CN113380320A - Molecular docking result screening method based on positive compound residue contribution similarity - Google Patents
Molecular docking result screening method based on positive compound residue contribution similarity Download PDFInfo
- Publication number
- CN113380320A CN113380320A CN202110748005.2A CN202110748005A CN113380320A CN 113380320 A CN113380320 A CN 113380320A CN 202110748005 A CN202110748005 A CN 202110748005A CN 113380320 A CN113380320 A CN 113380320A
- Authority
- CN
- China
- Prior art keywords
- compound
- energy
- contribution
- target
- residue
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B15/00—ICT specially adapted for analysing two-dimensional or three-dimensional molecular structures, e.g. structural or functional relations or structure alignment
- G16B15/30—Drug targeting using structural data; Docking or binding prediction
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C10/00—Computational theoretical chemistry, i.e. ICT specially adapted for theoretical aspects of quantum chemistry, molecular mechanics, molecular dynamics or the like
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/50—Molecular design, e.g. of drugs
Landscapes
- Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Medicinal Chemistry (AREA)
- Pharmacology & Pharmacy (AREA)
- Crystallography & Structural Chemistry (AREA)
- Biophysics (AREA)
- Medical Informatics (AREA)
- Evolutionary Biology (AREA)
- Biotechnology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention relates to a molecular docking result screening method based on positive compound residue contribution similarity, which belongs to the technical field of drug screening, and comprises the steps of constructing a positive compound library, a target spot, optimizing a compound structure and optimizing a docking result screening method; the method increases the screening probability of the active compound in the process of finding the potential patent medicine compound, further fundamentally improves the precision of the screening method, improves the accuracy of screening the potential positive compound, and greatly saves the calculation cost and the time cost.
Description
Technical Field
The invention belongs to the technical field of drug screening, and particularly relates to a large-scale molecular docking result screening method based on positive compound residue contribution similarity.
Background
Molecular docking is a typical approach for structure-based drug design. The method utilizes a computer to obtain molecules which can be matched with a specific drug action target in geometry and chemistry from a three-dimensional structure database, and determines the most favorable binding conformation thereof, thereby realizing computer-assisted drug screening. With the increasing size of ligand databases, molecular docking has become one of the most computationally intensive and data intensive scientific applications in structure-based drug discovery. Molecular docking is generally used as a primary screening process for virtual screening, and a few ligands obtained by screening enter subsequent more complex screening steps such as molecular dynamics
Due to the complexity and diversity of research systems and the insufficiency of molecular docking systems in scoring functions, the accuracy of screening strategies is poor. At present, the problem of relatively accurate screening of active compounds is difficult to solve. In large-scale molecular docking calculations, selection of a screening strategy is crucial, limited to the computational speed problem of molecular dynamics simulation. A good screening strategy would allow accurate screening for false positive compounds, leaving a small number of possible active compounds for further screening and testing.
In order to improve the screening precision of the molecular docking result, the screening process is optimized by adopting molecular dynamics and energy analysis technology. Molecular dynamics simulation relies primarily on newton mechanics to more accurately simulate the motion of a molecular system, to extract samples in an ensemble consisting of different states of the molecular system, to calculate the configuration integral of the system, and to further calculate thermodynamic quantities and other macroscopic properties of the system based on the results of the configuration integral. Molecular dynamics simulation is an important computer simulation method for solving a multi-body problem at the atomic and molecular level, and can predict the dynamics characteristics on the nanometer scale. Molecular dynamics simulation can be used to simulate the fundamental process associated with the atomic motion path by solving the equations of motion for all particles.
To further understand the binding energy of the residues in the protein and the small molecules during molecular dynamics, etc., the results are analyzed for energy. By strictly decomposing the binding free energy into the original contributions from different atom or interaction types, a more accurate understanding of the contribution of the residues to the respective energy components of the binding can be obtained.
Although the current molecular docking result screening method can realize rapid large-scale screening, the screening precision is still insufficient.
1. Traditional screening of molecular docking results typically uses scoring as a single screening criterion. However, the scoring function needs to evaluate a large number of docking postures and rank different ligand configurations, and in consideration of computational efficiency, an approximate value scoring function is introduced in most docks, which generally affects the accuracy of prediction.
2. In order to improve the screening precision, the traditional screening method usually adopts a secondary scoring (consistency scoring) mode to eliminate false positives. By performing multiple docking calculations on a large-scale ligand library using different docking systems, the hit rate of the compound is improved, but the calculation cost is also increased.
The molecular docking results need to be analyzed after docking is complete, which is also the most important part of doing molecular docking. The existing molecular docking theory is not perfect, and in addition, the docking procedures are numerous and have different performances, and the success rate in actual use is often different. The molecular docking using computers is simple, the rapid development of supercomputer technology at the present stage, and the computing power of supercomputers is multiplied. Although the dataset of targets, compounds docked is in the order of millions, today's calculations can be done in a shorter time. But it is difficult to efficiently, accurately and reliably screen the results after docking. The method for screening the docking results is uncertain, the screening standards are various, and how to analyze the obtained data depends on the experience of the individual. The results obtained are very different, limited by the limitations of the breadth of the knowledge level and the short panel of personal experimental experience. The success rate of screening for a highly differentiated compound using such results is low. Therefore, how to improve the success rate by establishing and optimizing a screening scheme is an important research topic in virtual screening based on molecular docking.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a molecular docking result screening method based on positive compound residue contribution similarity, which improves the screening proportion of active compounds and accelerates the research and development progress of new drugs by constructing a positive compound library of a specific target spot, optimizing the target spot and the compound structure and optimizing the docking result screening method.
In order to solve the technical problems, the invention adopts the technical scheme that:
a molecular docking result screening method based on positive compound residue contribution similarity comprises the steps of constructing a positive compound library, a target point, optimizing a compound structure and optimizing a docking result screening method;
the positive compound library is constructed by the following steps: obtaining a crystal structure of a required target from a public and authoritative pharmaceutical data source, selecting a compound with good binding degree measured by multiple experiments according to the binding force condition of the target and a ligand compound obtained by different experimental methods disclosed by a database, and separating the crystal structure by using a molecular visualization technology to obtain the structural information of a positive compound;
the optimized target and ligand compound structure is as follows: firstly, checking an initial structure of a target spot of a ligand compound to be screened and carrying out hydrotreating; secondly, generating an accurate 3D molecular model aiming at the original structure of the ligand compound, carrying out format conversion on the 3D structure, and adding charges and required parameter items;
the optimization of the screening method of the docking results comprises the following steps: firstly, carrying out butt joint calculation on the optimized target spot and the ligand compound;
secondly, respectively carrying out contribution factors of the positive compound and the compound to be tested on the result after the butt joint calculation; obtaining contribution factors of residues in different compound-target complexes by energy decomposition MM/PBSA treatment;
energy breakdown, i.e., MM/PBSA, is the calculation of the degree of affinity of the binding between ligand and receptor, and the principle approach is to break down the energy driving the binding of both to every amino acid residue on the target in order to clearly see the specific energy contribution of each amino acid to the ligand binding, including VDW, solvation energy, electrostatic energy, etc.
Firstly, constructing a contribution factor of an existing positive compound aiming at a specific target. To obtain the main contributors to the compound, post-processing work on the docking results is required: adding a force field parameter required by molecular dynamics to a compound to generate a ligand protein complex;
due to the small number of positive compounds, all the contributing factors are obtained through steps of molecular dynamics simulation and energy decomposition. The energy of decomposition includes mainly molecular mechanical energy, potential energy in vacuum, van der waals and coulomb electrostatic interaction energy, polar solvation energy and non-polar solvation energy. Determining a main contribution factor according to the contribution value of the total energy of all the contribution factors;
secondly, for large-scale co-target molecule docking results, in order to improve the screening speed, after post-treatment, energy minimization treatment and energy decomposition are carried out on the co-target molecule docking results to obtain corresponding main contribution factors;
finally, the similarity of the residue contributing factors was analyzed. After the energy decomposition of the previous step we get the data of the contribution of each disability to the overall binding energy. A comparison method of hot spot residue energy variance is adopted in the process of residue similarity analysis, namely residue comparison. Since only a small fraction of the residues are decisive for the stability of the protein-ligand complex, the residues with a high contribution coefficient are used here as hot-spot residues for comparison with the residues of the protein-ligand complex to be determined. Sequencing energy contribution coefficients of residues after energy decomposition aiming at the butt joint result of a large-scale compound and a target spot, and obtaining a residue set with large stability contribution factors through a formula 1; rank (energy) indicates ranking by residue energy contribution;
r rank (energy) formula-1
The similarity of the residue in the compound-target complex to be measured and the same residue in the positive compound-target complex can be obtained through a formula 2; wherein EiRepresenting the amount of contributing energy of the test compound to the residue in the target complex; ejRepresenting the amount of contributing energy of the positive compound to the residue in the target complex; RMSE is the coefficient of similarity of contributions from residues, the smaller the RMSE the greater the similarity of contributions from residues, the larger the RMSE the smaller the similarity of contributions from residues;
if the RMSE does not reach the set threshold, the druggability of the compound is quite marginal. If the RMSE reaches a set threshold, the similarity of the residues of the test compound-target complex and the positive compound-target complex remains a high match, so the results are screened for further drug discovery.
After molecular docking for large scale compound advancement, high throughput molecular dynamics simulation was combined with binding energy calculation using tools of high throughput calculation while estimating the contribution of each residue to the binding energy. And (3) for the obtained residue energy data, extracting by using a script tool according to the binding energy contribution of the final hot spot residue, and performing prediction classification on the molecular docking results by setting different similarity thresholds.
Compared with the prior art, the invention has the beneficial effects that:
(1) the invention adopts the main residue contribution of the positive compound of a specific target as the screening reference basis of the docking result of the same target. The method can evaluate the interaction between the compound and the positive compound target from different depths and extents by taking the docking scoring and the binding affinity as the reference basis for screening candidate compounds, and exerts the maximum advantage of each method. The screening probability of the active compound is increased in the process of finding the potential patent medicine compound, so that the precision of the screening method is fundamentally improved, and the accuracy of screening the potential positive compound is improved.
(2) The invention uses the result of one-time docking, and eliminates the time waste and the computing resource waste caused by adopting a secondary scoring (consistency) mode to eliminate false positive compounds. Compared with the traditional screening method, the method greatly saves the calculation cost and the time cost.
Drawings
FIG. 1 is a block diagram of a method for screening docking results based on the contribution of hot spot residues of positive compounds;
FIG. 2A, 2D structure of positive compound; B. 3D structure of positive compound;
FIG. 3 detailed structural information data for positive compounds; because of the excessive structural data of the selected positive compounds, the data information of only three elements in the picture is listed, and "…" represents the rest of the element information, which is not listed in the whole.
FIG. 4 is a flow chart of a molecular dynamics simulation program;
FIG. 5 is a residue similarity comparison screening model;
fig. 6 is an energy decomposition illustration.
Detailed Description
The present invention will be further described with reference to specific embodiments thereof, it being understood that the embodiments described are only a few, and not all, of the embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
A molecular docking result screening method based on positive compound residue contribution similarity is disclosed, a flow chart is shown in figure 1, and the specific steps are as follows:
1) construction of positive Compound library of EGFR target
The crystal structure of the EGFR target is obtained from a protein database, the database also provides the binding force condition of the target and a ligand compound obtained by Kd, IC50, Ki and other experimental methods of the crystal structure, and a compound with nanomolar binding degree and smaller experimental value is selected. The crystal structures were separated using molecular visualization tools to obtain structural information data for positive compounds, as shown in fig. 2 and 3. And (4) converting the obtained positive compound structure data into a format meeting the requirements by using a format conversion tool according to the requirements of docking software.
2) Structural optimization of targets and compounds
In order to improve the precision of molecular docking result screening, target spots and compounds are optimized. First, the target is hydrotreated. Secondly, aiming at the original structure of the compound database, a three-dimensional molecular model is generated efficiently and accurately by using a conversion tool, the three-dimensional structure is subjected to format conversion, charges are added, and the total number of all atomic charges and required parameter items are calculated.
3) Optimization of docking result screening method
Firstly, the optimized target is respectively butt-jointed with a positive compound and a compound in a target compound library for calculation. Residue contributing factors for positive compounds were constructed against the EGFR target. To obtain the main contributing factors, the post-processing work needs to be performed on the docking results: first, the force field parameters required for molecular dynamics simulation are added to the compound, as shown in fig. 4; endowing the receptor with a corresponding force field, adding a water solvent and ions to balance the ions, and finally forming a complex by the protein, the small molecules and the solvent. Due to the small number of positive compounds, all the contributing factors are obtained through complete molecular dynamics simulation and energy decomposition steps. The energy of decomposition includes mainly molecular mechanical energy, potential energy in vacuum, van der waals and coulomb electrostatic interaction energy, polar solvation energy and non-polar solvation energy. The main contributing factor is determined among all contributing factors based on the contribution of the total energy thereof, and residues with smaller total energy contributions are considered as the residues of main action. For the large-scale co-target molecule docking result, in order to improve the screening speed, after post-treatment, energy minimization treatment and energy decomposition are carried out on the co-target molecule docking result, and corresponding main contribution factors are obtained.
The optimization of the screening method of the docking results comprises the following steps: firstly, carrying out butt joint calculation on the optimized target spot and the ligand compound;
secondly, respectively carrying out contribution factors of the positive compound and the compound to be tested on the result after the butt joint calculation; the contribution factors of the residues in different compound-target complexes can be obtained through energy decomposition MM/PBSA treatment;
energy breakdown, i.e., MM/PBSA, is the calculation of the degree of affinity of the binding between ligand and receptor, and the main principle approach is to break down the energy driving the binding of both to every amino acid residue on the target in order to clearly see the specific energy contribution of each amino acid to the ligand binding, including VDW, solvation energy, electrostatic energy, etc.
Firstly, constructing a contribution factor of an existing positive compound aiming at a specific target. To obtain the main contributors to the compound, post-processing work on the docking results is required: adding a force field parameter required by molecular dynamics to a compound to generate a ligand protein complex;
due to the small number of positive compounds, all the contributing factors are obtained through steps of molecular dynamics simulation and energy decomposition. The energy of decomposition includes mainly molecular mechanical energy, potential energy in vacuum, van der waals and coulomb electrostatic interaction energy, polar solvation energy and non-polar solvation energy. Determining a main contribution factor according to the contribution value of the total energy of all the contribution factors;
secondly, for large-scale co-target molecule docking results, in order to improve the screening speed, after post-treatment, energy minimization treatment and energy decomposition are carried out on the co-target molecule docking results to obtain corresponding main contribution factors;
finally, the similarity of the residue contributing factors was analyzed, and as shown in FIG. 5, after the previous energy decomposition, we obtained data on the contribution of each disability to the overall binding energy. A comparison method of hot spot residue energy variance is adopted in the process of residue similarity analysis, namely residue comparison. Since only a small fraction of the residues are decisive for the stability of the protein-ligand complex, the residues with a high contribution coefficient are used here as hot-spot residues for comparison with the residues of the protein-ligand complex to be determined. And aiming at the butt joint result of the large-scale compound and the target, sequencing energy contribution coefficients of residues after energy decomposition, and obtaining a residue set with a large stability contribution factor through a formula 1. Rank (energy) indicates ranking by residue energy contribution;
r rank (energy) formula-1
The similarity of the residue in the compound-target complex to be measured and the same residue in the positive compound-target complex can be obtained through a formula 2; it is composed ofIn EiRepresenting the contribution energy value l of the test compound to the residue in the target complex; ejRepresenting the amount of contributing energy of the positive compound to the residue in the target complex; RMSE is the similarity of contributions of residues, the smaller the RMSE, the greater the similarity of contributions of residues, the larger the RMSE, the smaller the similarity of contributions of residues;
when the RMSE reaches a set threshold, we set a threshold of 0.96 in this example, to achieve a higher similarity with the positive compound, where not only does the similarity of the compound to the positive compound residue maintain a high match, but the energy contribution of the residue is also maximal. The druggability of this compound is quite marginal if the set threshold is not reached.
Finally, compounds in the target compound library that have a similarity threshold of 96% to the main effect residue contribution of the positive compound are screened out. The screening specific gravity of the positive compound is improved by about 50 percent. Fig. 6 is an exemplary energy decomposition diagram.
Claims (2)
1. A molecular docking result screening method based on positive compound residue contribution similarity is characterized in that the method comprises the steps of constructing a positive compound library, constructing a target, optimizing a compound structure and optimizing a docking result screening method;
the positive compound library is constructed by the following steps: obtaining a crystal structure of a required target from a public and authoritative pharmaceutical data source, selecting a compound with good binding degree measured by multiple experiments according to the binding force condition of the target and a ligand compound obtained by different experimental methods disclosed by a database, and separating the crystal structure by using a molecular visualization technology to obtain the structural information of a positive compound;
the optimized target and ligand compound structure is as follows: firstly, checking an initial structure of a target spot of a ligand compound to be screened and carrying out hydrotreating; secondly, generating an accurate 3D molecular model aiming at the original structure of the ligand compound, carrying out format conversion on the 3D structure, and adding charges and required parameter items;
the optimization of the screening method of the docking results comprises the following steps: firstly, carrying out butt joint calculation on the optimized target spot and the ligand compound;
secondly, respectively carrying out contribution factors of the positive compound and the compound to be tested on the result after the butt joint calculation; obtaining contribution factors of residues in different compound-target complexes by energy decomposition MM/PBSA treatment;
finally, the similarity of the residue contributing factors is analyzed; after the energy decomposition of the previous step is carried out, the value data of the contribution of each disability to the whole binding energy is obtained; a comparison method of hot spot residue energy variance is adopted in the process of residue similarity analysis, namely residue comparison; since only a small fraction of residues are decisive for the stability of the protein-ligand complex, the residues with a high contribution coefficient are taken as hot-spot residues for comparison reference with the residues of the protein-ligand complex to be tested; sequencing energy contribution coefficients of residues after energy decomposition aiming at the butt joint result of a large-scale compound and a target spot, and obtaining a residue set with large stability contribution factors through a formula 1; rank (energy) indicates ranking by residue energy contribution;
r rank (energy) formula-1
The similarity of the residue in the compound-target complex to be measured and the same residue in the positive compound-target complex can be obtained through a formula 2; wherein EiRepresenting the amount of contributing energy of the test compound to the residue in the target complex; ejRepresenting the amount of contributing energy of the positive compound to the residue in the target complex; RMSE is the coefficient of similarity of contributions from residues, the smaller the RMSE the greater the similarity of contributions from residues, the larger the RMSE the smaller the similarity of contributions from residues;
if the RMSE does not reach the set threshold, the druggability of the compound is quite minimal; if the RMSE reaches a set threshold, the similarity of the residues of the test compound-target complex and the positive compound-target complex remains a high match, so the results are screened for further drug discovery.
2. The method of claim 1, wherein the energy resolution (MM/PBSA) is used to calculate the degree of affinity of the binding between the ligand and the receptor by the following steps:
constructing contribution factors of existing positive compounds aiming at specific targets; to obtain the main contributors to the compound, post-processing work on the docking results is required: adding a force field parameter required by molecular dynamics to a compound to generate a ligand protein complex;
for a small number of positive compounds, all the contributing factors are obtained through steps including molecular dynamics simulation and energy decomposition; the energy of decomposition includes molecular mechanical energy, potential energy in vacuum, van der waals and coulomb electrostatic interaction energy, polar solvation energy and non-polar solvation energy; determining a main contribution factor according to the contribution value of the total energy of all the contribution factors;
secondly, for large-scale co-target molecule docking results, in order to improve the screening speed, after post-treatment, energy minimization treatment and energy decomposition are carried out on the co-target molecule docking results to obtain corresponding main contribution factors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110748005.2A CN113380320B (en) | 2021-07-01 | 2021-07-01 | Molecular docking result screening method based on positive compound residue contribution similarity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110748005.2A CN113380320B (en) | 2021-07-01 | 2021-07-01 | Molecular docking result screening method based on positive compound residue contribution similarity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113380320A true CN113380320A (en) | 2021-09-10 |
| CN113380320B CN113380320B (en) | 2022-03-15 |
Family
ID=77580507
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110748005.2A Active CN113380320B (en) | 2021-07-01 | 2021-07-01 | Molecular docking result screening method based on positive compound residue contribution similarity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113380320B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024063582A1 (en) * | 2022-09-21 | 2024-03-28 | (주)신테카바이오 | Method for analyzing protein-compound inter-binding stability by artificial intelligence drug discovery platform using molecular dynamics simulation data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020177170A1 (en) * | 2001-04-17 | 2002-11-28 | Peizhi Luo | Structure-based selection and affinity maturation of antibody library |
| CN104031118A (en) * | 2014-06-19 | 2014-09-10 | 天津大学 | Novel affinity peptide ligand of murine polyoma capsomere as well as designing and screening method thereof |
| CN105308602A (en) * | 2013-06-13 | 2016-02-03 | Ucb生物制药私人有限公司 | Obtaining an improved therapeutic ligand |
| CN107111691A (en) * | 2014-10-27 | 2017-08-29 | 阿卜杜拉国王科技大学 | The method and system of recognition ligand protein binding site |
| CN112201313A (en) * | 2020-09-15 | 2021-01-08 | 北京晶派科技有限公司 | Automatic small molecule drug screening method and computing equipment |
-
2021
- 2021-07-01 CN CN202110748005.2A patent/CN113380320B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020177170A1 (en) * | 2001-04-17 | 2002-11-28 | Peizhi Luo | Structure-based selection and affinity maturation of antibody library |
| CN105308602A (en) * | 2013-06-13 | 2016-02-03 | Ucb生物制药私人有限公司 | Obtaining an improved therapeutic ligand |
| CN104031118A (en) * | 2014-06-19 | 2014-09-10 | 天津大学 | Novel affinity peptide ligand of murine polyoma capsomere as well as designing and screening method thereof |
| CN107111691A (en) * | 2014-10-27 | 2017-08-29 | 阿卜杜拉国王科技大学 | The method and system of recognition ligand protein binding site |
| CN112201313A (en) * | 2020-09-15 | 2021-01-08 | 北京晶派科技有限公司 | Automatic small molecule drug screening method and computing equipment |
Non-Patent Citations (2)
| Title |
|---|
| YING LIU: "A Promising method for fast evaluating protein structure similarity", 《2012 9TH INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS AND KNOWLEDGE DISCOVERY》 * |
| 王哲: "基于分子对接的虚拟筛选方法的评测、优化和应用", 《中国优秀博硕士学位论文全文数据库(博士)医药卫生科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024063582A1 (en) * | 2022-09-21 | 2024-03-28 | (주)신테카바이오 | Method for analyzing protein-compound inter-binding stability by artificial intelligence drug discovery platform using molecular dynamics simulation data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113380320B (en) | 2022-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Amaro et al. | Ensemble docking in drug discovery | |
| Wang et al. | Analysis of deep learning methods for blind protein contact prediction in CASP12 | |
| Ju et al. | CopulaNet: Learning residue co-evolution directly from multiple sequence alignment for protein structure prediction | |
| Yu et al. | Designing template-free predictor for targeting protein-ligand binding sites with classifier ensemble and spatial clustering | |
| Bowman et al. | Progress and challenges in the automated construction of Markov state models for full protein systems | |
| Gao et al. | Alphadesign: A graph protein design method and benchmark on alphafolddb | |
| Kosciolek et al. | Accurate contact predictions using covariation techniques and machine learning | |
| Xu et al. | Induced fit docking, and the use of QM/MM methods in docking | |
| Villa et al. | Finding the right fit: chiseling structures out of cryo-electron microscopy maps | |
| Stahl et al. | EPSILON-CP: using deep learning to combine information from multiple sources for protein contact prediction | |
| CN106778065A (en) | A kind of Forecasting Methodology based on multivariate data prediction DNA mutation influence interactions between protein | |
| Godzien et al. | Metabolite annotation and identification | |
| Guo et al. | Self-supervised pre-training for protein embeddings using tertiary structures | |
| JPWO2019240289A1 (en) | Methods and systems for identifying the structure of compounds | |
| CN113380320B (en) | Molecular docking result screening method based on positive compound residue contribution similarity | |
| Li et al. | Protein inter‐residue contact and distance prediction by coupling complementary coevolution features with deep residual networks in CASP14 | |
| Orlando et al. | Observation selection bias in contact prediction and its implications for structural bioinformatics | |
| Niu et al. | Predicting DNA binding proteins using support vector machine with hybrid fractal features | |
| Liu et al. | De novo protein structure prediction by incremental inter-residue geometries prediction and model quality assessment using deep learning | |
| Evteev et al. | Siteradar: utilizing graph machine learning for precise mapping of protein–ligand-binding sites | |
| Pfeiffenberger et al. | Refinement of protein‐protein complexes in contact map space with metadynamics simulations | |
| CN112634997A (en) | Sterol database establishment and sterol analysis method | |
| Ikram et al. | Neural network based cloud computing platform for bioinformatics | |
| Hasan et al. | PredSucc-site: Lysine succinylation sites prediction in proteins by using support vector machine and resolving data imbalance issue | |
| Fu et al. | Simulate time-integrated coarse-grained molecular dynamics with multi-scale graph networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |