Reconstruction method, device and equipment of high-quality metagenome and storage medium

Abstract

The invention belongs to the technical field of medical detection, solves the problems of contig error clustering, error segmentation and pollution introduction in a binning process in the existing process of recovering a high-quality genome from a metagenome, and provides a reconstruction method, device and equipment of the high-quality metagenome and a storage medium. Comprising the following steps: performing binning on each contig of a first sample by using a plurality of binning tools to generate a plurality of sample boxes, screening out a high-quality first sample box group from the plurality of sample boxes, and then finishing reassembling of a first sample genome based on the contigs of the first sample and / or sequencing sequences of other similar samples, the genome of the first sample with high quality and high integrity is obtained; according to the method, contigs of a single sample are subjected to binning and fusion by utilizing a plurality of binning tools, so that a high-quality sample box is obtained, and then the genome recombination of a target sample is completed by adopting high-quality sequencing data of a single sample or a plurality of similar samples.

Description

Technical Field

[0001] The present invention relates to the field of computer medical detection technology, and in particular to a high-quality metagenome reconstruction method, device, equipment and storage medium. Background Art

[0002] Microbial communities are widely distributed in nature and are vital to ecosystem functions and human health. Metagenomics is a culture-free technology that can analyze microbial classification and function from environmental samples. In recent years, the recovery of high-quality metagenome-assembled genomes (MAGs) has become the mainstream of microbial community research. Recovery processes such as metaWRAP have developed rapidly, which support microbial community research. However, accurately reconstructing high-quality single genomes from complex samples remains a challenge. Existing binning technologies are based on sequence features or abundance features, but the performance of a single method is limited, and there are problems such as incorrect clustering of overlapping groups, incorrect genome segmentation, and introduction of contamination, which limits the accuracy and reliability of recovering high-quality microbial genomes from metagenomes. Summary of the invention

[0003] In view of this, the embodiments of the present invention provide a method, device, equipment and storage medium for reconstructing a high-quality metagenome, so as to solve the problems in the prior art of recovering high-quality microbial genomes from metagenomes, such as incorrect clustering of overlapping groups during the binning process, incorrect segmentation of the genome and introduction of contamination.

[0004] The technical solution adopted by the present invention is:

[0005] In one aspect, the present invention provides a method for reconstructing a high-quality metagenome, the method comprising:

[0006] In response to a first reconstruction instruction of a high-quality microbial genome corresponding to a first sample, obtaining sample boxes corresponding to each contig of the first sample, wherein any contig of the first sample includes a plurality of sample boxes corresponding one-to-one to each binning tool;

[0007] Using a sample box fusion strategy to perform redundancy removal and quality screening on all the sample boxes corresponding to each overlapping group, to obtain a high-quality first sample box set;

[0008] In response to the second reconstruction instruction for reassembling the first sample, selecting at least one sample from a plurality of target samples, extracting a sequencing sequence for reassembling the first sample, and reassembling the sample to obtain the target contig;

[0009] The first sample is one of the multiple target samples, and the target contiguous group is the contiguous group of the genome corresponding to the first sample.

[0010] Preferably, the initial binning of contigs includes:

[0011] In response to the first reconstruction instruction, obtaining each of the contigs and a plurality of binning tools corresponding to the sequencing data of the first sample;

[0012] Each of the overlapping groups is binned using the binning tools to obtain a plurality of sample boxes corresponding to each overlapping group of the first sample.

[0013] Preferably, the correspondence between overlapping groups and sample boxes includes:

[0014] In response to the first reconstruction instruction, acquiring target sequencing data of the first sample, a first naming rule, and a second naming rule;

[0015] Using the first naming rule to name each contig file corresponding to the target sequencing data, to obtain each contig;

[0016] Using the second naming rule to name the sample box files corresponding to each of the overlapping groups, to obtain each sample box;

[0017] According to the first naming rule and the second naming rule, a file name relationship table between each overlapping group and each sample box is established.

[0018] Preferably, generating the first sample box set by using a first fusion strategy of sample box fusion corresponding to DSL TOOL includes:

[0019] Obtain the file name relationship table of sample boxes corresponding to overlapping groups and each binning tool;

[0020] De-redundancy processing is performed on multiple sample boxes corresponding to each overlapping group using DSL TOOL according to the file name relationship table to obtain a high-quality first candidate sample box set;

[0021] The quality of each sample box in the first candidate sample box set is calculated to screen out the first sample box set with high quality.

[0022] Preferably, the second fusion strategy corresponding to the sample box fusion of the primary binning results of multiple binning tools is performed based on sequence characteristics to generate the first sample box set, including:

[0023] Arrange and combine multiple binning tools to generate multiple combined bins, and obtain sample bin information corresponding to each overlapping group in the primary binning result according to the file name relationship table;

[0024] Re-binning each overlapping group according to the sample box information and the intersection principle of the primary binning results to obtain a second candidate sample box set consisting of each of the combined boxes;

[0025] Calculating the quality of each sample box in the second candidate sample box set, selecting a combined box that meets the requirements to represent a corresponding overlapping group, and obtaining a third candidate sample box set;

[0026] The file name relationship table corresponding to the third candidate sample box set is updated to obtain the first sample box set.

[0027] Preferably, DSL TOOL generates the first sample box set by performing a third fusion strategy corresponding to sample box fusion based on the permutation and combination results based on sequence characteristics, including:

[0028] Obtaining a fourth candidate sample box set and a fifth candidate sample box set consisting of a plurality of non-redundant sample boxes corresponding to each overlapping group;

[0029] Generating a plurality of similar clusters according to the similarities between the sample boxes in the fourth candidate sample set and the fifth candidate sample set;

[0030] Selecting a representation of a sample box for each of the similar clusters to obtain a sixth candidate sample box set consisting of the representation sample boxes of each of the similar clusters;

[0031] Deduplication is performed according to the similarity of each sample box in the sixth candidate sample box set to obtain the first sample box set;

[0032] Among them, the fourth candidate sample box set is a non-redundant sample box set generated by merging sample boxes corresponding to DSL TOOL, and the fifth candidate sample box set is a non-redundant sample box set generated by arranging and combining the primary binning results of each overlapping group based on sequence characteristics according to multiple binning tools.

[0033] Preferably, obtaining the target overlapping group comprises:

[0034] In response to the second reconstruction instruction, obtaining the first sample box sets corresponding to all target samples;

[0035] Performing quality screening on a sample box set consisting of the first sample box set of all target samples to obtain a target sample box and an assembly scheme for reconstruction;

[0036] According to the assembly scheme, selecting sequencing data for reconstruction from a reference genome corresponding to at least one sample among all the target samples to assemble the genome of the first sample to obtain the target contig;

[0037] The reference genome of each target sample contains all overlapping nucleic acid sequences of the target sample box.

[0038] On the other hand, the present invention also provides a high-quality metagenome reconstruction device, the device comprising:

[0039] A data response module: configured to respond to a first reconstruction instruction of a high-quality microbial genome corresponding to a first sample, and obtain sample boxes corresponding to each contig of the first sample, wherein any contig of the first sample includes a plurality of sample boxes corresponding one-to-one to each binning tool;

[0040] Data integration module: used for removing redundancy and quality screening of all the sample boxes corresponding to each overlapping group by using a sample box fusion strategy to obtain a high-quality first sample box set;

[0041] A data reorganization module: configured to respond to the second reconstruction instruction of the first sample reassembly, select each target sample box for reassembly from the first sample box set corresponding to a plurality of target samples, and reassemble to obtain a target overlapping group;

[0042] The first sample is one of the multiple target samples, and the target contiguous group is the contiguous group of the genome corresponding to the first sample.

[0043] On the other hand, the present invention also provides an electronic device, at least one processor, at least one memory and computer program instructions stored in the memory, which implement any of the above methods when the computer program instructions are executed by the processor.

[0044] On the other hand, the present invention further provides a computer-readable storage medium having computer program instructions stored thereon, which implement any of the above-mentioned methods when the computer program instructions are executed by a processor.

[0045] In summary, the beneficial effects of the present invention are as follows:

[0046] The high-quality metagenome reconstruction method, device, equipment and storage medium provided by the present invention include that each overlapping group of the first sample is binned by multiple binning tools, and then the sample boxes after each overlapping group of the first sample processed by the multiple binning tools are de-redundant and quality screened to obtain a high-quality first sample box group, and then the first sample genome is reassembled based on the overlapping groups of the first sample and / or the sequencing sequences of similar other samples, so as to obtain a genome of the first sample with high quality and high integrity; the present invention obtains high-quality sample boxes by binning and fusing the overlapping groups of a single sample using multiple binning tools, and then uses the high-quality sequencing data of a single sample or similar multiple samples to complete the recombination of the target single sample genome, so as to obtain a high-quality and high-integrity genome. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the drawings required for use in the embodiment of the present invention will be briefly introduced below. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work, and these are all within the protection scope of the present invention.

[0048] Figure 1 Schematic diagram of the process of reconstructing a high-quality metagenome in Example 1 of the present invention;

[0049] Figure 2 This is a schematic diagram of a process for obtaining a file name relationship table in Embodiment 1 of the present invention;

[0050] Figure 3 This is a schematic diagram of the process of obtaining a target overlapping group in Example 1 of the present invention;

[0051] Figure 4 A logical schematic diagram of the sequence data reconstruction of each sample box in Example 1 of the present invention;

[0052] Figure 5 This is a schematic diagram of the structure of a high-quality metagenome reconstruction device in Example 2 of the present invention;

[0053] Figure 6 Schematic diagram of the structure of an electronic device in Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described in conjunction with the drawings in the embodiment of the present invention. It should be noted that, in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. In the description of the present invention, it should be understood that the orientation or position relationship indicated by the terms "center", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc. is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. Moreover, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such a process, method, article or device. In the absence of further restrictions, the elements defined by the phrase "comprising..." do not exclude the existence of other identical elements in the process, method, article or device comprising the elements. If there is no conflict, the various features of the present invention and the embodiments can be combined with each other, all within the protection scope of the present invention.

[0055] Example 1

[0056] The high-quality metagenome reconstruction method of the present invention includes fusion binning, sample binning refinement and reassembly and a unified processing framework, wherein the fusion binning is to bin the overlapping groups of the same gene sample using multiple binning tools, and the binning results are fused to screen out high-quality sample boxes, the sample binning refinement and reassembly is to refine the high-quality sample boxes to further improve the quality of the sample boxes, and then the sequencing data of the refined sample boxes are reorganized to obtain the target overlapping groups of high-quality gene samples; the unified processing architecture is to realize the fusion binning and sample binning refinement and reassembly functions, and to construct a DSL-based workflow framework to modularly integrate the binning, refinement and reassembly tasks into a working environment, which can realize the flexible matching of various analysis processes and realize information sharing between multiple tasks and multiple links to meet the combination requirements of high-quality gene samples, and at the same time, reduce manual intervention and improve the speed of metagenome analysis.

[0057] See also Figure 1 , Figure 1 The figure is a schematic diagram of a process for reconstructing a high-quality metagenome according to the present invention, wherein the method comprises:

[0058] S1: in response to a first reconstruction instruction of a high-quality microbial genome corresponding to a first sample, obtaining sample boxes corresponding to each contig of the first sample, wherein any contig of the first sample includes a plurality of sample boxes corresponding one-to-one to each binning tool;

[0059] In one embodiment, the S1 includes:

[0060] S11: in response to the first reconstruction instruction, obtaining each of the contigs and a plurality of binning tools corresponding to the sequencing data of the first sample;

[0061] S12: using each of the binning tools to perform binning processing on each of the overlapping groups to obtain a plurality of the sample boxes corresponding to each of the overlapping groups of the first sample.

[0062] Specifically, the first reconstruction instruction is used to restore the high-quality genome of the first sample. Based on the first reconstruction instruction, the system obtains the overlapping groups of the first sample that needs genome reconstruction after gene sequencing and the sample boxes corresponding to the overlapping groups. It should be noted that each overlapping group of the first sample is binned by at least two binning tools, that is, the present invention uses a variety of binning tools to bin all overlapping groups of the first sample separately, so that each overlapping group will obtain multiple sample boxes after binning by different binning tools; wherein, the binning tools include but are not limited to: metaBAT2, MaxBin2, CONCOCT, MetaBinner, binny, COMEBin, SemiBin and other binning tools.

[0063] S2: using a sample box fusion strategy to perform redundancy removal and quality screening on all the sample boxes corresponding to each overlapping group, to obtain a high-quality first sample box set;

[0064] Specifically, each overlapping group of the first sample will be binned using multiple binning tools, so each overlapping group will obtain multiple binning results, and each binning result of any overlapping group corresponds to each binning tool one by one, and then the binning results of each overlapping group are integrated using the sample box fusion strategy, which includes de-redundancy of sample boxes, quality assessment, abundance calculation, etc., so as to obtain a first sample box set consisting of de-redundant and high-quality sample boxes; wherein, the sample box fusion strategy includes quality optimization based on DAS TOOL (Dereplication, Aggregation, and Scoring Tool is an automated tool designed for genome analysis metagenomics), sample box fusion based on the arrangement and combination of primary binning results of multiple binning tools based on sequence characteristics, and sample box fusion based on DSL TOOL combined with the arrangement and combination results based on sequence characteristics (the two are combined for optimization). For the convenience of description in this article, the sample box fusion corresponding to DSLTOOL is recorded as the first fusion strategy, the sample box fusion based on the arrangement and combination of primary binning results of multiple binning tools based on sequence characteristics is recorded as the second fusion strategy, and the DSL TOOL combines the permutation and combination results based on sequence characteristics to perform sample box fusion, which is recorded as the third fusion strategy.

[0065] S3: In response to the second reconstruction instruction for reassembling the first sample, selecting at least one sample from a plurality of target samples, extracting a sequencing sequence for reassembling the first sample, and performing reassembly to obtain the target contig;

[0066] The first sample is one of the multiple target samples, and the target contiguous group is the contiguous group of the genome corresponding to the first sample.

[0067] Specifically, in response to the second reconstruction instruction of the genome of the first sample, each target sample box corresponding to the overlapping group used to reconstruct the genome of the first sample is screened out from the first sample box set of all target samples, and by extracting the test sequence of the genome corresponding to the first sample from multiple target samples, as many test sequences of the genome corresponding to the first sample as possible can be obtained, thereby assembling a genome with higher completeness. Specifically, sequencing data for reconstructing the first sample is selected from the reference genome corresponding to at least one sample among all target samples, and the first sample is reconstructed one by one in units of overlapping groups; redundancy removal, abundance calculation, classification annotation and other operations are performed on the first sample box set of each overlapping group of all target samples, so as to obtain a genome that satisfies the representativeness of each overlapping group. Sample boxes, and further screening out target boxes corresponding to the genome reconstruction or reconstruction reference of the first sample (the target box contains all the overlapping groups that need to be reconstructed for the first sample), such as: after a plurality of representative sample box sets are screened out from the first sample box sets of all target samples, a value assessment is performed on each sample box, such as selecting sample boxes with a completeness between 50% and 90%, a whole genome coverage read number greater than 100,000, and an average coverage depth greater than 1 to enter the next reconstruction analysis, and then use assembly tools to reconstruct and assemble the sequencing data corresponding to the sample boxes that meet the requirements to obtain the final target overlapping groups of the genome of the first sample; wherein, the assembly tools include but are not limited to metaSPAdes or MEGAHIT.

[0068] The method for reconstructing a high-quality metagenome provided by the present invention comprises that each overlapping group of a first sample is binned by a plurality of binning tools, and then the sample boxes after each overlapping group of the first sample processed by the plurality of binning tools are de-redundanted and quality screened to obtain a high-quality first sample box group, and then the genome of the first sample is reassembled based on the overlapping groups of the first sample and / or the sequencing sequences of other similar samples, so as to obtain a genome of the first sample with high quality and high integrity. The present invention obtains high-quality sample boxes by binning and fusing the overlapping groups of a single sample using a plurality of binning tools, and then the high-quality sequencing data of a single sample or similar multiple samples is used to complete the recombination of the genome of the target single sample, so as to obtain a genome with high quality and high integrity.

[0069] In one embodiment, see Figure 2 , the S1 comprises:

[0070] S13: In response to the first reconstruction instruction, obtaining target sequencing data of the first sample, a first naming rule, and a second naming rule;

[0071] S14: naming each contig file corresponding to the target sequencing data using the first naming rule to obtain each contig;

[0072] S141: In response to the first reconstruction instruction, obtaining target sequencing data and a first naming rule of the first sample;

[0073] S142: Performing genome assembly on the target sequencing data to obtain each contig corresponding to the target sequencing data of the first sample, and naming each contig using the first naming rule.

[0074] S15: using the second naming rule to name the sample box files corresponding to each of the overlapping groups to obtain each sample box;

[0075] S16: According to the first naming rule and the second naming rule, a file name relationship table between each overlapping group and each sample box is established.

[0076] Specifically, based on the first reconstruction instruction of the genome reconstruction of the first sample, the target sequencing data of the first sample is assembled into each contig using an assembly tool, and each contig is named according to a first naming rule, the assembly tool includes but is not limited to: metaSPAdes or MEGAHIT, and the naming rule is such as: gene sample ID + serial number. Each contig of the first sample is binned, and each contig is binned by a plurality of different binning tools, so that each contig can obtain a plurality of sample boxes binned by different binning tools, and the nucleic acid sequence file (fasts format) corresponding to each sample box is named according to a second naming rule, and the naming rule is such as: gene sample ID + binning tool name + serial number.

[0077] Furthermore, the target sequencing data (clean reads) and the contig sequence file (fasta format) after genome assembly are simultaneously input into multiple binning tools. After the binning is completed, a file name relationship table of the contig ID and the sample box ID is generated. Similarly, this step will output the nucleic acid sequence file in fasta format corresponding to the sample box obtained by each binning tool for each contig binning process.

[0078] See also Figure 2 , Figure 2 The logic diagram for fusing multiple binning results to obtain a high-quality fused sample set:

[0079] In one embodiment, the first sample box set is generated by using a first fusion strategy for sample box fusion corresponding to DSL TOOL, and S1 includes:

[0080] S21: Obtain the file name relationship table of sample boxes that correspond one-to-one between overlapping groups and each binning tool;

[0081] S22: De-redundancy processing is performed on multiple sample boxes corresponding to each overlapping group using DSL TOOL according to the file name relationship table to obtain a high-quality first candidate sample box set corresponding to each overlapping group one by one;

[0082] S23: Calculate the quality of each sample box in the first candidate sample box set, and screen out the first sample box set with high quality.

[0083] Specifically, DSL TOOL is used to receive the file name relationship table. DSL TOOL removes redundancy from multiple sample boxes obtained by binning of each overlapping group by different binning tools according to the file name relationship table, thereby obtaining a non-redundant bin (sample box) set, and then uses the CheckM2 tool to perform quality assessment on each sample box in the non-redundant bin (sample box) set, and screens out sample boxes that meet the requirements to form a first sample box set; it should be noted that: the redundancy of the first sample is to remove redundancy from the sample boxes of all target samples together, wherein the target samples are from the same source, and this method of redundancy removal can ensure that the first sample box sets selected by multiple samples are non-redundant, thereby improving the sequencing data selected from multiple samples for the first sample during genome reconstruction is non-redundant, thereby improving the efficiency of genome reconstruction.

[0084] In one embodiment, the first sample box set is generated by performing a second fusion strategy corresponding to sample box fusion on the primary boxing results of multiple boxing tools based on sequence characteristics, and S2 includes:

[0085] S24: performing permutation and combination of multiple binning tools to generate multiple combined bins, and obtaining sample bin information corresponding to each overlapping group in the primary binning result according to the file name relationship table;

[0086] S25: re-binning each overlapping group according to the sample box information and the intersection principle of the primary binning results, to obtain a second candidate sample box set consisting of each of the combined boxes;

[0087] S26: Calculate the quality of each sample box in the second candidate sample box set, select a combined box that meets the requirements to represent a corresponding overlapping group, and obtain a third candidate sample box set;

[0088] S27: Update the file name relationship table corresponding to the third candidate sample box set to obtain the first sample box set.

[0089] Specifically, the integration module receives multiple binning tool file name relationship tables and lists all possible permutations and combinations; further, each combination is re-binned according to the co-occurrence principle, and the steps include:

[0090] Step 1: Obtain the binning result information of each contig from the file name relationship table;

[0091] Step 2: Sort overlapping groups in descending order according to sequence length; in this order, count the sample boxes (referred to as primary boxes) corresponding to the primary bins corresponding to each overlapping group in turn, and then divide the overlapping groups contained in all primary boxes involved in the combination box into the combination box according to the correspondence between each sample box corresponding to the primary bin. Among them, the longer the sequence, the higher the accuracy of the longer sequence in statistical probability, and the higher the probability of its real existence. Therefore, with it as the center, find other overlapping groups in the same box. This can reduce the problem of confusion in the attribution of boxes in disordered allocation and shorten the calculation time.

[0092] Step 3: If the overlapping group has been re-divided, refer to the result of the previous step. If the overlapping group set has not been divided into the specified sample box, repeat the above steps until all overlapping groups have completed the re-division of the overlapping group set of the sample box.

[0093] After obtaining the first sample box set of all permutations and combinations of sample boxes, invalid sample box (bin) combinations are eliminated according to the preset sample box quality threshold (first quality threshold) to obtain the third candidate sample box set, such as: eliminating bins with a total length less than 0.5M or greater than 20M to ensure that the bin size is consistent with the size range of the general genome of microorganisms, and obtaining a third candidate sample box set that meets the requirements; then use the CheckM2 tool to perform quality assessment on the sample boxes of the third candidate sample box set, screen out sample boxes that meet the requirements, and form the first sample box set.

[0094] In one embodiment, four binning tools, metaBAT2 (tool A), CONCOCT (tool B), MetaBinner (tool C), and SemiBin (tool D) are required for specimen 1. Assume that specimen 1 is assembled into 10 contigs, contig1, contig2, contig3, contig4, contig5, contig6, contig7, contig8, contig9, and contig10 (this order is already in descending order of length).

[0095] One of the permutations is MetaBinner and SemiBin.

[0096] MetaBinner binning results in 3 bins: MetaBinner-bin1 includes contig1, contig3, and contig5; MetaBinner-bin2 includes contig2, contig4, contig6, and contig8; MetaBinner-bin3 includes contig7, contig9, and contig10.

[0097] SemiBin binning resulted in two bins: SemiBin-bin1 included contig1, contig3, and contig5; SemiBin-bin2 included contig2, contig4, contig6, and contig10; contig8, contig7, and contig9 could not be binned and belonged to the unbinned overlapping group.

[0098] The boxes are re-divided starting from contig1. Contig1 is in MetaBinner-bin1 and SemiBin-bin1. Contigs appearing in these two boxes include contig1, contig3, and contig5. Then contig1, contig3, and contig5 are re-divided into the MetaBinner-SemiBin-bin1 box.

[0099] Then contig2 is re-divided into boxes. Contig2 is in MetaBinner-bin2 and SemiBin-bin2. Contigs appearing in these two boxes include contig2, contig4, and contig6. Then contig2, contig4, and contig6 are re-divided into the MetaBinner-SemiBin-bin2 box.

[0100] Then contig3, contig4, contig5, and contig6 have been divided in the previous step and do not need to be divided again.

[0101] Finally, contig8, contig7, contig9, and contig10 were classified as unbinnable contigs because they did not meet the co-occurrence principle.

[0102] Finally, in the combination of MetaBinner and SemiBin, specimen 1 obtained two fused bins, MetaBinner-SemiBin-bin1 bin (contig1, contig3, contig5) and MetaBinner-SemiBin-bin2 bin (contig2, contig4, contig6).

[0103] Similarly, suppose that in the combination of MetaBinner and CONCOCT, specimen 1 gets 2 fused bins, MetaBinner-CONCOCT-bin1 bin (contig1, contig3, contig5), MetaBinner-CONCOCT-bin2 (contig2, contig4, contig6, contig8).

[0104] Subsequently, CheckM2 performs quality assessment on MetaBinner-SemiBin-bin1, MetaBinner-SemiBin-bin2, MetaBinner-CONCOCT-bin1, and MetaBinner-CONCOCT-bin2. Because MetaBinner-SemiBin-bin1 and MetaBinner-CONCOCT-bin1 contain more than 80% identical overlapping groups, they are considered to be similar clusters. Similarly, MetaBinner-SemiBin-bin2 and MetaBinner-CONCOCT-bin2 are a similar cluster. According to the CheckM2 results, MetaBinner-SemiBin-bin1 has higher quality than MetaBinner-CONCOCT-bin1, and the first sample box set retains MetaBinner-CONCOCT-bin1. The quality of MetaBinner-CONCOCT-bin2 is higher than that of MetaBinner-SemiBin-bin2, and the first sample box set retains MetaBinner-CONCOCT-bin2.

[0105] Finally, the second fusion strategy generates the first sample bin set MetaBinner-CONCOCT-bin1 (contig1, contig3, contig5) and MetaBinner-CONCOCT-bin2 (contig2, contig4, contig6, contig8).

[0106] In one embodiment, DSL TOOL generates the first sample box set by combining the results of permutation and combination based on sequence characteristics with a third fusion strategy corresponding to sample box fusion, and S2 includes:

[0107] S021: Obtain a fourth candidate sample box set and a fifth candidate sample box set consisting of a plurality of non-redundant sample boxes corresponding to each overlapping group;

[0108] S022: generating a plurality of similar clusters according to the similarities between the sample boxes in the fourth candidate sample set and the fifth candidate sample set;

[0109] S023: performing representation selection of sample boxes for each of the similar clusters to obtain a sixth candidate sample box set consisting of the representation sample boxes of each of the similar clusters;

[0110] S024: removing duplicates according to the similarity of each sample box in the sixth candidate sample box set to obtain the first sample box set;

[0111] The fourth candidate sample box set is a non-redundant sample box set generated by fusing sample boxes corresponding to DSL TOOL, and the fifth candidate sample box set is a non-redundant sample box set generated by fusing sample boxes through permutations and combinations based on gene characteristics.

[0112] Specifically, DSL TOOL is first used to remove redundancy and eliminate low-quality samples (such as eliminating sample boxes with a completion rate of less than 50% or a contamination rate of more than 10%) from the sample box set of each overlapping group to ensure that the quality of the sample boxes participating in the integration meets the preliminary requirements, thereby obtaining the fourth candidate sample box set. Similarly, the sample box sets of each overlapping group are arranged and combined in the same way based on sequence characteristics to obtain the fifth candidate sample box set. Then, similar clusters are constructed based on the fourth candidate sample box set and the fifth candidate sample box set. The construction method of the similar clusters is as described above and will not be repeated here. Further, the sample boxes characterized are screened based on similar clusters, and all sample boxes selected as representatives from each similar cluster are recorded as the sixth candidate sample box set. Then, the similarity of each sample box in the sixth candidate sample box set is used to remove duplication, thereby obtaining a first sample box set with de-redundancy and high quality. Similarly, each sample box of the first sample box set is subjected to quality assessment, that is, the final output file includes the nucleic acid sequence file (fasta format) and quality report of the first sample box set.

[0113] Furthermore, the construction of similar clusters includes:

[0114] S0221: traverse the fifth candidate sample box set using each sample box in the fourth candidate sample box set to obtain a plurality of first similar clusters of sample boxes and a fifth candidate sample box subset corresponding to each sample box in the fifth candidate sample set that does not form a similar cluster;

[0115] S0222: Traverse the interior of each sample box in the fifth candidate sample box subset to obtain multiple second similar clusters of the sample boxes;

[0116] S0223: Select and characterize each similar cluster of the first similar cluster and the second similar cluster to obtain a sixth candidate sample box set.

[0117] Specifically, firstly, a similar cluster is constructed for each sample box in the fourth candidate sample box set, each sample box in the fourth candidate sample box set is traversed, and the similarity between each sample box in the fourth candidate sample box set and each sample box in the fifth candidate sample box set is calculated in turn. If the similarity is greater than the first similarity threshold, two sample boxes whose similarity is greater than the first similarity threshold are recorded as the first similarity cluster, thereby obtaining multiple first similarity clusters; secondly, a similar cluster is constructed for each sample box in the fifth candidate sample box set that does not form a similar cluster, and each sample box in the fifth candidate sample box set that does not form a similar cluster is recorded as the fifth candidate sample box subset, and each sample box in the fifth candidate sample box subset is traversed, and the similarity between each sample box in the fifth candidate sample box subset and each sample box in the fifth candidate sample box subset is calculated in turn. If the similarity is greater than the first similarity threshold, the two sample boxes whose similarities are greater than the first similarity threshold are recorded as the second similar clusters, thereby obtaining multiple second similar clusters; again, representative sample boxes are selected for all the first similar clusters and the second similar clusters, and the quality (such as completeness and contamination rate, etc.) of each sample box in the similar cluster is calculated based on the similar cluster, and the sample boxes with high quality are used as representatives of the similar cluster, and the sample boxes representing each similar cluster are used as the sixth candidate sample box set. In a preferred embodiment, the sixth candidate sample box set also includes the sample boxes in the fourth candidate sample box set and the fifth candidate sample box subset that do not form similar clusters; finally, the sixth candidate sample box set is deduplicated and quality evaluated to obtain a high-quality, non-redundant first sample box set, and each sample box in the first sample box set has the characteristics of uniqueness and non-redundancy.

[0118] It should be noted that the final results include the first sample box set, the nucleic acid sequence file of the sample box and the quality report of the sample box. The quality report includes the integrity, contamination rate and other evaluation results (such as abundance table, classification annotation table) of each sample box.

[0119] In one embodiment, see Figure 3 , said S3 comprises:

[0120] S31: in response to the second reconstruction instruction, obtaining the first sample box sets corresponding to all target samples;

[0121] S32: performing quality screening on the sample box set consisting of the first sample box set of all target samples to obtain target sample boxes and assembly schemes for reconstruction;

[0122] S33: According to the assembly scheme, sequencing data for reconstruction is selected from the reference genome corresponding to at least one sample among all the target samples to assemble the genome of the first sample to obtain the target contig.

[0123] The reference genome of each target sample contains all overlapping nucleic acid sequences of the target sample box, and the assembly scheme includes co-assembly involving multiple samples and assembly involving a single sample.

[0124] Specifically, Figure 4 As shown, after the target sample box of the first sample is determined, the target genome of the target sample box is first determined, and the sequencing data that can be aligned to this genome is extracted from the sequencing data for reference sequences (references) and recorded as the first target genome. The first target genome includes all overlapping group nucleic acid sequences of the target sample box of the first sample. When there is a GTDB reference genome similar to the target sample box of the first sample, this GTDB reference genome will also be included.

[0125] Next, the reassembly strategy is determined. If only one sample sequencing data is used for reassembly, it is considered as the first reassembly strategy. If multiple sample sequencing data are used for joint reassembly, it is considered as the second reassembly strategy.

[0126] Subsequently, the sequencing data required for assembling the target sample box is determined, and the sequencing data used for assembling the target sample box is recorded as the first sequencing data.

[0127] If it is the first reassembly strategy, use BWA and samtools tools, refer to the first target genome, and extract paired sequencing data from the first sample sequencing data as the first sequencing data.

[0128] If it is the second reassembly strategy, first select the target sample box of the first sample and the second candidate reassembly sample that meets the abundance requirement, use BWA and samtools tools, refer to the first target genome, extract paired sequencing data from the second candidate reassembly sample sequencing data as the second candidate reassembly sample sequencing data, and similarly extract paired sequencing data from the first sample sequencing data as the first reference sequencing data. Use the Mash tool to calculate the distance between the second candidate reassembly sample sequencing data and the first reference sequencing data, so as to select multiple second candidate reassembly sample sequencing data that are most similar to the first reference sequencing data, and use them as the first sequencing data.

[0129] Finally, the first sequencing data is assembled to obtain the target contiguous group of the reassembled target sample box of the first sample.

[0130] In a preferred embodiment, the overlapping groups obtained after assembling the first sequencing data are recorded as reference overlapping groups, and the obtained reference overlapping groups are further quality-controlled and filtered to obtain target overlapping groups, wherein the quality control boiler of the reference overlapping groups includes: using the IQR method, the weighted Z score method, the modified Z score and other methods to remove overlapping groups with outliers in coverage depth and overlapping groups with a length of less than 2000bp to reduce the reassembly noise. The sample boxes corresponding to each target overlapping group of the first sample after quality control will be re-evaluated using CheckM2, and the sample boxes corresponding to overlapping groups that do not meet the requirements will be subjected to at least one round of quality control, multi-strategy fusion, refinement and reconstruction, such as: bins with a completion degree of >90% and contamination of <5% will be directly output as high-quality overlapping groups, and sample boxes with a completion degree of <50% will no longer be subjected to secondary analysis. For sample boxes that do not meet the standards, SemiBin2 (single coverage strategy) and MetaBAT2 (multiple coverage strategy) will be used for re-binning, combined with DAS After the optimization and integration of the tool, CheckM2 is used to re-evaluate the quality. For the sample boxes that still do not meet the standards, the DeePurify tool is used to refine them again to obtain new sample boxes, and the quality improvement of the sample boxes is evaluated. After completing the given round of reconstruction, if the quality is not improved after refinement, it is considered that the optimization has failed and the analysis is exited, and the sample boxes that do not meet the requirements are finally eliminated, that is: the target overlapping group of the reorganized first sample will also be filtered for quality control to obtain overlapping groups that meet the requirements and secondary overlapping groups that need further processing; the above binning, selection and reconstruction procedures are repeated for the secondary overlapping groups to obtain the target overlapping group of the reconstructed first sample, and the quality of the reconstructed overlapping group is evaluated again using CheckM2 to obtain target overlapping groups that meet the requirements and secondary overlapping groups that need further processing. Repeat the above operations until the number of repetitions reaches the threshold, exit the quality improvement program of the sample box, and output the final target overlapping group. Through multiple rounds of refinement, quality control and reconstruction, an optimization path is finally provided for low-quality sample boxes, improving the effectiveness of the overall analysis.

[0131] The method for reconstructing a high-quality metagenome provided by the present invention comprises that each overlapping group of a first sample is binned by a plurality of binning tools, and then the sample boxes after each overlapping group of the first sample processed by the plurality of binning tools are de-redundanted and quality screened to obtain a high-quality first sample box group, and then the genome of the first sample is reassembled based on the overlapping groups of the first sample and / or the sequencing sequences of other similar samples, so as to obtain a genome of the first sample with high quality and high integrity. The present invention obtains high-quality sample boxes by binning and fusing the overlapping groups of a single sample using a plurality of binning tools, and then the high-quality sequencing data of a single sample or similar multiple samples is used to complete the recombination of the genome of the target single sample, so as to obtain a genome with high quality and high integrity.

[0132] Example 2

[0133] See also Figure 5 The present invention further provides a high-quality metagenome reconstruction device based on the high-quality metagenome reconstruction method of Example 1, and the device comprises:

[0134] A data response module: configured to respond to a first reconstruction instruction of a high-quality microbial genome corresponding to a first sample, and obtain sample boxes corresponding to each contig of the first sample, wherein any contig of the first sample includes a plurality of sample boxes corresponding one-to-one to each binning tool;

[0135] Data integration module: used for removing redundancy and quality screening of all the sample boxes corresponding to each overlapping group by using a sample box fusion strategy to obtain a high-quality first sample box set;

[0136] A data reassembly module: configured to select at least one sample from a plurality of target samples in response to a second reconstruction instruction for reassembling the first sample, extract a sequencing sequence for reassembling the first sample, and reassemble the sample to obtain the target contig;

[0137] The first sample is one of the multiple target samples, and the target contiguous group is the contiguous group of the genome corresponding to the first sample.

[0138] In one embodiment, initial binning of contigs includes:

[0139] A first instruction unit: configured to obtain, in response to the first reconstruction instruction, each of the contigs and a plurality of binning tools corresponding to the sequencing data of the first sample;

[0140] The first binning unit is used to perform binning processing on each of the overlapping groups using the binning tools to obtain a plurality of sample boxes corresponding to each overlapping group of the first sample.

[0141] In one embodiment, the correspondence between overlapping groups and sample boxes includes:

[0142] A first parameter unit: configured to obtain target sequencing data, a first naming rule, and a second naming rule of the first sample in response to the first reconstruction instruction;

[0143] A first processing unit is used to name each contig file corresponding to the target sequencing data using the first naming rule to obtain each contig;

[0144] A second processing unit is used for naming the sample box files corresponding to each of the overlapping groups by using the second naming rule to obtain each sample box;

[0145] The first integration unit is used to establish a file name relationship table between each overlapping group and each sample box according to the first naming rule and the second naming rule.

[0146] In one embodiment, generating the first sample box set by using a first fusion strategy of sample box fusion corresponding to DSL TOOL includes:

[0147] The first data unit: a file name relationship table for obtaining the sample boxes corresponding to the overlapping groups and each binning tool;

[0148] The third processing unit is used for performing redundancy removal on the multiple sample boxes corresponding to each overlapping group by using DSL TOOL according to the file name relationship table to obtain a high-quality first candidate sample box set corresponding to each overlapping group one by one;

[0149] The second integration unit is used to calculate the quality of each sample box in the first candidate sample box set, and screen out the first sample box set with high quality.

[0150] In one embodiment, the second fusion strategy corresponding to the sample box fusion of the primary binning results of multiple binning tools is performed based on sequence characteristics to generate the first sample box set, including:

[0151] The second data unit is used to arrange and combine multiple binning tools to generate multiple combined bins, and obtain sample bin information corresponding to each overlapping group in the primary binning result according to the file name relationship table;

[0152] A fourth processing unit is used for re-binning each overlapping group according to the sample box information and the intersection principle of the primary binning result to obtain a second candidate sample box set composed of each combination box;

[0153] A third integration unit is used to calculate the quality of each sample box in the second candidate sample box set, select a combined box that meets the requirements to represent a corresponding overlapping group, and obtain a third candidate sample box set;

[0154] Information updating unit: used for updating the file name relationship table corresponding to the third candidate sample box set to obtain the first sample box set;

[0155] The calculating of the quality of each sample box in the third candidate sample box set includes: performing quality assessment on a plurality of sample boxes generated by different binning tools corresponding to each overlapping group, and selecting a sample box that meets the requirements to represent a corresponding overlapping group.

[0156] In one embodiment, the DSL TOOL generates the first sample box set by performing a third fusion strategy corresponding to the sample box fusion based on the permutation and combination results based on sequence characteristics, including:

[0157] A data acquisition unit: used for acquiring a fourth candidate sample box set and a fifth candidate sample box set consisting of a plurality of non-redundant sample boxes corresponding to each overlapping group;

[0158] A data comparison unit: used for generating a plurality of similar clusters according to the similarity between the sample boxes in the fourth candidate sample set and the fifth candidate sample set;

[0159] A data selection unit: used for selecting the representation of sample boxes for each of the similar clusters to obtain a sixth candidate sample box set consisting of the representation sample boxes of each of the similar clusters;

[0160] A data processing unit: configured to remove duplicates according to the similarity of each sample box in the sixth candidate sample box set to obtain the first sample box set;

[0161] The fourth candidate sample box set is a non-redundant sample box set generated by fusing sample boxes corresponding to DSL TOOL, and the fifth candidate sample box set is a non-redundant sample box set generated by fusing sample boxes through permutations and combinations based on gene characteristics.

[0162] In one embodiment, obtaining a target contig includes:

[0163] A second instruction unit: configured to obtain the first sample box set corresponding to all target samples in response to the second reconstruction instruction;

[0164] A second screening unit is used to perform quality screening on the sample boxes in the first sample box set to obtain target sample boxes and assembly schemes for reconstruction;

[0165] Genome reconstruction unit: used for selecting sequencing data for reconstruction from at least one sample among all the target samples according to the assembly scheme to assemble the genome of the first sample to obtain the target overlapping group.

[0166] The high-quality metagenome reconstruction device provided by the present invention comprises that each overlapping group of the first sample is binned by multiple binning tools, and then the sample boxes after each overlapping group of the first sample processed by the multiple binning tools are de-redundant and quality screened to obtain a high-quality first sample box group, and then the first sample genome is reassembled based on the overlapping groups of the first sample and / or the sequencing sequences of similar other samples, so as to obtain a genome of the first sample with high quality and high integrity; the present invention obtains high-quality sample boxes by binning and fusing the overlapping groups of a single sample by using multiple binning tools, and then uses the high-quality sequencing data of a single sample or similar multiple samples to complete the recombination of the target single sample genome, so as to obtain a genome with high quality and high integrity.

[0167] Example 3

[0168] The present invention provides an electronic device and a storage medium, see Figure 6 , comprising at least one processor, at least one memory and computer program instructions stored in the memory.

[0169] Specifically, the above-mentioned processor may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of an embodiment of the present invention, and the electronic device includes at least one of the following: a computer, a mobile terminal, a PC, a tablet computer, etc.

[0170] The memory may include a large capacity memory for data or instructions. By way of example and not limitation, the memory may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. Where appropriate, the memory may include a removable or non-removable (or fixed) medium. Where appropriate, the memory may be inside or outside the data processing device. In a particular embodiment, the memory is a non-volatile solid-state memory.

[0171] In one example, the electronic device may further include a communication interface and a bus, wherein the processor, the memory, and the communication interface are connected via the bus and communicate with each other.

[0172] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for reconstructing a high-quality metagenome, characterized in that: The method comprises: In response to a first reconstruction instruction of a high-quality microbial genome corresponding to a first sample, obtaining sample boxes corresponding to each contig of the first sample, wherein any contig of the first sample includes a plurality of sample boxes corresponding one-to-one to each binning tool; Using a sample box fusion strategy to perform redundancy removal and quality screening on all the sample boxes corresponding to each overlapping group, to obtain a high-quality first sample box set; In response to the second reconstruction instruction for reassembling the first sample, selecting at least one sample from a plurality of target samples, extracting a sequencing sequence for reassembling the first sample, and reassembling the sample to obtain the target contig; The first sample is one of the multiple target samples, and the target contiguous group is the contiguous group of the genome corresponding to the first sample.

2. The method for reconstructing a high-quality metagenome according to claim 1, characterized in that: The initial binning of contigs includes: In response to the first reconstruction instruction, obtaining each of the contigs and a plurality of binning tools corresponding to the sequencing data of the first sample; Each of the overlapping groups is binned using the binning tools to obtain a plurality of sample boxes corresponding to each overlapping group of the first sample.

3. The method for reconstructing a high-quality metagenome according to claim 1, characterized in that: The correspondence between overlapping groups and sample boxes includes: In response to the first reconstruction instruction, acquiring target sequencing data of the first sample, a first naming rule, and a second naming rule; Using the first naming rule to name each contig file corresponding to the target sequencing data, to obtain each contig; Using the second naming rule to name the sample box files corresponding to each of the overlapping groups, to obtain each sample box; According to the first naming rule and the second naming rule, a file name relationship table between each overlapping group and each sample box is established.

4. The method for reconstructing a high-quality metagenome according to claim 3, characterized in that: Generating the first sample box set by using the first fusion strategy of sample box fusion corresponding to DSL TOOL includes: Obtain the file name relationship table of sample boxes corresponding to overlapping groups and each binning tool; De-redundancy processing is performed on multiple sample boxes corresponding to each overlapping group using DSL TOOL according to the file name relationship table to obtain a high-quality first candidate sample box set; The quality of each sample box in the first candidate sample box set is calculated to screen out the first sample box set with high quality.

5. The method for reconstructing a high-quality metagenome according to claim 3, characterized in that: The second fusion strategy corresponding to the sample box fusion of the primary binning results of multiple binning tools is performed based on sequence characteristics to generate the first sample box set, including: Arrange and combine multiple binning tools to generate multiple combined bins, and obtain sample bin information corresponding to each overlapping group in the primary binning result according to the file name relationship table; Re-binning each overlapping group according to the sample box information and the intersection principle of the primary binning results to obtain a second candidate sample box set consisting of each of the combined boxes; Calculating the quality of each sample box in the second candidate sample box set, selecting a combined box that meets the requirements to represent a corresponding overlapping group, and obtaining a third candidate sample box set; The file name relationship table corresponding to the third candidate sample box set is updated to obtain the first sample box set.

6. The method for reconstructing a high-quality metagenome according to claim 3, characterized in that: DSL TOOL combines the results of permutations and combinations based on sequence characteristics to generate the first sample box set using a third fusion strategy corresponding to sample box fusion, including: Obtaining a fourth candidate sample box set and a fifth candidate sample box set consisting of a plurality of non-redundant sample boxes corresponding to each overlapping group; Generating a plurality of similar clusters according to the similarities between the sample boxes in the fourth candidate sample set and the fifth candidate sample set; Selecting a representation of a sample box for each of the similar clusters to obtain a sixth candidate sample box set consisting of the representation sample boxes of each of the similar clusters; Deduplication is performed according to the similarity of each sample box in the sixth candidate sample box set to obtain the first sample box set; Among them, the fourth candidate sample box set is a non-redundant sample box set generated by merging sample boxes corresponding to DSL TOOL, and the fifth candidate sample box set is a non-redundant sample box set generated by arranging and combining the primary binning results of each overlapping group based on sequence characteristics according to multiple binning tools.

7. The method for reconstructing a high-quality metagenome according to any one of claims 1 to 6, characterized in that: Obtaining target contigs includes: In response to the second reconstruction instruction, obtaining the first sample box sets corresponding to all target samples; Performing quality screening on a sample box set consisting of the first sample box set of all target samples to obtain a target sample box and an assembly scheme for reconstruction; According to the assembly scheme, selecting sequencing data for reconstruction from a reference genome corresponding to at least one sample among all the target samples to assemble the genome of the first sample to obtain the target contig; The reference genome of each target sample contains all overlapping nucleic acid sequences of the target sample box.

8. A high-quality metagenome reconstruction device, characterized in that: The device comprises: A data response module: configured to respond to a first reconstruction instruction of a high-quality microbial genome corresponding to a first sample, and obtain sample boxes corresponding to each contig of the first sample, wherein any contig of the first sample includes a plurality of sample boxes corresponding one-to-one to each binning tool; Data integration module: used for removing redundancy and quality screening of all the sample boxes corresponding to each overlapping group by using a sample box fusion strategy to obtain a high-quality first sample box set; A data reassembly module: configured to select at least one sample from a plurality of target samples in response to a second reconstruction instruction for reassembling the first sample, extract a sequencing sequence for reassembling the first sample, and reassemble the sample to obtain the target contig; The first sample is one of the multiple target samples, and the target contiguous group is the contiguous group of the genome corresponding to the first sample.

9. An electronic device, characterized in that: At least one processor, at least one memory and computer program instructions stored in the memory, when the computer program instructions are executed by the processor, implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images