of the lengths of the Loss_risk regions in the object / the sum of the lengths of all the regions in the object whose copy numbers are calculated; The third mutation region ratio=the sum of the Gain region lengths in the object / the sum of the region lengths of all the calculated copy numbers in the object; The fourth mutation region ratio=the sum of the lengths of the Gain_risk regions in the object / the sum of the lengths of all the regions in the object where the copy numbers are calculated; Wherein, the object is: the entire chromosome, the chromosome arm, the second target region or the third target region.

13. The device according to claim 12, characterized in that The judging module comprises: a chromosome overall copy number variation type judgment unit, configured to establish a first scoring model using the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio at the chromosome overall level to judge the chromosome overall copy number variation type; a chromosome arm copy number variation type judgment unit, configured to establish a second scoring model using the average copy number, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio at the chromosome arm level to judge the copy number variation type of the chromosome arm; a chromosome region copy number variation type judgment unit, configured to establish a third scoring model based on the copy number variation type of the entire chromosome and the copy number variation type of the chromosome arm, using the average copy number of the second target region, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio, so as to judge the copy number variation type of the chromosome region; The chromosome band copy number variation type judgment unit is configured to establish a fourth scoring model based on the copy number variation type of the chromosome as a whole and the copy number variation type of the chromosome arm, using the average copy number of the third target region, the first mutation region ratio, the second mutation region ratio, the third mutation region ratio and the fourth mutation region ratio, so as to judge the copy number variation type of the chromosome band.

14. The device according to claim 13, characterized in that The judgment criteria of the copy number variation type in the chromosome whole copy number variation type judgment unit and the chromosome arm copy number variation type judgment unit include: When the average copy number of the object to be judged is ≤ the baseline average copy number*0.8 and the first mutation region ratio of the object to be judged is ≥85%, the copy number variation type of the object to be judged is Loss; When the average copy number of the judgment object is ≤ the baseline average copy number*0.9 and the sum of the first mutation region ratio of the judgment object and the second mutation region ratio of the judgment object is ≥85%, the copy number variation type of the judgment object is loss risk Loss_risk; When the average copy number of the judgment object is ≥ the baseline average copy number*1.2 and the third mutation region ratio of the judgment object is ≥85%, the copy number variation type of the judgment object is amplification Gain; When the average copy number of the judgment object is ≥ the baseline average copy number*1.1 and the third mutation region ratio of the judgment object and the fourth mutation region ratio of the judgment object are ≥85%, the copy number variation type of the judgment object is amplification risk Gain_risk; The judgment object is the entire chromosome or the chromosome arm.

15. The device according to claim 13, characterized in that The judgment criteria of the copy number variation type of the chromosome region in the chromosome region copy number variation type judgment unit include: When the copy number variation type of the chromosome arm in the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ baseline average copy number*0.8 and the first mutation region ratio of the second target region is ≥85%, then the copy number variation type of the chromosome region is deletion Loss; When the copy number variation type of the chromosome arm in the chromosome region is not amplification Gain, when the average copy number of the second target region is ≤ baseline average copy number*0.9 and the sum of the first mutation region ratio of the second target region and the second mutation region ratio of the second target region is ≥85%, then the copy number variation type of the chromosome region is deletion risk Loss_risk; When the copy number variation type of the chromosome arm in the chromosome region is not Loss, when the average copy number of the second target region is ≥ baseline average copy number*1.2 and the third mutation region ratio of the second target region is ≥85%, then the copy number variation type of the chromosome region is Gain; When the copy number variation type of the chromosome arm in the chromosome region is not loss, when the average copy number of the second target region is ≥ baseline average copy number*1.1 and the ratio of the third mutation region in the second target region to the fourth mutation region in the second target region is ≥85%, then the copy number variation type of the chromosome region is amplification risk Gain_risk.

16. The device according to claim 13, characterized in that The judgment criteria of the chromosome band copy number variation type in the chromosome band copy number variation type judgment unit include: When the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ baseline average copy number*0.8 and the first mutation region ratio of the third target region is ≥85%, the copy number variation type of the chromosome band is deletion Loss; When the copy number variation type of the chromosome arm of the chromosome band is not amplification Gain, when the average copy number of the third target region is ≤ baseline average copy number*0.9 and the sum of the first mutation region ratio of the third target region and the second mutation region ratio of the third target region is ≥85%, then the copy number variation type of the chromosome band is deletion risk Loss_risk; When the copy number variation type of the chromosome arm in the chromosome band is not Loss, when the average copy number of the third target region is ≥ baseline average copy number*1.2 and the third mutation region ratio of the third target region is ≥85%, then the copy number variation type of the chromosome band is Gain; When the copy number variation type of the chromosome arm of the chromosome band is not loss, when the average copy number of the third target region is ≥ baseline average copy number*1.1 and the ratio of the third mutation region of the third target region to the ratio of the fourth mutation region of the third target region is ≥85%, then the copy number variation type of the chromosome band is amplification risk Gain_risk.

17. A computer-readable storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute the method for detecting tumor molecular typing based on low-depth whole genome sequencing as described in any one of claims 1 to 8.

18. A processor, characterized in that: The processor is used to run a program, wherein the program runs the method for detecting tumor molecular typing based on low-depth whole genome sequencing as described in any one of claims 1 to 8.

Images