An algorithm for identifying circular RNA encoded by plant mitochondrial genomes
Through the MeCi algorithm, blast comparison and specific screening conditions are used to solve the problem that the existing technology cannot effectively identify the circRNA of plant mitochondrial genomes, achieving higher recognition accuracy and credibility, and significantly improving the circRNA of plant mitochondrial genomes recognition ability.
Patent Information
- Application Number
- CN202210276265.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-21
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2042-03-21
AI Technical Summary
Existing circRNA recognition algorithms cannot effectively identify circular RNAs encoded by plant mitochondrial genomes, especially due to the unique characteristics of plant mitochondrial genomes, such as class II intron self-splicing, RNA editing sites and repeat sequences, resulting in recognition accuracy and ability. Low reliability.
An algorithm called MeCi was developed to identify circRNAs encoded by plant mitochondrial genomes using blast alignment and specific screening conditions, including mismatches of the fragments and overlap or vacant nucleotides that allow reverse alignment of fragments and overlap or vacant nucleotides at cyclization sites. , avoid relying on intron reverse splicing rules and improve the credibility of identification.
The MeCi algorithm significantly improves the recognition ability of the plant mitochondrial genome encoding circRNA. Compared with existing algorithms such as CIRI2 and find_circ, it can more accurately identify more real circRNAs, enhancing the credibility and accuracy of recognition.
Smart Images

Figure CN114708912B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of bioinformatics, and in particular to an identification algorithm for circular RNA encoded by plant mitochondrial genome. Background Art
[0002] Ribonucleic acid (RNA) is a ribonucleic acid polymer transcribed from deoxyribonucleic acid (DNA), including messenger RNA (mRNA), transfer RNA (tRNA), ribosomal RNA (rRNA) and small nuclear RNA (snRNA). RNA has both linear molecules and circular structure molecules connected head to tail in the form of covalent bonds, namely circular RNA (circRNA). The closed circular structure of circRNA molecules makes them more stable than linear RNA and not easily degraded by ribonuclease exonucleases. Transcriptome sequencing (RNA-sequencing, RNA-seq) technology is the basis for high-throughput identification of circRNA. RNA-seq can be divided into two types: single-end sequencing and double-end sequencing. Single-end sequencing reads the RNA sequence from one end, while double-end sequencing reads the sequence from both ends of the RNA, and then combines the double-end sequencing sequences together through the overlapping sequences at both ends. Double-end sequencing can obtain longer reads.
[0003] Identifying circRNA from high-throughput sequencing results is a key step in its functional study, which requires the assistance of appropriate algorithms. Existing studies have shown that intron back splicing is the main pathway for the generation of circRNA. Commonly used circRNA identification algorithms include CIRI2 [1] 、CIRI [2] 、find_circ [3] 、CIRCexplorer [4] and UROBORUS [5] These algorithms first identify the reverse splicing products through different principles, and then filter them according to the conserved sequences of the nuclear intron splicing sites or gene annotation information, so as to obtain candidate circRNAs with higher feasibility.
[0004] There are differences between plant mitochondrial genomes and nuclear genomes, mainly including the following parts: (1) Most of the introns in the plant mitochondrial genome are self-splicing class II introns, and there is no report showing that this type of intron has a reverse splicing pathway; (2) Compared with the nuclear genome, the plant mitochondrial genome is very small (for example, the mitochondrial genome size of corn B73 is about 570kb, while its nuclear genome size is about 2.18Gb); (3) There are a large number of RNA editing sites in the coding region of the plant mitochondrial genome, which do not exist in the nuclear genome. These differences make the circular RNA (circRNA) encoded by the plant mitochondrial genome have different characteristics from the circRNA encoded by the nuclear genome. Currently, most commonly used algorithms are designed for circRNA encoded by the nuclear genome, and they cannot effectively identify circRNA encoded by the plant mitochondrial genome.
[0005] Take CIRI2 as an example. CIRI2 is one of the most widely used circRNA identification algorithms. It uses maximum likelihood estimation based on multiple seed matching to identify reads containing back-spliced junctions (BSJs), and reduces false positives caused by repetitive sequences and base mismatches by calculating the false discovery rate (FDR). CIRI2 has several problems in identifying circRNAs encoded by plant mitochondrial genomes, including: (1) CIRI2 screens circRNAs by identifying intron back-splicing products, while no back-splicing is found in class II introns of plant mitochondrial genomes; (2) There are a large number of RNA editing sites in plant mitochondrial genomes, which leads to an excessively high FDR when CIRI2 predicts that plant mitochondrial genomes encode circRNAs, which in turn causes a large number of real circRNAs to be filtered out; (3) Plant mitochondrial genomes contain a large number of repetitive sequences, and CIRI2 will filter out circRNAs originating from these repetitive regions.
[0006] References:
[0007] [1]Gao Y, Zhang J, Zhao F. Circular RNA identification based on multipleseed matching[J]. Briefings in Bioinformatics.2018,19(5):803-810.
[0008] [2]Gao Y, Wang J, Zhao F. CIRI: an efficient and unbiased algorithm for de novo circular RNA identification[J]. Genome Biology.2015,16(1).
[0009] [3]Memczak S, Jens M, Elefsinioti A, et al. Circular RNAs are a largeclass of animal RNAs with regulatory potency[J].Nature.2013,495(7441):333-338.
[0010] [4]Ma
[0011] [5]Song Summary of the invention
[0012] The technical problem to be solved by the present invention is how to identify circular RNA (circRNA) encoded by plant mitochondrial genome.
[0013] In order to solve the above technical problems, the present invention first provides a method for identifying circRNA encoded by plant mitochondrial genome. The method comprises the following steps: using blast tool to align RNA-seq data to reference genome to obtain an alignment result file; performing sequence feature screening on the data in the alignment result file to obtain a candidate circRNA file; and performing conditional screening on the candidate circRNA file to obtain a circRNA encoded by plant mitochondrial genome.
[0014] In the method described above, the sequence feature screening can be to remove the reads in the alignment result file in which only one sequence fragment is aligned to the reference genome, and retain the reads in which only two sequence fragments can be aligned to the reference genome in opposite directions.
[0015] In the above method, the sequence feature screening may include the following steps:
[0016] A1) removing the read lengths in which only one sequence fragment is aligned to the reference genome in the alignment result files, and retaining the alignment result files in which two or more sequence fragments can be aligned to the read lengths of the reference genome in opposite directions.
[0017] A2) extracting the position information data of the two or more sequence fragments in the comparison result file and the positive and negative strand information data of the sequence fragments corresponding to the reference genome.
[0018] A3) processing the positive and negative strand information data, retaining only the read lengths of two sequence fragments aligned to the reference genome in opposite directions, sorting and integrating the sequence fragments according to the position information, and obtaining candidate circRNAs.
[0019] In the above method, the position information may be the ID information of the read length corresponding to the sequence fragment. The position information may also include the position information of the sequence fragment on the read length, and the position information of the sequence fragment aligned to the reference genome sequence. The direction may be the direction from 5' to 3' of the nucleotide sequence.
[0020] In the above method, the number of nucleotides in the sequence fragment is greater than or equal to 11.
[0021] In the method described above, A3) processing the positive and negative chain information may include the following steps: filling the positive and negative chain information data of each sequence fragment into two files, a positive chain file and a negative chain file, respectively; integrating the sequence fragments in the positive chain file to obtain a positive chain integration file, exchanging the starting and ending position information of the sequence fragments in the negative chain file, and then integrating them to obtain a negative chain integration file.
[0022] In the above method, the "E" parameter of the blast tool can be set in the range of 10 -5 ~5.
[0023] The setting of the "E" parameter can specifically be 2.
[0024] In the above method, the screening conditions may be as follows:
[0025] B1) The number of overlapping nucleotides and missing nucleotides at the cyclization site should be less than or equal to 3.
[0026] B2) The length of the candidate circRNA is less than or equal to 10,000 nucleotides and greater than or equal to the length of the corresponding read length.
[0027] The cyclization site may be the junction point in the read length of two sequence fragments aligned to the reference genome in opposite directions. The overlapping nucleotides may be the positions where two sequence fragments aligned to the reference genome in opposite directions have overlapping nucleotides. The vacant nucleotides may be the nucleotides that cannot be aligned to the reference genome at the junction of two sequence fragments aligned to the reference genome in opposite directions.
[0028] In the above method, the length of the candidate circRNA can be the number of nucleotides between the starting nucleotide and the ending nucleotide of the cyclization site corresponding to the reference genome. The starting nucleotide can correspond to the 5' terminal nucleotide aligned to the reference genome. The ending nucleotide can correspond to the 3' terminal nucleotide aligned to the reference genome.
[0029] In the method described above, the RNA-seq data may be a fasta format file. The fasta format file may be a split RNA-seq data file.
[0030] In the above method, the sequence of the reference genome may be a formatted sequence of the reference genome. The formatting tool may be formatdb.
[0031] In order to solve the above technical problems, the present invention also provides a device for identifying circular RNA (circRNA) encoded by plant mitochondrial genome. The device may include the following modules:
[0032] C1) Sequence alignment module: used to align plant mitochondrial transcriptome sequencing (RNA-seq) data to the reference genome using the blast tool to obtain an alignment result file.
[0033] C2) Sequence feature screening module: used to perform sequence feature screening on the data in the comparison result file to obtain a candidate circRNA file.
[0034] C3) Condition screening module: used to perform condition screening on the candidate circRNA file to obtain the final circRNA.
[0035] The sequence feature screening can be established by a method comprising the following steps: removing the reads in the alignment result file in which only one sequence fragment is aligned to the reference genome, and retaining only the reads in which two sequence fragments can be aligned to the reference genome in opposite directions.
[0036] The "E" parameter of the blast tool can be set in the range of 10 -5 ~5.
[0037] The setting of the "E" parameter can specifically be 2.
[0038] In the above-mentioned device, the condition screening module can be established by a method comprising the following steps:
[0039] B1) The number of overlapping nucleotides and missing nucleotides at the cyclization site should be less than or equal to 3.
[0040] B2) The length of the candidate circRNA is less than or equal to 10,000 nucleotides and greater than or equal to the length of the corresponding read length.
[0041] The cyclization site may be the junction point of two sequence fragments aligned to the reference genome in opposite directions on the read length. The overlapping nucleotides may be the overlapping nucleotides of the positions of two sequence fragments aligned to the reference genome in opposite directions. The vacant nucleotides may be the nucleotides that exist at the junction of two sequence fragments aligned to the reference genome in opposite directions and cannot be aligned to the reference genome.
[0042] In the device described above, the length of the candidate circRNA may be the number of nucleotides between the start nucleotide and the end nucleotide of the cyclization site corresponding to the reference genome. The start nucleotide may correspond to the 5' end nucleotide of the reference genome. The end nucleotide may correspond to the 3' end nucleotide of the reference genome.
[0043] In the above device, the RNA-seq data may be a fasta format file. The fasta format file may be a split RNA-seq data file.
[0044] In the above-mentioned device, the sequence of the reference genome may be a formatted sequence of the reference genome. The formatting tool may be formatdb.
[0045] In order to solve the above technical problem, the present invention further provides a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to establish the steps of the method described above or the computer program enables a computer to establish the modules of the device described above.
[0046] In order to solve the above technical problem, the present invention further provides a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the steps of the method described above or the computer program enables a computer to execute the steps of the module of the device described above.
[0047] In view of the shortcomings of the existing methods, the present invention aims to develop an identification algorithm suitable for plant mitochondrial genome-encoded circRNAs, namely mitochondrion-encoded circRNAidentifier (MeCi), by combining the characteristics of plant mitochondrial genomes. The algorithm is used to identify high-confidence mitochondrial genome-encoded circRNAs from plant RNA-seq data.
[0048] CIRI2 identifies circRNAs by recognizing the reverse splicing products of introns, while most of the introns in plant mitochondrial genomes are class II introns, and there are no reports showing that such introns have reverse splicing pathways. MeCi detects circularization junctions (i.e., the junctions between two reverse alignment sequences in the read length) in RNA-seq reads based on blast alignment; MeCi's recognition of circRNAs is not limited to the reverse splicing of introns. There are a large number of RNA editing sites in plant mitochondrial genomes. If CIRI2 is used to predict mitochondrial genome-encoded circRNAs, the FDR will be too high, resulting in a large number of circRNAs being filtered out. MeCi uses a larger blastn E value (the default setting is 2, and the adjustable range is 10 -5 ~5), retaining both the alignment fragments of smaller length (≥11nt) and the read lengths with mismatches. There are a certain number of repetitive sequences in the plant mitochondrial genome. MeCi allows a mitochondrial genome-encoded circRNA to be aligned to different locations in the genome, while CIRI2 will delete these circRNAs. CIRI2 does not allow circRNAs with overlapping or missing nucleotides at the circularization site (the junction of two reverse alignment sequences on the read length), but many plant mitochondrial genomes contain overlapping or missing nucleotides at the junction of circRNAs. MeCi allows circRNAs to have overlapping and missing nucleotides at the circularization site, and limits their number to no more than 3.
[0049] The innovative features of the plant mitochondrial genome encoding circRNA method MeCi established in the present invention are as follows:
[0050] 1. MeCi is based on blastn alignment and uses e-values to control mismatches and alignment fragment length.
[0051] 2. MeCi's screening of circRNAs is based on the characteristics of plant mitochondrial genomes and does not rely on intron backsplicing rules. The specific conditions are as follows:
[0052] a. A read length is required to contain only two reverse fragments aligned to the reference genome;
[0053] b. There are no more than 3 overlapping or missing nucleotides at the junction of the two reverse alignment fragments of the read length;
[0054] The default cE value is set to 2 and the adjustable range is 10 -5 ∼5, and multiple mismatched nucleotides were allowed;
[0055] d. The length of the two reverse alignment fragments of the read length is not less than 11 nucleotides;
[0056] e. The number of nucleotides between the start nucleotide and the end nucleotide of the circularization site between the two reverse alignment fragments of the read length on the reference genome is greater than the number of nucleotides of the read length.
[0057] Compared with the prior art, the present invention has the following beneficial effects:
[0058] The existing circRNA recognition algorithms are mainly used to identify the reverse splicing products from nuclear introns, while plant mitochondrial introns are mostly class II introns, and there is no reverse splicing pathway. MeCi is designed according to the characteristics of plant mitochondrial genomes. First, MeCi aligns RNA-seq data to the reference genome through blast, and screens the reads with only two reverse alignment fragments; then, through a series of screening conditions, a high-confidence read is obtained. The specific screening conditions are as follows: multiple mismatched bases are allowed in the alignment fragments, no more than 3 bases are allowed to overlap or be missing at the cyclization site (the connection point between the two reverse alignment sequences on the read length), a circRNA is allowed to be aligned to multiple locations in the genome, and the circRNA length (the number of nucleotides between the first nucleotide and the last nucleotide of the cyclization site on the reference genome) is greater than the read length (the number of nucleotides in the read length).
[0059] In the embodiment of the present invention, three circRNA identification algorithms, namely MeCi, find_circ and CIRI2, were used to identify mitochondrial genome encoded circRNAs in the maize and Arabidopsis RNA-seq data downloaded from NCBI. The results showed that in the maize 3-times repeated data, MeCi identified 32686, 23560 and 2638 mitochondrial genome encoded circRNAs, find_circ identified 1782, 1197 and 1256, and CIRI2 identified 263, 219 and 184, respectively. In the Arabidopsis 2-times repeated data, MeCi identified 49772 and 60289 mitochondrial genome encoded circRNAs, find_circ identified 2333 and 2782, respectively, and CIRI2 identified 453 and 593, respectively. The results of RT-PCR amplification verification by designing divergent primers showed that most of the mitochondrial genome-encoded circRNAs of maize and Arabidopsis newly detected by MeCi were correct, thus proving that MeCi has a high degree of credibility in predicting mitochondrial genome-encoded circRNAs of plants. The ability of the MeCi algorithm established by the present invention to identify mitochondrial genome-encoded circRNAs of plants is superior to the existing circRNA identification methods, and can effectively identify mitochondrial genome-encoded circRNAs of plants. BRIEF DESCRIPTION OF THE DRAWINGS
[0060] Figure 1 To determine and classify reads containing circularization sites. If a read contains two segments that are aligned to the reference genome in opposite directions, it is determined as a candidate circRNA, and the connection point of the two segments on the read is the circularization site. According to the difference in the circularization site, circRNA can be divided into three types. Type I: The circularization site contains nucleotides that are repeatedly aligned to the reference genome (i.e., overlapping nucleotides), Type II: The circularization site contains nucleotides that cannot be aligned to the reference genome (i.e., vacant nucleotides), and Type III: The circularization site does not contain overlapping or vacant nucleotides.
[0061] Figure 2 This is the parameter input and verification flow chart.
[0062] Figure 3 This is the sequence preprocessing flowchart.
[0063] Figure 4 Split flowchart for sequence files.
[0064] Figure 5 Flowchart of sequence alignment.
[0065] Figure 6 This is the flow chart of Blast alignment data analysis.
[0066] Figure 7 Flow chart for circular RNA screening.
[0067] Figure 8 Complete flow chart for identification of plant mitochondrial circular RNAs.
[0068] Fig. 9 Comparison of the number of high-confidence mitochondrial genome-encoded circRNAs predicted from RNA-seq data by MeCi, CIRI2, and find_circ together or individually.
[0069] Fig.10 RT-PCR validation of the newly predicted maize mitochondrial genome-encoded circRNAs by MeCi. PCR was used to validate the circRNAs derived from maize Zmrrn26, Zmcob, Zmnad2T2, and ZmtrnY. The gene structure is shown in the figure, the bold line indicates the intron (int), and the 5' and 3' end positions of the mature linear RNA are indicated by arrows. The black and blue lines represent the positions of the circRNAs detected by MeCi and RT-PCR at the site, respectively. The positions of the convergent primers and divergent primers are shown in the figure. Con: convergent primer, Di: divergent primer; gD: gDNA, - and +: RNase– and RNase+cDNA; the arrows point to the target fragments amplified by the convergent primers; the amplified bands of the divergent primers indicated by the broken lines were recovered, cloned, transformed, and sequenced, respectively. M: DNA marker.
[0070] Fig.11 RT-PCR validation of the newly predicted Arabidopsis mitochondrial genome-encoded circRNAs by MeCi. PCR was used to validate the circRNAs derived from the Arabidopsis Atatp1, Atcox3, Atnad4L, AttrnY and AttrnD loci. The gene structure is shown in the figure, and the 5' and 3' end positions of the mature linear RNA are indicated by arrows. The black and light gray lines represent the positions of the circRNAs detected by MeCi and RT-PCR at the locus, respectively. The positions of the convergent primers and divergent primers are shown in the figure. Con: convergent primer, Di: divergent primer; gD: gDNA, - and +: RNase– and RNase+cDNA; the arrows point to the target fragments amplified by the convergent primers; the bands amplified by the divergent primers indicated by the broken lines were recovered, cloned, transformed and sequenced respectively. M: DNA marker. DETAILED DESCRIPTION
[0071] The present invention is further described in detail below in conjunction with specific embodiments, and the examples provided are only for illustrating the present invention, rather than for limiting the scope of the present invention. The examples provided below can be used as a guide for further improvements by those of ordinary skill in the art, and do not constitute a limitation of the present invention in any way.
[0072] The experimental methods in the following examples, unless otherwise specified, are all conventional methods, and are performed according to the techniques or conditions described in the literature in the field or according to the product instructions. The materials, reagents, etc. used in the following examples, unless otherwise specified, can all be obtained from commercial channels.
[0073] Example 1: Establishment of MeCi, an algorithm for identifying circRNA encoded by plant mitochondrial genome
[0074] 1. Data input and data preprocessing
[0075] 1.1 Data Input
[0076] When the MeCi algorithm starts to operate, parameters need to be input and checked. The first step is to check whether the plant mitochondrial RNA-seq sequence file and the reference genome sequence file are correctly input. The second step is to check whether each option parameter (including "-in1", "-in2" and "-genome" parameters, Table 1) are set. If the parameters are not set, fill in the default setting parameters. The third step is to create a result output folder (the process is as follows Figure 2 shown).
[0077] 1.2 Data Preprocessing
[0078] RNA-seq is divided into two types: single-end sequencing and double-end sequencing. Because the processing methods of the two types of sequencing files are different, the input sequence files need to be preprocessed.
[0079] Single-end sequencing: Convert the sequence file from fastq format to fasta format for subsequent processing.
[0080] Paired-end sequencing: First, use the double-end sequencing file splicing program flash to splice the two sequence files of the double-end sequencing and obtain a spliced sequence file. Then, convert the spliced sequence file from fastq format to fasta format for subsequent processing (the process is as follows Figure 3 shown).
[0081] Table 1. Required input data and optional options for MeCi
[0082]
[0083]
[0084] 2. Sequence Alignment
[0085] 2.1 Sequence file splitting
[0086] Before blast alignment, the sequences in the fasta format sequence file are split into multiple files according to a certain size to improve the efficiency of subsequent sequence alignment. The size of the split file is 0.1% to 100% of the read length of the sequencing file.
[0087] Calculate the number of reads of the fasta format sequence file in step 1.2, and divide the number of reads by the sequence chunk size parameter (chunksize; generally set to 0.1% to 100% of the sequence file read length) to obtain the number of sequence chunks that need to be split. Create a split file and set the file name. Write and call a Perl language script to split the fasta format sequence file and fill it into the previously prepared split file to obtain the fasta format split file (the process is as follows Figure 4 shown).
[0088] 2.2 Sequence alignment
[0089] Sequence alignment was performed using the formatdb and blastall programs obtained from NCBI. The reference genome sequence was formatted using the formatdb program to obtain the formatted reference genome sequence, and then sequence alignment was started.
[0090] Create a multithreaded alignment, calculate the base number of each sequence in the split file in fasta format, use blastall to batch align the sequences in the split file in fasta format with the formatted reference genome sequence, and output the alignment result file (the process is as follows Figure 5 The default setting of E value in the comparison process using blastall is 2, and the range is 10. -5 ~5. Decreasing the E value will reduce the mismatch rate of the blastn alignment fragment (the sequence fragment aligned to the reference genome) and increase its length; increasing the E value will shorten the alignment fragment and increase the mismatch rate. The number of nucleotides in the read alignment fragment is greater than or equal to 11.
[0091] Table 2. Programs used by MeCi
[0092] name type Function blastall program Sequence alignment formatdb program Reference genome sequence formatting fastalength program Calculate the number of sequence bases FastqToFa program Convert Fastq format files to fasta format flash program Paired-end sequencing reads assembly
[0093] 3. Sequence feature screening and obtaining candidate circular RNA
[0094] This step processes the alignment result file obtained in step 2.2 to screen out alignment results that meet the characteristics of circRNA.
[0095] According to existing reports, circRNA recognition algorithms mainly identify circRNAs based on the sequence characteristics of reads, that is, a read can be divided into two segments that are mapped to the reference genome, and the two segments are mapped to the reference genome in opposite directions. The connection point of the two reverse-aligned sequences on the read is the circularization site of the circRNA ( Figure 1 ).
[0096] According to the differences in the cyclization sites, circRNA can be divided into three types. Type I: The cyclization site contains nucleotides that are repeatedly mapped to the reference genome (i.e., overlapping nucleotides), Type II: The cyclization site contains nucleotides that cannot be mapped to the reference genome (i.e., vacant nucleotides), and Type III: The cyclization site does not contain overlapping nucleotides or vacant nucleotides ( Figure 1 ).
[0097] In the algorithm, there are mainly the following steps (the process is as follows Figure 6 shown):
[0098] The first step is to analyze the data in the blast comparison result file and delete the reads that have only one sequence fragment that can be aligned to the reference genome, that is, remove the reads that are completely aligned with the reference genome, and only retain the candidate reads (that is, the reads that have more than two sequence fragments that are aligned to the reference genome).
[0099] The second step is to extract the position information of the sequence fragments on each read that can be matched to the reference genome and their position information on the reference genome sequence, as well as the ID information data of the read, from the comparison result file containing the candidate reads; calculate the positive and negative chain information corresponding to each sequence fragment on the reference genome, and fill the data into the positive chain and negative chain files.
[0100] The third step is to process the positive chain file and the negative chain file separately. For the positive chain file, the sequence fragments therein are integrated to obtain the positive chain integration file; for the negative chain file, the start and end position information of the sequence fragments therein are exchanged, and then integrated to obtain the negative chain integration file. The sequences in the positive chain integration file and the negative chain integration file are sorted by the ID of the read length (the ID information of the read length comes from the original RNA-seq data), and the read lengths of the sequence fragments that are aligned to the reference genome in two and only two directions are retained. The two sequence fragments on the same read length are integrated together to obtain the candidate circRNA file.
[0101] 4. Conditional screening of candidate circRNAs to obtain the final plant mitochondrial genome-encoded circRNA
[0102] The candidate circRNAs in step 3 are obtained by screening sequence features, and there may be false positives. In order to obtain highly reliable circRNAs, it is necessary to combine the characteristics of plant mitochondrial genomes and set screening conditions to filter candidate circRNAs, so as to obtain the final plant mitochondrial genome-encoded circRNAs (process as shown in Figure 7 The filtering conditions are as follows:
[0103] First, the number of overlapping nucleotides (i.e., there are several overlapping nucleotides between the two sequence fragments aligned to the reference genome in opposite directions) and missing nucleotides (i.e., there are nucleotides that cannot be aligned to the reference genome between the two sequence fragments aligned to the reference genome in opposite directions) at the circularization site (the junction point in the read length of the two sequence fragments aligned to the reference genome in opposite directions) should be less than or equal to 3 ( Figure 1 ).
[0104] Second, the length of the circRNA (the number of nucleotides between the start nucleotide and the end nucleotide of the circularization site corresponding to the reference genome) is less than or equal to 10,000 nucleotides, and greater than or equal to the length of the corresponding read length (number of sequence nucleotides). Among them, the start nucleotide corresponds to the 5' end nucleotide of the reference genome; the end nucleotide corresponds to the 3' end nucleotide of the reference genome.
[0105] The plant mitochondrial genome-encoded circRNA obtained by the above screening conditions is filled into the output table, and two result files are generated: one file containing the final plant mitochondrial genome-encoded circRNA data, named circrna_details.xls; the other file containing the final circRNA corresponding sequencing read length data, named read_details.xls.
[0106] The complete flow chart of the plant mitochondrial genome encoded circRNA identification method MeCi established by the present invention is as follows: Figure 8 shown.
[0107] Example 2: Comparison of application effects of MeCi and prior art
[0108] 1. Comparison of three circRNA identification algorithms in predicting circRNAs encoded by plant mitochondrial genomes
[0109] RNA-seq data of maize and Arabidopsis mitochondrial circRNA were downloaded from NCBI. The maize and Arabidopsis RNA-seq data contained 3 and 2 biological replicates, respectively.
[0110] Table 3. Transcriptome data of mitochondrial circRNAs in maize and Arabidopsis
[0111]
[0112] Three circRNA identification algorithms, namely MeCi, find_circ and CIRI2, were used to identify mitochondrial genome-encoded circRNAs in maize and Arabidopsis from two sets of data. In the three replicates of maize, MeCi identified 32686, 23560 and 2638 mitochondrial genome-encoded circRNAs, find_circ identified 1782, 1197 and 1256, and CIRI2 identified 263, 219 and 184, respectively (Table 4). In the two replicates of Arabidopsis, MeCi identified 49772 and 60289 mitochondrial genome-encoded circRNAs, find_circ identified 2333 and 2782, and CIRI2 identified 453 and 593, respectively.
[0113] In order to improve the reliability of prediction, circRNAs that appeared in at least two replicates were selected as high-confidence mitochondrial genome-encoded circRNAs. According to this standard, MeCi, find_circ and CIRI2 identified 7524, 674 and 66 mitochondrial genome-encoded circRNAs in maize, respectively, and 9819, 685 and 105 in Arabidopsis, respectively. Among them, 37 and 42 mitochondrial genome-encoded circRNAs were jointly predicted by the three algorithms in maize and Arabidopsis, respectively (Table 4 and Fig. 9 ); 7482 and 9749 were newly discovered by MeCi in maize and Arabidopsis, respectively, which were not identified by the existing algorithms find_circ and CIRI2 ( Fig. 9 ).
[0114] Table 4. Comparison of the number of mitochondrial genome-encoded circRNAs predicted by different circRNA identification methods in the same RNA-seq data
[0115]
[0116] Among them, the main steps of the find_circ algorithm are as follows:
[0117] (1) Using Bowtie2, each read of the RNA-seq sequence file was aligned with the reference genome, and reads that completely matched the genome were discarded;
[0118] (2) Take 20 base sequences at the 5' and 3' ends of the remaining read as anchors and align them with the genome. If the two anchors can be aligned to the reference genome and the order of the anchors in the read and the reference genome is opposite, then the alignment is extended.
[0119] (3) Align the full-length sequence of this read with the genome. If the alignment result contains the two alignment sequences in step (2), and the flanking sequence of the circularization site (the junction of the two reverse alignment sequences on the read) is a GT-AG conserved site, then this is considered to be a candidate circRNA.
[0120] (4) Candidate circRNAs were further screened according to the following conditions. First, the circularization site (the connection point between two reverse-aligned sequences on the read length) should be clearly located and there should be only one; second, each circRNA should be supported by at least two independent read lengths; third, the anchor point should only be able to be aligned to one position on the genome; fourth, the length of the circRNA (the number of nucleotides between the first and last nucleotides of the circularization site on the reference genome) should not exceed 100 kb; fifth, when the two read length sequences are aligned with the genome, a maximum of two mismatched bases are allowed.
[0121] The main steps of the CIRI2 algorithm are as follows:
[0122] (1) Use BWA-MEM to align each read of the RNA-seq sequence file with the reference genome. When a partial fragment of a read (fragment A) can be aligned to the genome, but another flanking fragment (fragment B) cannot be accurately aligned to the reference genome sequence, retain the read for the next step of analysis.
[0123] (2) Segment B is divided into multiple small fragments, and each small fragment is aligned to the reference genome. When the small fragment can be aligned to the reference genome and is in the opposite order to fragment A in the read length and the reference genome, it is a reverse splicing type small fragment. When the small fragment can be aligned to the reference genome and is in the same order as fragment A in the read length and the reference genome, it is counted as a forward splicing type small fragment. By adjusting the FDR threshold, the number and length of the small fragments and their mismatch rate when aligned to the reference genome can be controlled.
[0124] (3) The number of forward-splicing and reverse-splicing small fragments was counted respectively, and the maximum likelihood estimation method was used to determine whether the read length was a reverse-splicing read length.
[0125] (4) Calculate the circRNA information corresponding to the reverse splicing read length. Under the default parameters, the algorithm requires that the circRNA length (the number of nucleotides between the first and last nucleotides of the circularization site on the reference genome) is no more than 200 kb and that the flanking sequence of the circularization site (the junction of the two reverse alignment sequences on the read length) contains a GT-AG conserved site.
[0126] 2. Result verification
[0127] 2.1 Experimental verification method
[0128] 2.1.1 Main reagents and their formula or source:
[0129] (1) DNA extraction solution (formula see Table 5)
[0130] Table 5. DNA extraction solution formula
[0131] Components Final concentration Urea 7.0M Sodium chloride (NaCl) 0.3M Tris-HCl (pH 8.0) 50mM EDTA(pH8.0) 24mM Sarkosyl 1%
[0132] (2) Mitochondrial extraction buffer (formula see Table 6)
[0133] Table 6. Mitochondrial extract formula
[0134] Components Final concentration sucrose 0.3M <![CDATA[Sodium pyrophosphate (Na4P2O7)]]> 5mM <![CDATA[Potassium dihydrogen phosphate (KH2PO4)]]> 10mM Polyvinylpyrrolidone (PVP) 1%(w / v) EDTA 2mM Bovine serum albumin (BSA) 1%(w / v) Cysteine 5mM ascorbic acid 20mM
[0135] The pH was adjusted to 7.3 using H3PO4.
[0136] (3) Mitochondrial washing buffer (formula see Table 7)
[0137] Table 7. Mitochondrial washing solution formula
[0138] Components Final concentration sucrose 0.3M EGTA 1mM MOPS 10mM
[0139] The pH was adjusted to 7.2 using H3PO4.
[0140] (4) Main reagents and consumables
[0141] DNA polymerase 2x Taq Master Mix was purchased from Novozymes Biotech Co., Ltd. (Nanjing, China), RNA and DNA purification and recovery kits were purchased from Tiangen Biochemical Technology Co., Ltd. (Beijing, China), TRIzol reagent was purchased from Invitrogen, USA, PrimeScriptTMII reverse transcriptase was purchased from TaKaRa, Beijing, China, RNase R was purchased from Epicentre, USA, and Miracloth was purchased from Calbiochem, USA.
[0142] 2.1.2 Plant materials and planting environment
[0143] The genetic background of the maize (Zea mays) material was the W22 inbred line, and the ecotype of the Arabidopsis (Arabidopsis thaliana) material was Columbia. Maize was planted in the experimental field of South China Agricultural University, while Arabidopsis was planted in an artificial climate chamber (22°C, dark culture).
[0144] 2.1.3 Genomic DNA (gDNA) extraction
[0145] About 0.2 g of corn 15DAP (days after self-pollination) grains or Arabidopsis seedlings cultured in the dark for 5DAG (days after germination) were taken and placed in a mortar, and gDNA of corn and Arabidopsis was extracted using a DNA extraction solution.
[0146] 2.1.4 Enrichment of mitochondria in maize and Arabidopsis
[0147] (1) Take 10 g of maize kernels at 15 DAP or Arabidopsis seedlings at 5 DAP and place them in an ice-cold 50 ml centrifuge tube.
[0148] (2) Add 3-4 ml of mitochondrial extraction buffer to a pre-cooled mortar, pour all the tissue in the centrifuge tube into the mortar, and grind thoroughly on ice.
[0149] (3) Filter and grind the homogenate through Miracloth, collect it in an ice-cold 50-ml centrifuge tube, and use about 10 ml of mitochondrial extraction buffer to rinse the residual tissue on the Miracloth.
[0150] (4) Centrifuge at 8000 g, 4°C for 10 min.
[0151] (4) Transfer the supernatant to a new 50 ml centrifuge tube and centrifuge at 20,000 g, 4°C for 10 min.
[0152] (5) Carefully remove the supernatant; the precipitate is the mitochondria.
[0153] (6) Add 3 ml of mitochondrial washing buffer to the pellet, shake gently to mix, and divide into three 1.5 ml RNase-free centrifuge tubes. Centrifuge at 20,000 g, 4°C, for 10 min.
[0154] (7) Discard the supernatant, quickly freeze the precipitate using liquid nitrogen, and store it at -80°C for subsequent experiments.
[0155] 2.1.5 Mitochondrial RNA extraction
[0156] (1) Take one tube of mitochondrial pellet stored at -80°C obtained in step 2.1.4, add 1 ml of TriZol reagent, vortex thoroughly to mix, and place on ice for 10 min.
[0157] (2) Add 200 μl of chloroform, mix thoroughly, and place on ice for 5 min.
[0158] (3) Centrifuge at 12000 rpm for 10 min. Transfer the supernatant to a new 1.5 ml RNase-free centrifuge tube.
[0159] (4) Add an equal volume of isopropanol, mix thoroughly, and place on ice for 30 min.
[0160] (5) Centrifuge at 14,000 rpm for 15 min and remove the supernatant. The white precipitate in the tube is RNA.
[0161] (6) Wash the white precipitate twice with 80% ethanol and remove the remaining liquid with a pipette.
[0162] (7) Incubate at room temperature for 5 min and add 30 μl of RNase-free water.
[0163] (8) Use liquid nitrogen to quickly freeze the RNA and store it in a -80°C freezer for later use.
[0164] 2.1.6 Removal of gDNA contamination
[0165] DNase I was used to remove gDNA contamination in mitochondrial RNA. The reaction system is shown in Table 8.
[0166] Table 8. DNase I treatment of mitochondrial RNA
[0167] Components Final concentration DNase I 3μl 10x DNase I buffer 3μl Mitochondrial RNA 20 μg RNase-free water Add water to 30 μl
[0168] The reaction conditions were as follows: incubate at 37°C for 30 min, add 3 μl EDTA, and incubate at 75°C for 10 min to inactivate DNase I. The reaction solution was recovered using an RNA purification kit for subsequent RNase R treatment.
[0169] 2.1.7 Mitochondrial RNA delinearization
[0170] Mitochondrial RNA was treated with RNase R to remove linear RNA therein. The reaction system is shown in Table 9.
[0171] Table 9. RNase R delinearization
[0172] Components Final concentration RNase R 2μl 10x RNase R buffer 2μl Mitochondrial RNA 20 μg RNase-free water Add water to 40 μl
[0173] Reaction conditions: incubate at 37°C for 30 min. The reaction solution was recovered using an RNA purification kit for subsequent cDNA synthesis.
[0174] 2.1.8 First-strand cDNA synthesis
[0175] PrimeScriptTMII reverse transcriptase and random primers were used to synthesize two first-strand cDNAs with RNase R treated and untreated mitochondrial RNA as templates, which were named RNase+ and RNase-, respectively. The reverse transcription reaction system is shown in Table 10.
[0176] Table 10. Reverse transcription reaction system
[0177] Components Final concentration volume Random Primers 50uM 1.25μl dNTP Mix 10mM 1.25μl Template RNA 200ng 1μl RNase inhibitors 40μ / μl 0.5μl Reverse transcriptase 200μ / μl 1.25μl 5x Reverse Transcriptase Buffer 1x 5.0μl RNase-free water / 14.75 total / 25μl
[0178] Reaction conditions: incubate at 30°C for 10 min, at 42°C for 45 min, and at 70°C for 15 min. The reaction solution was directly used for PCR amplification.
[0179] 2.1.9 PCR primer design
[0180] Eleven mcircRNA production sites of maize and Arabidopsis were randomly selected, and divergent and convergent primers were designed (Table 11). Convergent primers were used as positive controls to amplify DNA and linear RNA reverse transcription products from gDNA and cDNA templates, respectively, while divergent primers amplified reverse transcription products from circRNA.
[0181] Table 11. RT-PCR primer sequences and expected fragment sizes
[0182]
[0183] 2.1.10 PCR amplification
[0184] Using the gDNA from step 2.1.2 and the RNase+ and RNase-cDNA from step 2.1.7 as templates, perform PCR amplification using 2x TaqMaster Mix reagent. The PCR reaction system and conditions are shown in Tables 12 and 13, respectively.
[0185] Table 12. PCR reaction system
[0186] PCR reaction components Final concentration volume Deionized water / 7.9μl 2x PCR master mix 1x 10.0μl Forward primer 10μM 0.8μl Reverse primer 10μM 0.8μl Reverse transcription product / 0.5μl Total / 20.0μl
[0187] Table 13. PCR reaction conditions
[0188]
[0189] 2.1.11 PCR product separation, recovery and cloning
[0190] The PCR product was purified by gel recovery and connected to the pMD18-T vector to transform Escherichia coli DH5α. The monoclonal bacteria grown after transformation were picked for PCR identification, and the monoclonal bacteria containing the expected fragment size were selected and sent to the company for sequencing.
[0191] 2.1.12 Determination of the cyclization site of PCR amplification products
[0192] Use NCBI's blastn function to compare the sequence with the reference genome. Based on the comparison results, determine whether the insert fragment of the single clone contains a circularization junction.
[0193] 2.2 PCR amplification and sequencing results analysis
[0194] To verify whether the newly predicted mitochondrial genome-encoded circRNAs by MeCi in step 1 were accurate, 11 loci from the maize and Arabidopsis mitochondrial genomes were randomly selected (maize: Zmrrn26, Zmcob, Zmnad2T2, and ZmtrnY; Arabidopsis: Atatp1, Atcox3, Atnad4L, Atrpl5, AttrnY, and AttrnD), and RT-PCR was used to amplify the circRNAs encoded by these loci ( Fig.10 and 11 ). According to the MeCi prediction results, mitochondrial genomic sites can generally produce multiple circRNA isoforms ( Fig.10 and 11 , Table 14). For example, the number of circRNA isoforms from maize Zmnad2T2 is 322, while the number of isoforms from Arabidopsis Atatp1 is as high as 982 (Table 17). In addition, the 5' ends of the circularization junctions of most mitochondrial genome-encoded circRNA isoforms are relatively close, and their main difference lies in the different positions of the 3' ends ( Fig.10 and 11 ).
[0195] Table 14. Analysis of sequencing results of RT-PCR monoclonal
[0196]
[0197] The agarose gel electrophoresis results of PCR amplification products showed that each pair of convergent primers could amplify a single fragment from gDNA, RNase-cDNA and RNase+cDNA, and the length was consistent with the predicted size. In contrast, divergent primers could only amplify circRNA from RNase- and RNase+cDNA templates, and the amplified fragments were diffuse, which was consistent with the above-mentioned circRNA isomer characteristics, that is, circRNA isomers from the same site had similar 5' ends and different 3' ends. The fragments amplified by different divergent primers were sent to the company for sequencing after PCR product recovery, vector cloning, Escherichia coli competent transformation, positive clone identification and other steps. The results of single clone sequencing and sequence alignment showed that 144 of the 226 circRNAs identified by RT-PCR were predicted by the MeCi algorithm, with a verification ratio of 63.72%. In contrast, the number of circRNAs identified by RT-PCR predicted by CIRI2 and find_circ algorithms was 0. These results indicate that most of the mitochondrial genome-encoded circRNAs newly detected by MeCi are correct, thus proving that MeCi has a high degree of credibility in predicting plant mitochondrial genome-encoded circRNAs.
[0198] In summary, the ability of the MeCi algorithm established in the present invention to identify circRNA encoded by plant mitochondrial genome is superior to existing circRNA identification methods.
[0199] The present invention has been described in detail above. It will be apparent to those skilled in the art that the present invention may be implemented in a wide range under equivalent parameters, concentrations and conditions without departing from the spirit and scope of the present invention and without the need for unnecessary experimentation. Although the present invention provides specific embodiments, it should be understood that further improvements may be made to the present invention. In short, according to the principles of the present invention, this application intends to include any changes, uses or improvements to the present invention, including changes made by conventional techniques known in the art that depart from the scope disclosed in this application. Applications of some of the basic features may be made within the scope of the following appended claims.
Claims
1. A method for identifying circular RNA encoded by plant mitochondrial genome, characterized in that: The method comprises the following steps: using a blast tool to align transcriptome sequencing data to a reference genome to obtain an alignment result file; performing sequence feature screening on the data in the alignment result file to obtain a candidate circRNA file; performing conditional screening on the candidate circRNA file to obtain a circular RNA encoded by a plant mitochondrial genome; The sequence feature screening is to remove the read lengths in which only one sequence fragment in the alignment result file is aligned to the reference genome, and retain the read lengths in which only two sequence fragments can be aligned to the reference genome in opposite directions; The sequence feature screening comprises the following steps: A1) removing the read length of only one sequence fragment aligned to the reference genome in the alignment result file, and retaining the alignment result files of the read lengths of more than two sequence fragments that can be aligned to the reference genome in opposite directions; A2) extracting the position information data of the two or more sequence fragments in the comparison result file and the positive and negative strand information data of the sequence fragments corresponding to the reference genome; A3) processing the positive and negative strand information data, retaining only the read lengths of two sequence fragments aligned to the reference genome in opposite directions, and sorting and integrating the sequence fragments according to the position information to obtain candidate circRNAs; The "E" parameter of the blast tool is set in the range of 10 -5 ~5; The screening conditions are as follows: B1) The number of overlapping nucleotides and missing nucleotides at the cyclization site should be less than or equal to 3; B2) The length of the candidate circRNA is less than or equal to 10,000 nucleotides and greater than or equal to the length of the corresponding read length; The circularization site is the connection point in the read length of two sequence fragments aligned to the reference genome in opposite directions; the overlapping nucleotides are the overlapping nucleotides at the position where two sequence fragments are aligned to the reference genome in opposite directions; the vacant nucleotides are the nucleotides that cannot be aligned to the reference genome at the connection point of two sequence fragments aligned to the reference genome in opposite directions.
2. A device for identifying circular RNA encoded by plant mitochondrial genome, characterized in that: The device comprises the following modules: C1) Sequence alignment module: used to align the transcriptome sequencing data of plant mitochondria to the reference genome using the blast tool to obtain an alignment result file; C2) Sequence feature screening module: used to perform sequence feature screening on the data in the comparison result file to obtain a candidate circRNA file; C3) Condition screening module: used to conditionally screen the candidate circRNA files to obtain the final circRNA; The sequence feature screening is established by a method comprising the following steps: removing the reads in the alignment result file in which only one sequence fragment is aligned to the reference genome, and only retaining the reads in which two sequence fragments can be aligned to the reference genome in opposite directions; The "E" parameter of the blast tool is set in the range of 10 -5 ~5; The condition screening module is established by a method comprising the following steps: B1) The number of overlapping nucleotides and missing nucleotides at the cyclization site should be less than or equal to 3; B2) The length of the candidate circRNA is less than or equal to 10,000 nucleotides and greater than or equal to the length of the corresponding read length; The circularization site is the connection point in the read length of two sequence fragments aligned to the reference genome in opposite directions; the overlapping nucleotides are the overlapping nucleotides at the positions of two sequence fragments aligned to the reference genome in opposite directions; the vacant nucleotides are the nucleotides that exist at the connection of two sequence fragments aligned to the reference genome in opposite directions and cannot be aligned to the reference genome.
3. A computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute the steps of the method according to claim 1.
Citation Information
Patent Citations
Organelle genome screening method based on total DNA sequencing results
CN107784199A
Method for predicting RNA binding protein based on circRNA position information
CN113096729A
Bioinformatics methods of in silico validation and selection of circrnas
US20200239939A1
