RNA evolution precursor screening method based on evolutionary biology and computational biology

Abstract

The invention relates to the technical field of biology, and discloses an RNA (Ribonucleic Acid) evolution precursor screening method based on evolutionary biology and computational biology, which comprises the following steps: constructing a phylogenetic tree, screening an amino acid sequence, and predicting a three-dimensional structure of a dimer formed by bacterial transcriptase. Constructing a nucleotide analogue small molecule database; docking the predicted transcriptase dimer with a nucleoside analogue, and screening small molecules with an excellent binding mode; molecules similar to original nucleotide-protein interaction are screened out to serve as potential candidate molecules; and carrying out double-chain structure prediction on the screened candidate molecules to obtain the RNA small molecule evolution precursor. According to the method, nucleoside analogues are systematically screened out by combining evolutionary biology and computational biology methods to serve as precursor candidate small molecules of RNA in the RNA world, compared with a traditional RNA precursor experimental research method, the method is more systematic, and compared with life origin exploration purely based on evolutionary biology, theoretical result support of computational biology is provided.

Description

Technical Field

[0001] The present invention relates to the field of biotechnology, and in particular to a method for screening RNA evolution precursors based on evolutionary biology and computational biology. Background Art

[0002] At the end of the last century, with the discovery of the catalytic activity of RNA ribozymes, Nobel Prize winner in Chemistry Walter Gilbert first proposed the "RNA world" hypothesis. The hypothesis holds that before the emergence of modern life, proteins and DNA may not have existed, and RNA, as a key molecule, both transmits genetic information and has catalytic functions. Further assumptions suggest that RNA itself is a product of evolution, and there was an earlier "pre-RNA world" in which nucleoside analogs played a role as carriers of genetic information. If both hypotheses are true, then the earliest transcription proteins may have retained the ability to interact with RNA evolutionary precursors, and traces of these interactions still exist in the active sites. Therefore, by studying the interactions between RNA polymerases and RNA nucleotide analogs widely sampled in the bacterial tree of life, potential nucleoside small molecules can be screened, providing a basis for exploring molecules that perform information transmission functions in the pre-RNA world.

[0003] Although existing studies have made some progress in exploring the evolutionary precursors of RNA nucleosides, there are certain limitations. Some studies focus on small molecules of nucleoside analogs that have been synthesized in the laboratory, and experimentally verify the interaction between these artificially synthesized molecules and related proteins, and further speculate on their possibility as RNA nucleoside precursors based on these experimental results. This type of research is usually supported by solid experimental data and can provide relatively direct evidence to support its arguments. However, it is worth noting that this research method often adopts a reverse reasoning method. In other words, it starts from known, artificially synthesized molecules and reversely deduces the conclusion that they may be RNA nucleoside precursors. Although this reasoning method can reveal some potential relationships to a certain extent, it lacks molecular evidence from other sources to prove the possibility of RNA nucleoside precursors. In addition, this type of research lacks a certain degree of systematicity. Since the research focus is mainly on specific small molecules of nucleoside analogs, potential precursor molecules in a wider range may be ignored. At the same time, there is often a lack of sufficient connection and integration between different studies, which makes it difficult to form a comprehensive and systematic theoretical framework to explain the evolutionary precursors of RNA nucleosides.

[0004] Another part of the research adopted the methods of evolutionary biology when exploring the structure of proteins related to the origin of life. This type of research usually focuses on constructing an evolutionary tree of proteins, and reveals their relationship and evolutionary history by comparing and analyzing homologous protein sequences in different species. On this basis, researchers will further compare the three-dimensional structures of these proteins to explore their conservation and differences in structure and function. However, this type of research often stops at constructing evolutionary trees and comparing three-dimensional structures, and lacks further experimental data support. This means that although these studies can provide valuable information about the evolutionary history of proteins, their understanding of specific protein functions, active sites, and interactions with other molecules may be relatively limited.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In response to the problems in the related art, the present invention proposes an RNA evolution precursor screening method based on evolutionary biology and computational biology to overcome the above-mentioned technical problems existing in the existing related technology.

[0007] To this end, the specific technical solution adopted by the present invention is as follows:

[0008] The RNA evolution precursor screening method based on evolutionary biology and computational biology includes the following steps:

[0009] S1. Construct a phylogenetic tree based on the amino acid sequence of bacterial transcriptase, screen the amino acid sequence of bacterial transcriptase with broad-spectrum characteristics, and predict the three-dimensional structure of bacterial transcriptase dimer;

[0010] S2. Select ribose isomers and modern nucleotide analogs, and construct a complete nucleotide analog small molecule database through base phosphate substitution and spatial isomerization change method;

[0011] S3. Use molecular docking software to dock the predicted transcriptase dimer with nucleoside analogs, predict their binding modes, and screen small molecules with excellent binding modes;

[0012] S4. Through molecular interaction fingerprint analysis technology, molecules similar to the original nucleotide-protein interaction are screened out as potential candidate molecules;

[0013] S5. Use the double-strand prediction model to predict the double-strand structure of the screened candidate molecules, and obtain the RNA small molecule evolutionary precursor based on the prediction results.

[0014] Furthermore, the method of constructing a phylogenetic tree based on the amino acid sequence of the bacterial transcriptase, screening the amino acid sequence of the bacterial transcriptase with broad-spectrum properties, and predicting the three-dimensional structure of the bacterial transcriptase forming a dimer comprises the following steps:

[0015] S11, collecting amino acid sequences of several related bacterial transcriptases, and screening to obtain a preset number of amino acid sequences of bacterial transcriptases according to sequence lengths and annotated sequences;

[0016] S12. Use multiple sequence alignment software to align the amino acid sequences of the screened proteins, and construct a phylogenetic tree based on the maximum likelihood method and evolutionary model;

[0017] S13. Screen the amino acid sequences of bacterial transcriptases with broad-spectrum properties, and use Alphafold2 to predict the three-dimensional structure model of the protein for the screened amino acid sequences.

[0018] Furthermore, there are 20,000 amino acid sequences of related bacterial transcriptases, 2,300 amino acid sequences of a preset number of bacterial transcriptases, and 80 amino acid sequences of bacterial transcriptases with broad-spectrum properties.

[0019] Furthermore, the method of using multiple sequence alignment software to align the screened protein amino acid sequences and constructing a phylogenetic tree based on the maximum likelihood method and the evolutionary model includes the following steps:

[0020] S121, using multiple sequence alignment software to align the screened protein amino acid sequences, and inputting the sequence alignment results into bioinformatics software;

[0021] S122. Select the JTT evolution model, set parameters, and run the maximum likelihood algorithm according to the set parameters to obtain a phylogenetic tree.

[0022] Furthermore, the method of screening the amino acid sequence of bacterial transcriptase with broad-spectrum characteristics and predicting the protein three-dimensional structure model of the screened amino acid sequence by Alphafold2 comprises the following steps:

[0023] S131, based on the topological structure, branch length and bootstrap support information of the phylogenetic tree, evaluate the reliability of the phylogenetic tree and the evolutionary relationship between different sequences;

[0024] S132, screening out amino acid sequences of bacterial transcriptases with broad-spectrum characteristics according to the evaluation results, and saving the screened amino acid sequences in a preset format;

[0025] S133, inputting the amino acid sequence file in a preset format into the locally running Alphafold2 environment, and using the Alphafold2 model to predict the protein three-dimensional structure model for the screened amino acid sequence.

[0026] Furthermore, the selection of ribose isomers and modern nucleotide analogs, and the construction of a complete nucleotide analog small molecule database through base phosphate substitution and spatial isomerization change method include the following steps:

[0027] S21, selecting isomers of ribose and replacing their bases and phosphate groups to generate a partial small molecule structure space;

[0028] S22. Based on the cheminformatics toolkit and combined with selected modern nucleotide analogs, a complete nucleotide analog small molecule database was constructed.

[0029] Furthermore, the method of docking the predicted transcriptase dimer with the nucleoside analog using molecular docking software, predicting their binding modes, and screening small molecules with excellent binding modes comprises the following steps:

[0030] S31. Use molecular docking software to dock the predicted transcriptase dimer with the nucleoside analog to predict their binding mode;

[0031] S32. Scoring the docking changes through a scoring function, and screening out small molecules with excellent binding modes based on the scoring results.

[0032] Furthermore, the number of search spaces in each docking process is 24.

[0033] Furthermore, the method of screening out molecules similar to the original nucleotide-protein interaction using the molecular interaction relationship fingerprint analysis technology as potential candidate molecules includes the following steps:

[0034] S41, using screening tools to identify docking conformations, and screening ligand conformations that have reasonable hydrogen bonding interactions with template DNA based on the filters;

[0035] S42, based on the protein-ligand interaction fingerprint toolkit, generates binary interaction fingerprints to identify the interactions between ligands and the amino acid residues of proteins;

[0036] S43. Compare the molecular action fingerprint generated by each small molecule with the action fingerprint of the original nucleotide, calculate the similarity between the two, and select a preset number of molecules as potential candidate molecules based on the similarity results.

[0037] Furthermore, the double-stranded structure prediction of the selected candidate molecules using the double-stranded prediction model and obtaining the RNA small molecule evolution precursor based on the prediction result analysis includes the following steps:

[0038] S51, by using the parameters of the RNA A-form helix model as a guide, wherein the parameters include a helical rise height of 2.55 angstroms, a helical twist angle of 32.69, and a tilt angle of 22.62;

[0039] S52. Based on the test results of whether nucleic acid analogs can adopt a helical conformation similar to RNA, the RNA small molecule evolutionary precursors are obtained.

[0040] The beneficial effects of the present invention are:

[0041] 1) The present invention combines evolutionary biology and computational biology methods to systematically screen out nucleoside analogs as candidate small molecules for RNA precursors in the pre-RNA world. Compared with the traditional experimental method of studying RNA precursors, this method is more systematic, and compared with the exploration of the origin of life based solely on evolutionary biology, it provides theoretical support for computational biology results.

[0042] 2) The present invention first calculates the binding energy of candidate nucleotide molecules and the bacterial transcriptase of broad-spectrum screening through molecular docking, screens out candidate molecules with lower binding energy, and analyzes its reasonable binding mode. Then, molecular fingerprint comparison is utilized to screen out candidate molecules similar to the binding mode of the original nucleoside micromolecule. Finally, the energy and possibility of these candidate molecules to build double-stranded structures are calculated and predicted, thereby screening out the micromolecules most suitable as RNA nucleoside precursors. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 is a flow chart of a method for screening RNA evolution precursors based on evolutionary biology and computational biology according to an embodiment of the present invention;

[0045] Figure 2 It is a schematic diagram of the principle of the RNA evolution precursor screening method based on evolutionary biology and computational biology according to an embodiment of the present invention;

[0046] Figure 3 It is a technical roadmap for the RNA evolution precursor screening method based on evolutionary biology and computational biology according to an embodiment of the present invention. DETAILED DESCRIPTION

[0047] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0048] According to an embodiment of the present invention, a method for screening RNA evolution precursors based on evolutionary biology and computational biology is provided.

[0049] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1-Figure 3 As shown, the RNA evolution precursor screening method based on evolutionary biology and computational biology according to an embodiment of the present invention comprises the following steps:

[0050] S1. Construct a phylogenetic tree based on the amino acid sequence of bacterial transcriptase, screen the amino acid sequence of bacterial transcriptase with broad-spectrum characteristics, and predict the three-dimensional structure of bacterial transcriptase dimer;

[0051] Specifically, suitable protein sequences are screened and three-dimensional models are constructed. First, relevant bacterial transcriptase amino acid sequences are retrieved on UniProt, a public protein amino acid sequence website, to obtain 20,000 transcriptase amino acid sequences. Then, 2,300 amino acid sequences are screened out based on the sequence length (β chain length is not less than 600 amino acids and not more than 1,600 amino acids) and the annotation of the sequence. Next, multiple sequence alignment software is used to align the screened protein amino acid sequences, and a powerful bioinformatics software (such as MEGA, Molecular Evolutionary Genetics Analysis) is used to further use the maximum likelihood (ML) algorithm and a specific evolutionary model to construct a phylogenetic tree. The specific steps include: inputting the results of the sequence alignment into the bioinformatics software, selecting the JTT (Jones-Taylor-Thornton) evolutionary model, setting parameters, running the maximum likelihood algorithm, and allowing the bioinformatics software to calculate based on the provided sequence data and evolutionary model. Then, after the algorithm is run, the results of the phylogenetic tree are obtained. The topological structure, branch length and bootstrap support rate of the tree were carefully checked to evaluate the reliability of the tree and the evolutionary relationship between different sequences. Considering the practical limitations of the computational cost, it is impossible to perform detailed three-dimensional structure prediction and molecular docking analysis on the amino acid sequences of all 2300 bacterial transcriptases. Therefore, at a broad spectrum level, this embodiment carefully selected 80 representative bacterial transcriptase amino acid sequences. These sequences not only cover all the major branches and types in the bacterial transcriptase evolutionary map, but also can fully reflect the evolutionary relationship and structural characteristics between different bacterial transcriptases. Finally, Alphafold2 was further used to predict the protein three-dimensional structure model of the 80 protein amino acid sequences screened out. The specific steps include: saving the protein amino acid sequence screened out as FASTA format, submitting the prepared FASTA file to the locally running Alphafold2 environment, selecting the multimer module to model rpoβ and rpoβ' at the same time, selecting the prediction model number to be up to 5 times, and predicting that the TM score is not less than 0.85. The Alphafold2 model predicts the three-dimensional structure of the screened proteins through multiple modules such as multiple sequence alignment, convolutional neural network, and spatial conformation prediction.

[0052] S2. Select ribose isomers and modern nucleotide analogs, and construct a complete nucleotide analog small molecule database through base phosphate substitution and spatial isomerization change method;

[0053] Specifically, a suitable nucleotide analogue small molecule database is screened. A nucleotide is a three-part structure consisting of a ribosome + a phosphate + a base. In the process of constructing a nucleotide analogue database, 227 ribose isomers + 13 modern nucleotide analogues are selected in this embodiment, and further through base phosphate substitution and spatial isomerization changes (generated by the EmbedMultipleConfs function under RDKit, which is a very common and powerful chemical informatics toolkit), a nucleotide analogue small molecule database of 5964 is formed.

[0054] S3. Use molecular docking software to dock the predicted transcriptase dimer with nucleoside analogs, predict their binding modes, and screen small molecules with excellent binding modes;

[0055] Specifically, molecular docking. In the key step of molecular docking, the present embodiment uses docking software (such as SMINA software) for processing. The essence of molecular docking is to combine protein models with small molecules, and then the small molecules are dynamically changed by rotating, adjusting the structure, etc. on the active site. The docking software uses a specific empirical function algorithm to score these changes, and the lower the binding energy is, the higher the score is, which represents that the ligand is more closely combined with the target. In order to further improve the accuracy of docking, a custom scoring function is specially designed and optimized in the present embodiment. The core purpose of this function is to accurately capture the interaction between hydroxyl and magnesium ion, which is very important for screening nucleotide candidates that can produce more accurate interactions with magnesium ion active sites. In this docking process, the number of search spaces in each docking process is set to 24, that is, for each small molecule, 24 different conformations are generated to dock with proteins. Considering that there are 5964 small molecules and 80 protein models, this embodiment will eventually obtain a huge data set, including 5964 (number of small molecules) × 24 (number of conformations of each small molecule) × 80 (number of protein models) = million-level data points. Faced with such a huge data set, screening is particularly important. In this embodiment, a preliminary screening is first performed based on the scoring scores, and the molecules with the top 20% scores are selected as the candidate molecule set after the first step of screening. This step aims to screen out the molecules that are most likely to effectively bind to the target from the massive data, providing strong support for subsequent research.

[0056] The scoring function in this embodiment is an empirical scoring function formula provided by SMINA software: Docking_score = -0.035579*gauss1–0.005156*gauss2+0.840245*repulsion–0.035069*hydrophobic–0.587439*non_dir_h_bond, two Gaussian terms, one repulsion term, one hydrophobic term, and one non-directional hydrogen bond term, where Docking_score is the docking score, which represents the output result of the entire formula, representing the docking of the ligand and the receptor. The comprehensive evaluation score after the reaction is obtained. Gauss1 and gauss2 are two quantities related to the Gaussian function, which are used to simulate the interaction potential energy between atoms. Repulsion is the repulsion term, which represents the repulsion between the ligand and the receptor. When the atoms of the two are too close in space, the overlapping electron clouds will produce repulsion. Hydrophobic is the hydrophobic term, which represents the hydrophobic interaction between the ligand and the receptor. If the hydrophobic term is relatively small, it means that the hydrophobic interaction between the ligand and the receptor is good, which is conducive to binding. Non_dir_h_bond is the non-directional hydrogen bond term, which is used to quantify the contribution of hydrogen bonds to ligand-receptor binding.

[0057] S4. Through molecular interaction fingerprint analysis technology, molecules similar to the original nucleotide-protein interaction are screened out as potential candidate molecules;

[0058] Specifically, molecular action mode screening. Molecules with higher docking scores often have relatively reasonable binding conformations, but sometimes structures with lower energy are not the conformations when chemical reactions occur. It is more valuable to screen small molecules that have a protein binding mode that is closer to the original molecule (NTP). In this embodiment, LigGrep is first used as a screening tool (a free and open source program that is specifically used to screen and identify docking postures that match user-specified receptor / ligand interactions) to identify the correct docking conformation, that is, nucleotides need to transmit genetic information and need to produce base complementary pairing (hydrogen bond connection) with the corresponding DNA. The screened conformation requires that it can produce hydrogen bonds with the corresponding DNA base. A user-specified filter (written in JSON format) is designed to specifically screen ligand conformations that have reasonable hydrogen bond interactions with template DNA. Its format is as follows (in simple terms, the distance between the H atom of the specified DNA base and the O atom of the candidate small molecule is less than 4 angstroms):

[0059]

[0060] Then, a protein-ligand interaction fingerprint toolkit (e.g., ProLIF, Protein-Ligand Interaction Fingerprint) is used to generate a binary interaction fingerprint (a list of 0 and 1) to identify the interactions between the ligand and the amino acid residues of the protein, such as hydrogen bonds, hydrophobic interactions, and electrostatic interactions. The molecular action fingerprint generated for each small molecule is then compared with the action fingerprint of the original nucleotide, and the similarity between the two is calculated. The molecules with the top 20% similarity and that pass the LigGrep screening are further screened to enter the third round of screening.

[0061] In this embodiment, TC similarity (Tanimoto coefficient similarity) is used to compare the similarity of two molecular fingers, and the calculation formula is as follows:

[0062]

[0063] Where TC(A,B) represents the similarity between two molecules, A and B represent the two molecular fingerprints as two vectors, A·B represents the dot product of the vectors, and |A| and |B| represent the modulus of the vectors.

[0064] S5. Use the double-strand prediction model to predict the double-strand structure of the screened candidate molecules, and obtain the RNA small molecule evolutionary precursor based on the prediction results.

[0065] Specifically, double-stranded structure model prediction. What we want to screen is RNA nucleotide precursors, one of its biggest functions is to retain the ability to transmit genetic information, so it is necessary to be able to form a relatively stable double-stranded structure. Therefore, in this embodiment, the small molecules screened out in the fourth step are calculated for double-stranded structure model prediction. The double-stranded helix model of the screened nucleic acid analogs is constructed using a calculation package for the double-stranded system of nucleotide analogs (such as pNAB, The proto-Nucleic-Acid Builder). By using the parameters of the RNA A-type helix model (helix rise height 2.55 angstroms, helix torsion angle 32.69, tilt angle 22.62) as a guide, it is tested whether nucleic acid analogs can adopt a helical conformation similar to RNA. Finally, the molecules screened out in this embodiment are candidates for RNA small molecule evolutionary precursors that can form a stable double-stranded structure and have the potential to retain the ability to transmit genetic information.

[0066] In summary, with the help of the above technical scheme of the present invention, the present invention innovatively integrates the cutting-edge technologies of evolutionary biology and computational biology, and combines some characteristics of RNA nucleotides to construct a comprehensive and targeted RNA precursor screening method. First, by systematically collecting the evolutionary data of bacterial transcriptases and constructing their evolutionary trees, the three-dimensional structure modeling of the widely screened amino acid sequences is performed using advanced protein structure prediction technology. This step not only lays a solid evolutionary biology foundation for subsequent molecular docking experiments, but also ensures the breadth and depth of the screened molecules at the evolutionary level. Further, in the field of computational biology, traditional molecular docking technology is used to perform energy binding calculations on the screened broad-spectrum protein models and nucleotide analog small molecule databases. For the specific binding characteristics of RNA nucleotides and transcriptases, molecular fingerprint screening and double-stranded model prediction model screening methods are further designed to achieve refined screening of small molecules. Through this three-layer screening mechanism, the successfully screened small molecules not only show strong binding ability in a wide spectrum (with good average docking scores for more than 80 transcriptases), but also their binding mode between molecules and proteins is highly similar to that of the original nucleotides (NTP) and proteins (similar molecular fingerprints), and can form a stable double-stranded structure (double-stranded structure has low energy). This innovative screening method greatly improves the screening efficiency and accuracy of RNA precursors, and provides a new scientific basis for the discovery and application of RNA precursors.

[0067] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for screening RNA evolution precursors based on evolutionary biology and computational biology, characterized in that: The following steps are involved: S1. Construct a phylogenetic tree based on the amino acid sequence of bacterial transcriptase, screen the amino acid sequence of bacterial transcriptase with broad-spectrum characteristics, and predict the three-dimensional structure of bacterial transcriptase dimer; S2. Select ribose isomers and modern nucleotide analogs, and construct a complete nucleotide analog small molecule database through base phosphate substitution and spatial isomerization change method; S3. Use molecular docking software to dock the predicted transcriptase dimer with nucleoside analogs, predict their binding modes, and screen small molecules with excellent binding modes; S4. Through molecular interaction fingerprint analysis technology, molecules similar to the original nucleotide-protein interaction are screened out as potential candidate molecules; S5. Use the double-strand prediction model to predict the double-strand structure of the screened candidate molecules, and obtain the RNA small molecule evolutionary precursor based on the prediction results.

2. The RNA evolution precursor screening method based on evolutionary biology and computational biology according to claim 1, characterized in that: The method of constructing a phylogenetic tree according to the amino acid sequence of the bacterial transcriptase, screening the amino acid sequence of the bacterial transcriptase with broad-spectrum characteristics, and predicting the three-dimensional structure of the bacterial transcriptase forming a dimer comprises the following steps: S11, collecting amino acid sequences of several related bacterial transcriptases, and screening to obtain a preset number of amino acid sequences of bacterial transcriptases according to sequence lengths and annotated sequences; S12. Use multiple sequence alignment software to align the amino acid sequences of the screened proteins, and construct a phylogenetic tree based on the maximum likelihood method and evolutionary model; S13. Screen the amino acid sequences of bacterial transcriptases with broad-spectrum properties, and use Alphafold2 to predict the three-dimensional structure model of the protein for the screened amino acid sequences.

3. The RNA evolution precursor screening method based on evolutionary biology and computational biology according to claim 2, characterized in that: There are 20,000 amino acid sequences of related bacterial transcriptases, 2,300 amino acid sequences of a preset number of bacterial transcriptases, and 80 amino acid sequences of bacterial transcriptases with broad-spectrum properties.

4. The RNA evolution precursor screening method based on evolutionary biology and computational biology according to claim 2, characterized in that: The method of using multiple sequence alignment software to align the screened protein amino acid sequences and constructing a phylogenetic tree based on the maximum likelihood method and the evolutionary model includes the following steps: S121, using multiple sequence alignment software to align the screened protein amino acid sequences, and inputting the sequence alignment results into bioinformatics software; S122. Select the JTT evolution model, set parameters, and run the maximum likelihood algorithm according to the set parameters to obtain a phylogenetic tree.

5. The RNA evolution precursor screening method based on evolutionary biology and computational biology according to claim 2, characterized in that: The method of screening the amino acid sequence of a bacterial transcriptase with broad-spectrum characteristics and predicting the protein three-dimensional structure model of the screened amino acid sequence by Alphafold2 comprises the following steps: S131, based on the topological structure, branch length and bootstrap support information of the phylogenetic tree, evaluate the reliability of the phylogenetic tree and the evolutionary relationship between different sequences; S132, screening out amino acid sequences of bacterial transcriptases with broad-spectrum characteristics according to the evaluation results, and saving the screened amino acid sequences in a preset format; S133, inputting the amino acid sequence file in a preset format into the locally running Alphafold2 environment, and using the Alphafold2 model to predict the protein three-dimensional structure model for the screened amino acid sequence.

6. The RNA evolution precursor screening method based on evolutionary biology and computational biology according to claim 1, characterized in that: The selection of ribose isomers and modern nucleotide analogs, and the construction of a complete nucleotide analog small molecule database through base phosphate substitution and spatial isomerization change method include the following steps: S21, selecting isomers of ribose and replacing their bases and phosphate groups to generate a partial small molecule structure space; S22. Based on the cheminformatics toolkit and combined with selected modern nucleotide analogs, a complete nucleotide analog small molecule database was constructed.

7. The RNA evolution precursor screening method based on evolutionary biology and computational biology according to claim 1, characterized in that: The method of docking the predicted transcriptase dimer with the nucleoside analog using molecular docking software, predicting their binding modes, and screening small molecules with excellent binding modes comprises the following steps: S31. Use molecular docking software to dock the predicted transcriptase dimer with the nucleoside analog to predict their binding mode; S32. Scoring the docking changes through a scoring function, and screening out small molecules with excellent binding modes based on the scoring results.

8. The RNA evolution precursor screening method based on evolutionary biology and computational biology according to claim 7, characterized in that: The number of search spaces in each docking process is 24.

9. The RNA evolution precursor screening method based on evolutionary biology and computational biology according to claim 1, characterized in that: The method of screening out molecules similar to the original nucleotide-protein interaction using the molecular interaction relationship fingerprint analysis technology and using them as potential candidate molecules includes the following steps: S41, using screening tools to identify docking conformations, and screening ligand conformations that have reasonable hydrogen bonding interactions with template DNA based on the filters; S42, based on the protein-ligand interaction fingerprint toolkit, generates binary interaction fingerprints to identify the interactions between ligands and the amino acid residues of proteins; S43. Compare the molecular action fingerprint generated by each small molecule with the action fingerprint of the original nucleotide, calculate the similarity between the two, and select a preset number of molecules as potential candidate molecules based on the similarity results.

10. The RNA evolution precursor screening method based on evolutionary biology and computational biology according to claim 1, characterized in that: The double-stranded structure prediction model is used to predict the double-stranded structure of the selected candidate molecules, and the RNA small molecule evolution precursor is obtained based on the prediction result analysis. The following steps are involved: S51, by using the parameters of the RNA A-form helix model as a guide, wherein the parameters include a helical rise height of 2.55 angstroms, a helical twist angle of 32.69, and a tilt angle of 22.62; S52. Based on the test results of whether nucleic acid analogs can adopt a helical conformation similar to RNA, the RNA small molecule evolutionary precursors are obtained.

Images