An analysis method, system and device based on single-cell whole transcriptome sequencing data

Abstract

The embodiment of the present application discloses an analysis method, system, and device based on single-cell whole transcriptome sequencing data, which includes the following steps: lncRNA analysis: select lncRNA annotation comparison reference, assemble reads aligned to the genome into transcripts, analyze based on transcripts, discover new lncRNA molecules and determine the basic characteristics of lncRNA; single-cell whole transcriptome quantitative analysis: compare reads with the reference genome, annotate reads as specific genes; then correct and count the UMI to obtain an unfiltered feature-barcode matrix; subgroup up-regulated expression gene analysis: each subgroup screens specifically expressed gene markers, uses Seurat's rank sum test to perform differential gene expression analysis on different cell subgroups, and screens genes up-regulated in each cell subgroup. This application solves the problem that the analysis of single-cell whole transcriptome sequencing data, i.e., scFAST-seq data, in the prior art is not comprehensive and specific enough and the scope of involvement is not wide enough.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of bioinformatics, and specifically to an analysis method, system, and device based on single-cell whole transcriptome sequencing data. Background Art

[0002] High-throughput sequencing is one of the most widely used technologies in the field of basic research in life sciences today. It can help us reveal the genome composition of organisms and the relationship between genes, and promote the further development and application of genomics. With the continuous development of single-cell sequencing technology, most single-cell transcriptome sequencing technologies can only capture mRNA with a single-end polyA tail, and the data obtained have certain limitations. Single-cell full-length RNA Sequence (scFAST-seq) is a new technology for high-throughput sequencing at the single-cell level. Its technical principle is to use microfluidic digital droplets and Barcoded Beads coupled with random primers to achieve single-cell separation and random capture of the whole transcriptome in one step of oil-in-water. Single-cell whole transcriptome sequencing technology can capture full-length sequences of non-coding RNA, namely lncRNA and circRNA, at the single-cell level, so it can realize the study of the whole transcriptome at the single-cell level.

[0003] By combining random primers with high reverse transcription efficiency and efficient rRNA removal methods, random capture of the full-length sequence coverage of the single-cell transcriptome can be achieved, so that lncRNA without ploy A tails and circRNA captured by single-cell transcriptome technology can be captured at the single-cell level. This technology can make sequence research more refined, upstream and downstream regulatory mechanisms more complete, and cell pseudo-differentiation pathways more restored. There are still some shortcomings in the existing technology, which are affected by insufficient sequencing coverage, insufficient sequencing depth, small single-cell sample size, and factors such as reverse transcriptase, primer specificity, and amplification preference. In particular, the massive data processing and analysis process is relatively complex, and the accuracy of data analysis is limited. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the embodiments of the present application is to provide an analysis method, system, and device based on single-cell whole transcriptome sequencing data, through lncRNA analysis and circRNA analysis, single-cell whole transcriptome gene expression quantification, and finally subpopulation up-regulated expression gene analysis, to solve the problems in the prior art that the analysis of single-cell whole transcriptome sequencing data, i.e., scFAST-seq data is not comprehensive and specific enough and the scope of coverage is not wide enough.

[0005] In order to solve the above technical problems, the technical solutions adopted in the embodiments of the present application are as follows:

[0006] The present invention provides an analysis method based on single-cell whole transcriptome sequencing data, which comprises the following steps:

[0007] lncRNA analysis: select lncRNA annotation and alignment reference, assemble reads aligned to the genome into transcripts, analyze based on transcripts, discover new lncRNA molecules and determine the basic characteristics of lncRNA;

[0008] Quantitative analysis of the whole transcriptome of a single cell: reads are compared with the reference genome and reads are annotated as specific genes. After UMI correction and statistics, an unfiltered feature-barcode matrix is ​​obtained.

[0009] Analysis of upregulated genes in subpopulations: Gene markers with specific expression were screened for each subpopulation, and differential gene expression analysis was performed on different cell subpopulations using the Seurat rank-sum test to screen upregulated genes in each cell subpopulation.

[0010] As a further preferred solution, in some embodiments of the present application, the lncRNA analysis comprises the following steps:

[0011] lncRNA molecular annotation: The genome or known transcriptome data of the species corresponding to the sample is used as a reference for annotation comparison, and HISAT2 software is used for comparison based on the reference genome;

[0012] Prediction and identification of new lncRNAs: Reconstruct transcripts and use CPC2, CNCI, and feelnc software to predict the coding capacity of new lncRNAs. The intersection of the three softwares is used as the final lncRNA prediction result, and the prediction results are identified;

[0013] Classification of lncRNA transcripts: According to the location of new lncRNAs on the genome relative to protein-coding genes, new lncRNAs can be divided into intergenic long non-coding lncRNAs, bidirectional long non-coding lncRNAs, intronic long non-coding RNAs, antisense long non-coding RNAs, and sense long non-coding RNAs;

[0014] lncRNA expression analysis: SeekSoul Tools was used to quantify the expression of reads annotated as lncRNA by correcting the UMI to determine the number of lncRNAs detected in the sample.

[0015] As a further preferred solution, in some embodiments of the present application, the novel lncRNA prediction comprises the following steps:

[0016] stringtie reconstructed transcripts: reconstructed transcripts based on stringTie transcriptome analysis software;

[0017] Identification of lncRNA prediction results: transcripts with length ≥ 200 bp and exon number ≥ 2 were retained, and the coding ability of new transcripts was predicted using CPC2, CNCI, and feelnc software, and the intersection of transcripts without coding potential was taken as a reliable prediction result.

[0018] As a further preferred scheme, in some embodiments of the present application, the analysis method based on single-cell whole transcriptome sequencing data also includes circRNA analysis and statistics: after aligning the Reads to the reference genome, circRNA identification is performed, and the circRNA identification results are filtered to obtain highly reliable circRNAs, and the identified circRNAs are aligned to the circBank database through local alignment software, and the remaining species are aligned to the circBase database, and highly reliable annotated circRNAs are found and defined as existing circRNAs, otherwise they are newly predicted circRNAs.

[0019] As a further preferred solution, find_circ is used in some embodiments of the present application to identify circRNA, and the local software is blast software, and the filtering condition set for finding highly credible annotated circRNA is E value <e -10 .

[0020] As a further preferred solution, in some embodiments of the present application, in the subpopulation up-regulated expression gene analysis step, the screening conditions for up-regulated genes are:

[0021] (1) In the target subpopulation or control subpopulation, the gene is expressed in more than 25% of the cells;

[0022] (2) gene expression fold log2FC ≥ 0.36, i.e., gene up-regulation fold ≥ 1.28;

[0023] (3)P value ≤ 0.01.

[0024] The present application also provides an analysis system based on single-cell whole transcriptome sequencing data, including:

[0025] lncRNA analysis unit, used to perform lncRNA analysis functions, discover new lncRNA molecules and determine the basic characteristics of lncRNA;

[0026] Single-cell whole transcriptome quantitative analysis unit, used to perform single-cell whole transcriptome quantitative analysis function;

[0027] The subpopulation up-regulated expression gene analysis unit is used to perform the subpopulation up-regulated expression gene analysis function and screen the up-regulated expression genes of each cell subpopulation.

[0028] As a further preferred solution, in some embodiments of the present application, the lncRNA analysis unit comprises:

[0029] lncRNA molecule annotation subunit, used for comparison based on reference genome;

[0030] The new lncRNA prediction and identification subunit is used to predict the coding capacity of new transcripts and identify the prediction results;

[0031] The lncRNA transcript classification subunit is used to classify novel lncRNAs into different types of lncRNAs according to their positions on the genome relative to protein-coding genes;

[0032] The lncRNA expression analysis subunit is used to perform statistical expression quantification on the corrected UMI and determine the number of lncRNAs detected in the sample.

[0033] As a further preferred solution, in some embodiments of the present application, the analysis system based on single-cell whole transcriptome sequencing data also includes a circRNA analysis and statistical unit for finding highly reliable annotated circRNAs to distinguish known circRNAs from newly predicted circRNAs.

[0034] An embodiment of the present application also provides a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the analysis method based on single-cell whole transcriptome sequencing data described in an embodiment of the present application.

[0035] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0036] 1. The analysis method based on single-cell whole transcriptome sequencing data described in the embodiments of the present application can effectively identify and assemble new lncRNA transcripts from sequencing data by selecting lncRNA annotation comparison references, and then determine their basic characteristics, which helps to expand the research scope of lncRNA and provide new clues for understanding its role in biological processes.

[0037] 2. In the analysis method based on single-cell whole transcriptome sequencing data described in the embodiment of the present application, reads are compared with the reference genome and annotated as specific genes. Combined with the correction and statistics of UMI, this method can generate a high-quality unfiltered feature-barcode matrix, providing accurate quantification of gene expression at the single-cell level, laying a solid foundation for subsequent analysis.

[0038] 3. The analysis method based on single-cell whole transcriptome sequencing data described in the embodiment of the present application can systematically analyze the differential gene expression between different cell subpopulations by adopting the Seurat rank sum test, and screen out the genes upregulated in each cell subpopulation, which helps to reveal the functional differences and potential biological mechanisms between cell subpopulations.

[0039] 4. The analysis method based on single-cell whole transcriptome sequencing data described in the embodiments of the present application combines lncRNA analysis, circRNA analysis, single-cell whole transcriptome quantitative analysis and subpopulation up-regulated expression gene analysis, providing a comprehensive ability to understand cell status and function from multiple levels. This comprehensive analysis method helps to reveal complex biological phenomena and disease mechanisms.

[0040] 5. The analysis method based on single-cell whole transcriptome sequencing data described in the embodiments of the present application has high sensitivity and accuracy. It is applicable to a variety of biological and medical research fields, such as disease occurrence and development mechanisms, cell differentiation processes, drug response mechanisms, etc., and provides new tools and perspectives for research in related fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a Venn diagram of CPC2, CNCI, and Feelnc in the prediction and identification of new lncRNAs described in the examples of this application.

[0042] Figure 2 This is a statistical diagram of the classification of lncRNA transcripts described in the examples of this application.

[0043] Figure 3 This is a distribution diagram of the number of circRNA chromosomes described in the examples of this application.

[0044] Figure 4 This is a statistical distribution diagram of the circRNA chromosome described in the examples of this application. DETAILED DESCRIPTION

[0045] This specific implementation manner is merely an explanation of the embodiments of the present application, and it is not a limitation of the embodiments of the present application. After reading this specification, those skilled in the art can make modifications to the embodiments without creative contribution as needed, but as long as they are within the scope of the claims of the embodiments of the present application, they are protected by the patent law.

[0046] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the embodiments of the present application.

[0047] The term "comprise" and any variations thereof in the specification and claims of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0048] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0049] Example 1

[0050] The present invention provides an analysis method based on single-cell whole transcriptome sequencing data, which comprises the following steps:

[0051] lncRNA analysis: Select lncRNA annotation and alignment reference, assemble reads aligned to the genome into transcripts, perform analysis based on transcripts, discover new lncRNA molecules and determine the basic characteristics of lncRNA. The specific steps are as follows:

[0052] 1) lncRNA molecular annotation: Using the genome or known transcriptome data of the species corresponding to the sample (e.g., full-length transcriptome spliced ​​by third-generation technology) as the annotation comparison reference, the HISAT2 software global and local search methods can effectively compare to the spliced ​​reads in the RNA-Seq sequencing data, and perform comparison based on the reference genome;

[0053] 2) Prediction and identification of new lncRNAs: Since the sources of lncRNAs are very complex, different transcripts of the same gene produce lncRNAs with great differences. Therefore, lncRNAs will be analyzed based on transcripts. The coding capacity of new transcripts is predicted by using CPC2, CNCI, and feelnc software. The intersection of the three softwares is used to judge the final lncRNA prediction results, and the prediction results are identified. The identification process is based on the reconstruction of transcripts by stringtie transcriptome analysis software, retaining transcripts with a length of ≥200bp and an exon number of ≥2 (plant samples ≥1). The statistics of the number of stringtie reconstructed transcripts are shown in Table 1 below (filtered transcripts with a length of ≥200bp and an exon number of ≥2 (plant samples ≥1)):

[0054] Table 1: Statistics of the number of stringtie-reconstructed transcripts

[0055] Number of genes Number of transcripts 2373 2791

[0056] The intersection of transcripts without coding potential is taken as the reliable prediction result. Figure 1 shown.

[0057] The new lncRNA prediction results are shown in Table 2.

[0058] Table 2: New lncRNA prediction results

[0059]

[0060] 3) Classification of lncRNA transcripts: According to the location of novel lncRNAs relative to protein-coding genes on the genome, novel lncRNAs can be divided into intergenic long noncoding lncRNAs, bidirectional long noncoding lncRNAs, intronic long noncoding RNAs, antisense long noncoding RNAs, and sense long noncoding RNAs; the results are shown in Tables 3 and Figure 2 :

[0061] Table 3: Statistics of lncRNA types

[0062]

[0063] lncRNA expression analysis: SeekSoul Tools was used to quantify the expression of reads annotated as lncRNA by correcting the UMI, and the number of lncRNA detected in the samples was determined. The statistical results of lncRNA expression in the samples are shown in Table 4, the statistical results of the number of lncRNA detected in each group are shown in Table 5, and the statistical results of the number of lncRNA detected in each subgroup are shown in Table 6.

[0064] Table 4: Statistical results of lncRNA quantity in the detected samples

[0065]

[0066] Table 5: Statistical results of the number of lncRNAs detected in each group

[0067]

[0068] Table 6: Statistical results of lncRNA quantity detected in each subgroup

[0069]

[0070] Note: sample is the sample (grouping, subgroup) name, and all is the result of all merging;

[0071] Refer_lncs: the total number of lncRNAs in the reference genome (or reference gene set);

[0072] sequenced_Refer_lncs(%): the total number of Refer_lncs detected by sequencing results and the percentage of Refer_lncs;

[0073] The number of novel lncRNAs discovered by the Novel_lncs project;

[0074] sequenced_Novel_lncs(%): the total number of Novel_lncs detected by sequencing results and the percentage of Novel_lncs;

[0075] Total_lncs: the total number of all lncRNAs, including lncRNAs known from the reference genome and newly identified lncRNAs;

[0076] sequenced_Total_lncs(%): The total number of lncRNAs detected by sequencing results and their percentage in Total_lncs.

[0077] Quantitative analysis of the whole transcriptome of a single cell: Use SeekSoul Tools to align reads with the reference genome and annotate reads as specific genes. After correcting and counting the UMIs, an unfiltered feature-barcode matrix is ​​obtained. The number of gene types detected in each sample is shown in Table 7.

[0078] Table 7: Statistical results of the number of gene types detected in each sample

[0079]

[0080] Note: Sample name;

[0081] Total protein coding genes Detected The total number of protein coding genes detected in all cells;

[0082] Total lncRNA genes Detected: The total number of lncRNA genes detected in all cells;

[0083] Total circRNA genes Detected: the total number of circRNAs detected in all cells;

[0084] Median protein coding genes per Cell: The median number of protein coding genes detected in each cell;

[0085] Median lncRNA genes per Cell: the median number of lncRNA genes detected in each cell;

[0086] Median circRNA genes per Cell The median number of circRNAs detected in each cell.

[0087] Analysis of upregulated genes in subpopulations: Gene markers with specific expression were screened for each subpopulation, and differential gene expression analysis was performed on different cell subpopulations using the Seurat rank sum test to screen upregulated genes in each cell subpopulation; the screening conditions for upregulated genes were as follows:

[0088] (1) In the target subpopulation or control subpopulation, the gene is expressed in more than 25% of the cells;

[0089] (2) gene expression fold log2FC ≥ 0.36, i.e., gene up-regulation fold ≥ 1.28;

[0090] (3)P value ≤ 0.01.

[0091] Because scFAST-seq can detect both protein-coding genes and non-protein-coding genes, for ease of viewing, the statistical results of differentially expressed genes are classified and displayed according to protein coding genes and lncRNAs, as shown in Tables 8, 9, and 10.

[0092] Table 8: Statistics of the number of upregulated protein coding genes in each subgroup

[0093] Single cell subsets 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Number of up-regulated genes 84 173 153 230 320 335 250 137 268 77 222 100 81 78 68 154 70 165

[0094] Table 9: Statistics of the number of upregulated LncRNA genes in each subgroup

[0095] Single cell subsets 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Number of up-regulated genes 1 7 3 6 2 6 4 1 0 1 2 1 1 0 1 8 1 8

[0096] Table 10: Annotation table of differentially expressed genes in each subgroup

[0097]

[0098]

[0099] Note: Target Cluster, we will screen the upregulated genes in the target cluster;

[0100] Gene ID The ID number corresponding to the upregulated gene in the target subgroup;

[0101] Gene name: gene name, usually symbol number;

[0102] Gene Type, including protein coding, LncRNA and circRNA;

[0103] Target_Cluster_mean is the average expression of upregulated genes in the target cell subpopulation; Other_Cluster_mean is the average expression of upregulated genes in other cell subpopulations; Log2FC is the ratio of the average expression of genes in the target cell subpopulation to that in other cell subpopulations, in log2; Pvalue is the p-value of the difference test;

[0104] Qvalue difference test q-value;

[0105] Description: Description of the gene;

[0106] KEGG_A_class KEGG A-level classification name;

[0107] KEGG_B_class KEGG B-level classification name;

[0108] Pathway KEGG Pathway annotation (level C classification);

[0109] K_ID gene corresponds to KEGG protein annotation ID;

[0110] GO Component GO functional annotation_cellular component;

[0111] GO Function GO functional annotation_molecular function;

[0112] GO Process GO functional annotation_biological process.

[0113] Example 2

[0114] On the basis of the scheme of Example 1 above, in this example, the analysis method based on single-cell whole transcriptome sequencing data further includes circRNA identification and statistics: after the reads are aligned to the reference genome, find_circ is used to identify circular RNA, and the circular RNA identification results are filtered to obtain highly reliable circular RNA. The circular RNA identification results are shown in Tables 11, Figure 3 and Figure 4 .

[0115] Table 11: circRNA identification results

[0116]

[0117]

[0118] Note: Gene ID (gene_id) circular RNA ID;

[0119] circBank_ID: the ID of the circular RNA in the circBank database (human project);

[0120] circbase_ID is the ID of the circular RNA in the circBase database;

[0121] Annotation (annotation): the type of circular RNA included in circBase;

[0122] Best Transcript (best transcript): NCBI number of the circular RNA source gene;

[0123] Gene Symbol (gene symbol) The name of the gene from which the circular RNA originates.

[0124] To obtain more annotation information of circRNA, the identified circRNA was aligned to the circBank database using the local alignment software blast, and the remaining species were aligned to the circBase database. <e -10 As filtering conditions, highly credible annotated circRNAs were found and defined as existing circRNAs, otherwise they were newly predicted circRNAs; during the analysis, the statistics of circRNA annotations are shown in Table 11, the statistics of circRNA numbers in samples are shown in Table 12, the statistics of circRNA numbers in each group are shown in Table 13, and the statistics of circRNA numbers in each subgroup are shown in Table 14.

[0125] Table 12: Statistics of circRNA annotation

[0126] total_circRNA (number of all circRNAs) exist_circRNA (number of existing circRNAs) novel_circRNA (number of newly annotated circRNAs) 869 109 760

[0127] Table 13: CircRNA quantity in samples

[0128]

[0129] Table 14: CircRNA quantity in each group

[0130]

[0131] Table 15: CircRNA quantity in each subgroup

[0132]

[0133]

[0134] In this manual, some specific implementation details or methods are described in detail. However, this does not mean that they are necessary or unique. Within the scope of the present invention, any similar replacement, modification or variation is possible, as long as they do not deviate from the spirit and scope of the present invention. In addition, although some embodiments are described in detail in this manual, these descriptions should not limit the scope of the present invention, because many changes and modifications can be made within the scope of the present invention.

Claims

1. An analysis method based on single-cell whole transcriptome sequencing data, characterized in that: It includes the following steps: lncRNA analysis: select lncRNA annotation and alignment reference, assemble reads aligned to the genome into transcripts, analyze based on transcripts, discover new lncRNA molecules and determine the basic characteristics of lncRNA; specifically include 1) lncRNA molecular annotation: The genome or known transcriptome data of the species corresponding to the sample is used as the annotation comparison reference, and the HISAT2 software is used for comparison based on the reference genome; 2) Prediction and identification of new lncRNAs: Reconstruct transcripts based on stringtie transcriptome analysis software; keep transcript length ≥ 200bp and exon number ≥ 2, use CPC2, CNCI, and feelnc software to predict the coding ability of new transcripts, and use CPC2, CNCI, and feelnc software to predict the coding ability of new lncRNAs, and take the intersection of transcripts without coding potential as a reliable prediction result; use the intersection of the three software as the final lncRNA prediction result, and identify the prediction result; 3) Classification of lncRNA transcripts: According to the location of novel lncRNAs relative to protein-coding genes on the genome, novel lncRNAs can be divided into intergenic long noncoding lncRNAs, bidirectional long noncoding lncRNAs, intronic long noncoding RNAs, antisense long noncoding RNAs, and sense long noncoding RNAs; 4) lncRNA expression analysis: Use SeekSoul Tools to quantify the expression of reads annotated as lncRNA by correcting the UMI to determine the number of lncRNA detected in the sample; Quantitative analysis of the whole transcriptome of a single cell: reads are compared with the reference genome and reads are annotated as specific genes. After UMI correction and statistics, an unfiltered feature-barcode matrix is ​​obtained. Analysis of upregulated genes in subpopulations: Gene markers with specific expression were screened for each subpopulation, and differential gene expression analysis was performed on different cell subpopulations using the Seurat rank-sum test to screen upregulated genes in each cell subpopulation.

2. The analysis method based on single-cell whole transcriptome sequencing data according to claim 1, characterized in that: It also includes circ RNA analysis and statistics: after aligning the Reads to the reference genome, circRNA identification is performed, and the circRNA identification results are filtered to obtain highly reliable circRNAs. The identified circRNAs are aligned to the circBank database through local alignment software, and the remaining species are aligned to the circBase database. Highly reliable annotated circRNAs are found and defined as existing circRNAs, otherwise they are newly predicted circRNAs.

3. The analysis method based on single-cell whole transcriptome sequencing data according to claim 2, characterized in that: Use find_circ to identify circRNA. The local software is blast software. The filtering condition set for finding highly credible annotated circRNA is E value. <e -10 .

4. The analysis method based on single-cell whole transcriptome sequencing data according to any one of claims 1 to 3, characterized in that: In the step of analyzing up-regulated genes in subgroups, the screening conditions for up-regulated genes are: (1) In the target subpopulation or control subpopulation, the gene is expressed in more than 25% of the cells; (2) gene expression fold log2FC ≥ 0.36, i.e., gene up-regulation fold ≥ 1.28; (3)P value ≤ 0.

01.

5. An analysis system based on single-cell whole transcriptome sequencing data, characterized in that: include: lncRNA analysis unit, used to perform lncRNA analysis functions, discover new lncRNA molecules and determine the basic characteristics of lncRNA; include 1) The lncRNA molecular annotation subunit uses the genome or known transcriptome data of the species corresponding to the sample as the annotation comparison reference, and uses HISAT2 software to perform comparison based on the reference genome; 2) New lncRNA prediction and identification subunit, reconstruct transcripts based on stringtie transcriptome analysis software; keep transcript length ≥ 200bp and exon number ≥ 2, use CPC2, CNCI, feelnc three software to predict the coding ability of new transcripts, use CPC2, CNCI, feelnc software to predict the coding ability of new lncRNA, take the intersection of transcripts without coding potential as a reliable prediction result; use the intersection of the three software as the final lncRNA prediction result, and identify the prediction result; 3) lncRNA transcript classification subunits: According to the location of novel lncRNAs relative to protein-coding genes on the genome, novel lncRNAs can be divided into intergenic long noncoding lncRNAs, bidirectional long noncoding lncRNAs, intronic long noncoding RNAs, antisense long noncoding RNAs, and sense long noncoding RNAs; 4) lncRNA expression analysis subunit, using SeekSoul Tools to quantify the expression of reads annotated as lncRNA by correcting the UMI, and determining the number of lncRNA detected in the sample; Single-cell whole transcriptome quantitative analysis unit, used to perform single-cell whole transcriptome quantitative analysis function; The subpopulation up-regulated expression gene analysis unit is used to perform the subpopulation up-regulated expression gene analysis function and screen the up-regulated expression genes of each cell subpopulation.

6. The analysis system based on single-cell whole transcriptome sequencing data according to claim 5, characterized in that: It also includes a circRNA analysis and statistics unit, which is used to find highly confident annotated circRNAs to distinguish known circRNAs from newly predicted circRNAs.

7. A computer-readable storage medium, characterized in that: It stores a computer program, wherein the computer program enables a computer to execute the analysis method based on single-cell whole transcriptome sequencing data according to any one of claims 1 to 4.

Images