KR101884073B1 - micro-RNA ID analysis method through data mining of micro RNA - Google Patents

Abstract

The present invention relates to a method for analyzing miRNA IDs through data mining of miRNAs. In order to improve the analysis capability of miRNAs, accurate miRNAs are analyzed through comparative analysis of two or more databases, and miRNAs capable of predicting the expression of cancer cells Lt; RTI ID = 0.0 > miRNA < / RTI >
Also, the method of analyzing miRNA ID through data mining of miRNA according to the present invention can provide a therapeutic agent and a biomarker for breast cancer.

Description

(Micro-RNA ID analysis method through data mining of micro RNA)

The present invention relates to a method for analyzing miRNA IDs through data mining of microRNAs (miRNAs), and more particularly to a method for analyzing miRNA IDs by comparing and analyzing two or more databases, And to a method for analyzing miRNA IDs through data mining of miRNAs capable of predicting the expression of cancer cells.

Further, the present invention relates to miRNA ID biomarkers and therapeutic agents derived using miRNA ID analysis through data mining of miRNAs.

Many disease states are characterized by differences in the expression levels of various genes through changes in the number of genetic DNA copies or changes in transcription levels of particular genes. For example, the loss of genetic material plays an important role in malignant transformation and progression. MiRNAs are known to play the most important role among these genetic materials.

     miRNA was first discovered in 1993 by a team led by Harvard University professor V. Ambros. While searching for a gene that regulates the timing of Caenorhabditis elegans, it has been found that a short RNA fragment named lin-4 affects the synthesis of LIN-14 protein. The addition of an RNA fragment named let-7 led to increased interest in miRNA presence and function. Also, miRNAs are small single-strand nucleotides composed of 21 to 25 nucleotides (nucelotide), which are known to regulate the expression of various genes in eukaryotes. After miRNAs expressed in the C. elegans were discovered by the Ambrosia group in 1993, up to now, more than 700 species have been found in human cells.

     The biosynthesis of miRNA is largely driven by cleavage of two enzymes. First, genes containing miRNAs are transcribed by RNA polymerase II or III (RNA polymerase II / III), and various sizes of miRNA transcripts are synthesized. The primary miRNA (pri-miRNA) synthesized by this process has cap (7-methylguanylate cap) at the 5 'end and poly [A] tail at the 3' end. Pri-miRNA is processed into a precursor miRNA (pre-miRNA) with a length of about 70 nucleotides by a microprocessor complex consisting of the Drosha RNA cleavage enzyme in the nucleus and DGCR8 (DiGeorge critical region 8). Pre-miRNAs are exported to the cytoplasm via exportin5 and Ran-GTP, and then processed into mature double-stranded miRNAs composed of 20-25 nucleotides by the second cleavage enzymes Dicer and TRBP (transactivating response RNA binding protein).

     One of the double strands is broken down and the other strand is bound together with Ago (Argonaute) to form a RNA-induced silencing complex (RISC). TRBP induces the binding of Ago to miRNAs and allows miRNAs to regulate the expression of target genes. In general, miRNA binds to the 3 'untranslated region (3' UTR), which is not translated into a protein, which lowers the stability of the mRNA or lowers translation efficiency to inhibit the expression of the target gene.

     In recent years, miRNAs have been found to be important regulators of gene expression, and 15,172 miRNAs have been found in 32 species including plants and animals. Therefore, in order to process and analyze large amounts of data, Methods have been introduced. The BRCA1 gene, also known as the tumor suppressor gene, is a cancer-suppressing gene located on chromosome 17, which is known to be associated with the development of hereditary breast cancer.

     Hereditary breast cancer can be inherited from both the paternal and maternal lines, because the mutation of the parent's germ cell margins is inherited to its offspring. Recently, codon pattern mining has been used to predict the onset of cancer by detecting certain patterns of genes that cause hereditary breast cancer. In particular, we focused on the codon pattern among the gene patterns, taking into account the property that the same codons designate the same amino acid in all types of organisms studied, from bacteria to humans. One codon pattern designates one amino acid, which is a component of a protein. If a codon pattern shift occurs, the protein will lose its original function, resulting in an unexpected body abnormality. By applying a series of gene mining techniques to extract patterns of certain codons and finding mutated codon patterns, it can be used for early diagnosis of breast cancer.

     Databases for storing and managing miRNA sequences, information, and features have been constructed, including miRBase, ASRP, and miRNA Map. In addition, various algorithms for predicting miRNA gene candidates and target genes of miRNAs and applications for applying them are actively being developed.

     Methods for predicting miRNA gene candidates generally include RNA conformation-based searches, homology searches for miRNAs with similar sequences, and machine-learning approaches using miRNA traits Etc. are used. The search based on the RNA structure is a method of utilizing the physicochemical characteristics of the hairpin structure of the pre-miRNA, and it is a process of predicting the candidate miRNA by calculating whether the specific nucleotide sequence has a thermodynamically stable hairpin structure .

     Prediction by homology search is a method for predicting candidates by probabilistically calculating similarity of base sequences, and is very useful for predicting sequences of evolutionarily conserved miRNAs. Prediction through machine learning method is widely used in bioinformatics research. After repeated training using features such as base sequence, distribution, structural specificity and evolutionary conservativeness of known miRNAs, new sequence It is a method of predicting the result according to the learned information when the information is inputted.

     As a related art related to this, Patent Document 001 of "Patent Document 001" discloses a method of analyzing miRNA ID that can accurately analyze miRNA by improving the analysis capability of miRNA and predict the expression of cancer cells through it, And to provide miRNA biomarkers that regulate genes associated with colorectal cancer by regulating related genes.

     In addition, the following Patent Document 002 describes the detection of multiple miRNAs using a capillary electrophoresis system, but accurate miRNA characteristics can not be obtained by this method. The patent document 003 relates to a method for quantitative analysis of miRNA, but the accuracy with which the unknown miRNA can be identified is inferior. In addition, the following patent document 004 relates to a mi search automation system for miRNA, but it discloses only a method for simply identifying a miRNA using a mapping tool of a reference miRNA database (D / B).

Accordingly, the present inventors have conducted extensive research to solve the problems of the prior art. As a result, the present inventors have found that, as a method of analyzing a new miRNA ID through data mining of miRNA, a comparative analysis of two or more kinds of databases The present inventors completed the present invention by studying miRNA ID analysis method through data mining of miRNA which can accurately predict the expression of cancer cells through the analysis of miRNA.

Korean Patent No. 10-1583450 (2016. 01. 03.) Korean Patent Publication No. 10-2013-0122541 (Mar. 11, 2017) Korean Patent Laid-Open Publication No. 10-2014-0108913 (Apr. 15, 2014) Korean Patent Laid-Open Publication No. 10-2014-0114684 (Apr. 29, 2014)

The present invention relates to an optimized miRNA analysis method using data mining of miRNA. In order to improve the miRNA analysis ability, miRNAs are analyzed by comparing and analyzing two or more kinds of databases and miRNAs capable of predicting the expression of cancer cells To provide an optimized miRNA analysis method through data mining.

The present invention also relates to a therapeutic agent or a biomarker for regulating a cancer-related gene using the result of miRNA ID analysis through data mining of miRNA.

(A) preparing an unknown bio sample; (b) extracting an unknown microRNA (miRNA) from the sample prepared in the first step and sorting the sequenced miRNA into three groups of 5 fold or more, less than 5 folds, 2.5 folds or less, 2.5 folds or less A second step of providing miRNA data according to an expression level; (c) obtaining tumor miRNA data for each level of expression using a data platform from a whole milk map (TCGA); (d) comparing the expression level-specific miRNA data obtained in the second step with the expression level-specific miRNA data obtained in the third step to obtain miRNA results according to a common expression level; (e) removing the group of less than 2.5 folds from the group of less than 5 folds and less than 5 folds according to the difference of the expression level from the common expression level-based miRNA result obtained in the fourth step to obtain the miRNA result according to the regrouping expression level; (f) comparing the miRNA results of the regrouping expression levels obtained through the fifth step with a target prediction database to obtain a common result (I); (g) a seventh step of obtaining the miRNA ID by the sixth step; The present invention also provides a method for analyzing miRNA ID through data mining of miRNA.

     Also, the present invention is characterized in that, between the sixth step and the seventh step, (h) the miRNA results according to the regrouping expression level obtained through the step 5 are classified into three groups of 2.5 or more folds and 2.5 folds or less Obtaining a common result (II) by comparing with the reference database, and comparing the obtained results (I) and (II) to the reference database, thereby providing a method for analyzing miRNA IDs through data mining of miRNAs .

     The present invention further provides a method for analyzing miRNA ID through data mining of miRNA, which further comprises confirming the miRNA ID obtained in the seventh step using a PubMed database.

     In addition, the present invention provides a cancer therapeutic agent that utilizes the result of analysis of miRNA ID obtained by a method of analyzing miRNA ID through data mining of miRNA.

The present invention also provides a cancer biomarker using the result of analysis of miRNA ID obtained by a method of analyzing miRNA ID through data mining of miRNA.

The method of analyzing miRNA ID through data mining of miRNA according to the present invention enhances the efficiency of accurate miRNA analysis by comparing and analyzing two or more kinds of databases in order to improve the analysis ability of miRNA and accurately predicts the expression of cancer cells .

Also, it has an advantage that it can be provided as a miRNA biomarker or therapeutic agent for regulating a gene related to breast cancer using the result of analysis of miRNA ID through data mining of miRNA according to the present invention.

1 is a flow chart illustrating an optimized miRNA analysis method through data mining of miRNA according to the present invention.
Figure 2 is raw data for breast cancer downloaded by TCGA.
Figure 3 shows the results of miRNAs related to BRCA1 and BRCA2.
Figure 4 shows miRNA results related to Cyclin D1.
5 shows the results of miRAN related to N-RAS.
Fig. 6 shows the results of miRNAs related to FGF3, FGF4, HER2, and c-Myc.

Hereinafter, a method of analyzing miRNA ID through data mining of miRNA of the present invention will be described in detail with reference to preferred embodiments.

     The method of analyzing miRNA ID through data mining of miRNA according to the present invention comprises: (a) a first step of preparing an unknown bio sample; (b) extracting an unknown microRNA (miRNA) from the sample prepared in the first step and sorting the sequenced miRNA into three groups of 5 fold or more, less than 5 folds, 2.5 folds or less, 2.5 folds or less A second step of providing miRNA data according to an expression level; (c) obtaining tumor miRNA data for each level of expression using a data platform from a whole milk map (TCGA); (d) comparing the expression level-specific miRNA data obtained in the second step with the expression level-specific miRNA data obtained in the third step to obtain miRNA results according to a common expression level; (e) dividing the expression level of the mRNA by the common expression level obtained in the fourth step into three groups of 5 folds or more, less than 5 folds, 2.5 folds or more and 2.5 folds or less, A fifth step of removing a group less than 2.5 folds to obtain miRNA results according to regrouping expression levels; (f) comparing the miRNA results of the regrouping expression levels obtained through the fifth step with a target prediction database to obtain a common result (I); (g) a seventh step of obtaining the miRNA ID by the sixth step; Is performed.

     Between the sixth and seventh steps, (h) comparing the miRNA results according to the regrouping expression levels obtained in the step (5) to an RNA reference database to obtain a common result (II) And comparing the first and second threshold values with each other.

     The method may further include (i) reconfirming the miRNA ID obtained in the step (7) using a PubMed database.

     First, in the first step, the unknown bio sample may be obtained from fresh or frozen breast cancer tissues, cells, blood, serum or plasma, but is not limited thereto.

     Also, a method of extracting total RNA including an unknown miRNA from the sample of the second step can be performed by various methods known in the art, and can be preferably extracted using tripol or Triton X-100.

     Tumor miRNA data for stage 3 expression levels were obtained from tumor-specific miRNA data from patient cancer cells samples from the open-accessible, accessible Cancer Genome Atlas (TCGA) using the IluminaGA-miRNAseq platform, Level-specific tumor miRNA data contains a variety of information, including miRNA ID, read-count, and read per milion miRNA map.

     The target prediction database in the fifth step can be selected from at least one of miRWalk, MicroT4, miRBridge, miRBridge, miRMap, miRNAMap, PICTAR2, PITA, RNA22, RNAhybrid, Targetscan, But is not limited thereto.

     In (h), at least one of miRBase, ASRP, micro RNAMAP, miRGen, and CoGemiR may be selected and used in the reference database, but the present invention is not limited thereto.

     Further, the method may further comprise confirming the miRNA ID obtained in the seventh step using a PubMed database. The miRNA ID may be re-confirmed by searching for papers published in the pub med and confirming that the miRNA ID is obtained, The representative keyword is miR-, and it is possible to prove, but not limited to, the correlation between the miRNA ID and the miRNA of the target prediction database.

     Further, the oncogene and the tumor suppressor gene can be identified using miRNA ID analysis method using data mining of miRNA obtained by the present invention. Examples of the oncogene gene include HER2, Cyclin D1, N-RAS, FGF3, FGF4, C-Myc And tumor suppressor genes such as BRCA1 and BRCA2. And a comparative analysis can be carried out through a combination of at least one of them.

     BRCA genes include BRCA1 and BRCA2. The BRCA1 gene is located on chromosome 17, and the BRCA2 is located on chromosome 13. The BRCA gene, when damaged by DNA, reacts with other proteins and repairs them. Therefore, if the BRCA gene does not function normally, DNA repair problems will occur and the risk of cancer will increase. In particular, the problem is the gene mutation (Gene Mutation). BRCA gene mutations are associated with a dramatic increase in the risk of breast and ovarian cancer. A gene mutation is a change in DNA sequence that causes a trait that is not present in the parent to appear in the offspring.

     The risk of developing breast cancer and ovarian cancer depends on the BRCA gene mutation. Women with both BRCA1 and BRCA2 mutations have a risk of developing breast cancer of 60 to 85 percent. The risk of developing ovarian cancer increases when an abnormality is found in the BRCA1 gene. The BRCA2 gene mutation is known to be associated with the development of pancreatic cancer and stomach cancer, although its association with ovarian cancer is relatively small.

     Genetic studies on HER2 and cyclin D1 have found two genetic signals required for early breast cancer development by Harvard Medical School researchers. They introduced HER2 or cyclin D1 gene into the cultured breast tissue and thus tracked the effects of the three-dimensional model. Both HER2 and cyclin D1 promoted the proliferation of the breast cells to the depressed sites, and the cyclin D production tissues maintained a depressed location because proliferating cells were prone to death by autologous cell death. In contrast, HER2 provided a survival signal (or anti-death) signal to allow cells to fill the depression. This filled structure resembles the pathology of early breast tumors. Thus, it has been shown that the tumor cells maintain their glandular architecture in the early stage of breast cancer development by overcoming the death of normal cells as well as the proliferation of tumor cells.

     The N-RAS gene is one of the genes identified as a viral cancer gene (v-ras) of rodent retrovirus. Three types of mammalian cell genes c-H-ras1, c-K-ras2 and N-ras form the family. The 21 kDa protein (p21), which is present on the human chromosome at 11p11.5, 12p12.1 and 1p13, and consists of 4 coding exons and contains 189 amino acids that bind to the guanine nucleotide and localize in the cell membrane, is encoded . There are four kinds of Ras gene products, K-Ras 4A, K-Ras 4B, H-Ras and N-Ras.

     The FGF3, FGF4 gene is a type of fibroblast growth factor family that has various forms of biological processes including cleavage and cell survival activity and embryo development, cell growth, tissue repair, tumor growth and invasion, Was confirmed by conversion activity. The FGF4 gene was found in a wide variety of human tumors closely related to chromosome 11 cotransfection.

     Overexpression of the c-Myc oncogene is observed in most cancer cells, making it impossible to regulate the cell cycle. In normal cells, the c-Myc gene undergoes periodic testing by the p53-dependent ARF protein, a tumor suppressor gene. Recently, a direct effect of ARF on c-Myc has been newly discovered, and it is believed that it will be possible to develop new therapies for the treatment of cancer induced by abnormal regulation of c-Myc. Together with its interaction with P53 tumor suppressor, ARF binds c-Myc and does not induce c-Myc's target gene inhibition or apoptosis, it directly blocks c-Myc tumor suppressor function .

     The term " expression " as used herein also refers to the measurable expression level of a given nucleic acid or mRNA. The level of expression of the nucleic acid can be measured by a statistical method well known in the art. Fold also refers to the expression change value, which is well known in the art.

     A method for analyzing miRNA ID through data mining of miRNA can be performed by the following method.

     (1) miRNA sequence data includes quantitative expression levels of computerized miRNAs based on read counts in unknown samples. For miRNA expression comparisons in more than one sample, the level of expression in each sequencing sample is normalized by the total number of leads in the sample. The RPM is the value of the normalized expression level produced for each sample. Therefore, miRNA ID and RPM are used because they require expression levels of miRNAs in the data.

     (2) Single-molecule-based platforms, such as the Illumina genomic analyzer platform, typically exhibit inherent high errors because the platform detects all sequences containing errors. Higher error rates result in lower lead counts. miRNA IDs were classified as 3 groups at 5 folds or more (RPM> 5), less than 5 folds at 2.5 folds or more (5> RPM ≥ 2.5), less than 2.5 folds (RPM <2.5) ). Low-level expression (<2.5) and / or unmodified miRNA IDs among large amounts of collected data were removed to prevent errors.

     (3) To investigate the interaction of miRNA IDs and molecular targets, miRNA IDs available in the miRNA-target prediction databases mirWalk, DIANA-microT, miranda, miRBridge, miRDB, miRmap, miRNA Map, PITA, (MiRWalk, miranda, miRDB, RNA22, and TargetScan). Select three or more miRNA IDs based on the SUM value in five databases. By comparing five databases and checking twice, we can get three facts. First, there was a correlation between the target gene and miRNA IDs. The second was the reliability of predicted miRNAs belonging to specific genes. The last one increased the accuracy of extracted miRNA IDs through the complete removal of meaningless miRNA IDs.

     (4) Each target gene has more than five folds (RPM> 5), more than 2.5 folds at less than 5 folds (5> RPM ≥ 2.5), and multiple miRNA IDs labeled less than 2.5 folds. The miRNA IDs marked as less than 2.5 folds in the classification process were no longer present because they were all removed and the extracted miRNAs were presumed to have tumor miRNAs that could promote breast cancer development through the target gene. Each miRNA ID data is clustering to screen for common, new targets that are effective in breast cancer prevention.

     Furthermore, it can be used as a therapeutic agent or biomarker for regulating a breast cancer-related gene by miRNA ID analysis through data mining of miRNA.

     The following examples illustrate the present invention in more detail. These examples are for illustrating the present invention only and the scope of the present invention is not limited to the following examples.

<Experimental Example 1> RNA extraction method of frozen breast cancer tissue

     Prepare a biosample of frozen breast cancer tissue and extract the RNA from the prepared biosample using triazole. In the extraction method, 50-100 mg of the bio sample is finely crushed, and then added to 1 ml of triazole. After adding 0.2 ml of chloroform, the mixture is left at room temperature for 3 minutes and centrifuged at 12000 rpm for 15 minutes. Transfer the supernatant to a new tube, add 0.5 ml of isopropyl alcohol, and allow to stand for 10 minutes. After centrifugation at 12000 rpm for 10 minutes, the supernatant was discarded. The RNA pellet was subjected to DEPC-treated 75% ethanol (1 ml), taped, centrifuged at 12,000 rpm for 5 minutes, discarded, and the remaining RNA pellet was dried at room temperature for 10 minutes.

<Experimental Example 2> RNA extraction method of somatic cells

     A somatic cell biosample is prepared, and RNA is extracted from the prepared biosample using triazole. The extraction method is the same as in Experimental Example 1.

<Experimental Example 3> RNA extraction method of biological cells

     A biological cell sample is prepared, and RNA is extracted from the prepared biological cell sample using triazole. The extraction method is different only in that 50 to 100 mg of a biological cell sample is finely ground, and the remaining steps are the same as in Experimental Example 1.

     Using the RNA pellet prepared in Experimental Example 1, the nucleotide sequence of the RNA is determined through a known electrophoresis system. As shown in FIG. 1, an optimized miRNA analysis method was performed according to a flowchart showing an optimized miRNA analysis method through data mining of miRNA. Based on the identified nucleotide sequences, the groups were divided into three groups: 5 folds or more, less than 5 folds, 2.5 folds or less and 2.5 folds or less, depending on the level of expression level. Tumor miRNA data of each expression level was obtained using the data platform from the whole milk map (TCGA), and then miRNA data by expression level obtained from unknown sample and tumor miRNA data by expression level obtained from whole milk map were compared with each other, Obtain the result. Based on differences in expression levels, the obtained miRNA results were classified into three groups of 5 folds or more, 5 folds less than 2.5 folds and less than 2.5 folds, and groups less than 2.5 folds were removed from the classified groups to regroup Obtain miRNA results (result A) by expression level. After that, the results were compared with those of the target prediction groups, mirWalK, DIANA-microT, miranda, miRBridge, MiRDB, Mirmap, miRNA Map , PITA, RNA22, and Targetscan (result B).

     Subsequently, the results were compared with reference groups miRBase, ASRP, micro RNAMAP, miRGen, and CoGemiR, again with reference groups, in which the expression levels of miRNAs regrouped at more than 5 folds and less than 5 folds at 2.5 folds or more, c). The result B and C are compared with each other to obtain the miRNA ID from a common result. The obtained miRNA ID is re-confirmed using the public PubMed database to obtain the miRNA ID having high accuracy from the miRNA ID.

     Using the RNA pellet prepared from Experimental Example 2, the nucleotide sequence of the RNA is determined through a known electrophoresis system.

     Based on the identified nucleotide sequences, the groups were divided into three groups: 5 folds or more, less than 5 folds, 2.5 folds or less and 2.5 folds or less, depending on the level of expression level. Tumor miRNA data of each expression level was obtained using the data platform from the whole milk map (TCGA), and then miRNA data by expression level obtained from unknown sample and tumor miRNA data by expression level obtained from whole milk map were compared with each other, Obtain the result. Based on differences in expression levels, the obtained miRNA results were classified into three groups of 5 folds or more, 5 folds less than 2.5 folds and less than 2.5 folds, and groups less than 2.5 folds were removed from the classified groups to regroup Obtain miRNA results (result D) by expression level. After that, the result D, that is, the expression levels of miRNAs according to the regrouped expression levels were 5 fold or more and less than 5 folds and 2.5 fold or more were divided into target prediction databases mirWalK, DIANA-microT, miranda, miRBridge, MiRDB, Mirmap, miRNA Map , PITA, RNA22, and Targetscan (result E).

     The results were compared to the reference databases miRBase, ASRP, micro RNAMAP, miRGen, and CoGemiR, again with reference to the results D, that is, the expression level of miRNAs per regrouped expression level was 5 fold or more and less than 5 folds to 2.5 fold or more. F). The obtained results E and F are compared with each other to obtain the miRNA ID from a common result. The obtained miRNA ID is re-confirmed using the public PubMed database to obtain the miRNA ID having high accuracy from the miRNA ID.

     Using the RNA pellet prepared in Experimental Example 3, the nucleotide sequence of the RNA is determined through a known electrophoresis system. Based on the identified nucleotide sequences, the groups were divided into three groups: 5 folds or more, less than 5 folds, 2.5 folds or less and 2.5 folds or less, depending on the level of expression level. Tumor miRNA data of each expression level was obtained using the data platform from the whole milk map (TCGA), and then miRNA data by expression level obtained from unknown sample and tumor miRNA data by expression level obtained from whole milk map were compared with each other, Obtain the result. Based on differences in expression levels, the obtained miRNA results were classified into three groups of 5 folds or more, 5 folds less than 2.5 folds and less than 2.5 folds, and groups less than 2.5 folds were removed from the classified groups to regroup Obtain miRNA results (result G) by expression level. After that, the results were compared with those of the target prediction database, mirWalK, DIANA-microT, miranda, miRBridge, MiRDB, Mirmap, miRNA Map , PITA, RNA22, and Targetscan (result H).

     The results were compared with those of reference groups miRBase, ASRP, micro RNAMAP, miRGen, and CoGemiR. The results were compared with those of reference groups miRBase, ASRP, miRGen and CoGemiR I). The obtained results are compared with each other to obtain miRNA IDs from common results. The obtained miRNA ID is re-confirmed using the public PubMed database to obtain the miRNA ID having high accuracy from the miRNA ID.

     Attachment 2 is raw data for breast cancer downloaded by the TCGA. The list of miRNAs predicted in relation to oncogenes and tumor suppressor genes is sorted according to the expression levels (Figures 3 to 6).

     Fig. 3 relates to a list of putative miRNAs capable of regulating BRCA1 / 2, wherein BRCA1 occupies 11% of the total [52 miRNAs], followed by 2.5 folds or more (> 2.5) Have 1.5% [7 miRNAs].

     Figure 4 shows that Cyclin D1, which regulates miRNAs, occupies 18% of the total and 5% or more (> 5) miRNAs occupy 2% of the [83 miRNAs] "He said.

     Figure 5 relates to a tumor gene, N-RAS, which regulates miRNAs, with over 5 fold (≥5) miRNAs occupying 23% of the total [102 miRNAs] and over 2.5 folds (> 2.5) of 2.9% [13 miRNAs ].

     FIG. 6 is a tumor gene that regulates miRNAs, which is related to FGF3, FGF4, HER2, and c-Myc. FGF3 occupies 0.4% of the total miRNAs over 5 folds (≥5) [2 miRNAs] Over fold (> 2.5) occupied 0.2% [1 miRNA]. In the case of FGF4, over 5 fold (≥5) miRNAs occupied 0.4% [2 miRNAs] and over 2.5 folds (> 2.5) occupied 0.2% [1 miRNA]. For HER2, over 5 fold (≥5) miRNAs occupied 2.9% [13 miRNAs] and over 2.5 folds (> 2.5) occupied 0.4% [2 miRNAs]. For C-Myc, 0.2% miRNAs were 5 fold or greater (≥5).

The present invention has been described in detail by way of examples and experimental examples. The above embodiments are merely illustrative examples for explaining the present invention in more detail, and the nature of the technical idea of the present invention is not changed or reduced. It will be obvious to those skilled in the art that various embodiments and applications not shown in the above embodiments are possible.

Claims (14)

(a) a first step of preparing an unknown bio sample;
(b) extracting an unknown microRNA (miRNA) from the sample prepared in the first step and sorting the sequenced miRNA into three groups of 5 fold or more, less than 5 folds, 2.5 folds or less, 2.5 folds or less A second step of providing miRNA data according to an expression level;
(c) obtaining tumor miRNA data for each level of expression using a data platform from a whole milk map (TCGA);
(d) comparing the expression level-specific miRNA data obtained in the second step with the expression level-specific miRNA data obtained in the third step to obtain miRNA results according to a common expression level;
(e) dividing the expression level of the mRNA by the common expression level obtained in the fourth step into three groups of 5 folds or more, less than 5 folds, 2.5 folds or more and 2.5 folds or less, A fifth step of removing a group less than 2.5 folds to obtain miRNA results according to regrouping expression levels;
(f) comparing the miRNA results of the regrouping expression levels obtained through the fifth step with a target prediction database to obtain a common result (I);
(g) obtaining the miRNA ID by the sixth step; and performing seventh step of obtaining the miRNA ID by the sixth step.
The method according to claim 1,
Wherein the unknown bio sample in the first step is obtained from fresh or frozen breast cancer tissues, cells, blood, serum or plasma.
The method according to claim 1,
Wherein miRNA ID is analyzed by data mining of miRNA using trizol or Triton X-100 to extract the miRNA of the second step.
The method according to claim 1,
The target prediction database of the sixth step is selected from at least one of mirWalK, DIANA-microT, miranda, miRBridge, MiRDB, Mirmap, miRNA Map, PITA, RNA22 and Targetscan. How to analyze ID.
The method according to claim 1,
Between the sixth and seventh steps, (h) comparing the miRNA results according to the regrouping expression level obtained in the step (5) with an RNA reference database to obtain a common result (II) Wherein the method comprises the steps of: (a) comparing the miRNA IDs of the miRNAs obtained in step
6. The method of claim 5,
And confirming the miRNA ID using the PubMed database by confirming the miRNA ID obtained in the seventh step with the data mining method of miRNA.
6. The method of claim 5,
Wherein the reference database is selected from one or more of miRBase, ASRP, micro RNAMAP, miRGen, and CoGemiR.
(a) a first step of preparing an unknown bio sample;
(b) extracting an unknown microRNA (miRNA) from the sample prepared in the first step and sorting the sequenced miRNA into three groups of 5 fold or more, less than 5 folds, 2.5 folds or less, 2.5 folds or less A second step of providing miRNA data according to an expression level;
(c) obtaining tumor miRNA data for each level of expression using a data platform from a whole milk map (TCGA);
(d) comparing the expression level-specific miRNA data obtained in the second step with the expression level-specific miRNA data obtained in the third step to obtain miRNA results according to a common expression level;
(e) after re-classification into three groups of 5 fold or more, less than 5 folds, 2.5 folds or less, 2.5 folds or less, depending on the difference in expression level from the common expression level miRNA results obtained in the fourth step, A fifth step of removing a group less than 2.5 folds to obtain miRNA results according to regrouping expression levels;
(f) comparing the miRNA results of the regrouping expression levels obtained through the fifth step with a target prediction database to obtain a common result (I);
(g) obtaining the miRNA ID by the sixth step; and (7) obtaining the miRNA ID by the sixth step. The miRNA data mining method characterized by identifying the tumor gene and the tumor suppressor gene by using the optimized miRNA analysis method through data mining of the miRNA performing the miRNA ID RTI ID = 0.0 &gt; miRNA &lt; / RTI &gt;
9. The method of claim 8,
Between the sixth and seventh steps, (h) comparing the miRNA results according to the regrouping expression level obtained in the step (5) with an RNA reference database to obtain a common result (II) Wherein the method comprises the steps of: (a) comparing the miRNA IDs of the miRNAs obtained in step
9. The method of claim 8,
And confirming the miRNA ID using the PubMed database by confirming the miRNA ID obtained in the step 7 in the data mining of the miRNA.
9. The method of claim 8,
Wherein the oncogene is analyzed for one or more of HER2, Cyclin D1, N-RAS, FGF3, FGF4, and C-Myc.
9. The method of claim 8,
Wherein the tumor suppressor gene is one or more of BRCA1 and BRCA2.
delete delete

Images