CN112750510A - Method for predicting permeability of blood brain barrier of medicine - Google Patents
Method for predicting permeability of blood brain barrier of medicine Download PDFInfo
- Publication number
- CN112750510A CN112750510A CN202110060117.9A CN202110060117A CN112750510A CN 112750510 A CN112750510 A CN 112750510A CN 202110060117 A CN202110060117 A CN 202110060117A CN 112750510 A CN112750510 A CN 112750510A
- Authority
- CN
- China
- Prior art keywords
- brain barrier
- drug
- permeability
- neural network
- blood
- 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.)
- Pending
Links
- 239000003814 drug Substances 0.000 title claims abstract description 83
- 230000008499 blood brain barrier function Effects 0.000 title claims abstract description 63
- 210000001218 blood-brain barrier Anatomy 0.000 title claims abstract description 63
- 230000035699 permeability Effects 0.000 title claims abstract description 61
- 238000000034 method Methods 0.000 title claims abstract description 43
- 229940079593 drug Drugs 0.000 claims abstract description 48
- 238000013528 artificial neural network Methods 0.000 claims abstract description 47
- 238000012549 training Methods 0.000 claims description 28
- 238000012360 testing method Methods 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 230000004913 activation Effects 0.000 claims description 9
- 229910052739 hydrogen Inorganic materials 0.000 claims description 6
- 239000001257 hydrogen Substances 0.000 claims description 6
- 210000002569 neuron Anatomy 0.000 claims description 6
- 238000003062 neural network model Methods 0.000 claims description 4
- 102100033350 ATP-dependent translocase ABCB1 Human genes 0.000 claims description 3
- 102000004506 Blood Proteins Human genes 0.000 claims description 3
- 108010017384 Blood Proteins Proteins 0.000 claims description 3
- 108010047230 Member 1 Subfamily B ATP Binding Cassette Transporter Proteins 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000005192 partition Methods 0.000 claims description 3
- 239000000758 substrate Substances 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 3
- 238000000638 solvent extraction Methods 0.000 claims 1
- 238000002474 experimental method Methods 0.000 abstract description 4
- 230000007547 defect Effects 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 14
- 150000001875 compounds Chemical class 0.000 description 8
- 238000012546 transfer Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 239000000126 substance Substances 0.000 description 5
- 210000004556 brain Anatomy 0.000 description 4
- 210000003169 central nervous system Anatomy 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 238000004617 QSAR study Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 125000000524 functional group Chemical group 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012897 Levenberg–Marquardt algorithm Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 239000008280 blood Substances 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 230000002490 cerebral effect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000012634 fragment Substances 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 208000012902 Nervous system disease Diseases 0.000 description 1
- 208000025966 Neurological disease Diseases 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000004071 biological effect Effects 0.000 description 1
- 239000003576 central nervous system agent Substances 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000009510 drug design Methods 0.000 description 1
- 238000012362 drug development process Methods 0.000 description 1
- 210000002889 endothelial cell Anatomy 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000000338 in vitro Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 150000002632 lipids Chemical class 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 229940126586 small molecule drug Drugs 0.000 description 1
- 230000001225 therapeutic effect Effects 0.000 description 1
- 230000001988 toxicity Effects 0.000 description 1
- 231100000419 toxicity Toxicity 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- 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
- G16C20/70—Machine learning, data mining or chemometrics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/082—Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
-
- 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
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H70/00—ICT specially adapted for the handling or processing of medical references
- G16H70/40—ICT specially adapted for the handling or processing of medical references relating to drugs, e.g. their side effects or intended usage
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Chemical & Material Sciences (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Medicinal Chemistry (AREA)
- Crystallography & Structural Chemistry (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Pharmacology & Pharmacy (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Medical Informatics (AREA)
- Toxicology (AREA)
- Public Health (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention discloses a method for predicting permeability of a medicament blood brain barrier, which utilizes a group contribution method and an artificial neural network to establish a nonlinear model for predicting the permeability of the medicament blood brain barrier. The invention establishes a nonlinear prediction model for predicting the permeability of the blood brain barrier of the drug based on a group contribution method and an artificial neural network, predicts the permeability of the blood brain barrier of the drug by utilizing the molecular structure of the drug, overcomes the defects of large difficulty, time consumption, high cost and the like of the traditional experiment, and provides a prediction method which is more in line with the nonlinear relation between the molecular structure of the drug and the permeability of the blood brain barrier, high in accuracy and strong in applicability for the prediction of the permeability of the blood brain barrier of the drug.
Description
Technical Field
The invention relates to the technical field of calculation of permeability of a blood brain barrier of a medicament, in particular to a method for predicting the permeability of the blood brain barrier of the medicament based on a group contribution method and an artificial neural network, which is suitable for predicting the permeability of the blood brain barrier according to molecular structure information of the medicament.
Background
Quantitative structure-activity relationship (QSAR) is an important method for researching the relationship between the structural characteristics and the activity of a compound, the final purpose is to determine the influence of parameters on the biological activity, and the prediction, evaluation, screening and the like of the properties or the activity of unknown chemicals are realized by establishing a mathematical prediction model, so that basic data are provided for further safety evaluation. Free and Wilson in the sixties of the twentieth century proposed a QSAR study approach, the Free-Wilson group contribution method. The theoretical basis for this approach is to structurally divide a compound into functional groups or structural fragments, and to assume that each functional group or structural fragment contributes independently to a specific effect of the compound. It assumes that the contributions of the same functional group are identical in different molecules or mixtures, and the nature of the pure substances or mixtures is considered to be the sum of this property of the groups that constitute them. With the development of molecular models, structural design and computer simulation techniques, the method is widely applied to research on aspects such as drug design and activity prediction, and achieves remarkable results.
The blood-brain barrier, the central nervous system barrier present in the brain, was found to be a major obstacle before drugs reach targets in the development of drugs for the treatment of neurological diseases. The blood-brain barrier primarily regulates the permeability of drugs to the brain. Its presence limits the penetration of almost 100% of large and 98% of small molecule drugs. Therefore, early screening of new compound entities for blood brain barrier permeability is critical not only for central nervous system compounds (to assess whether therapeutic targets are reached) but also for peripherally active compounds that can penetrate to the brain in order to avoid side effects in the development of new central nervous system drugs.
In the development of CNS drugs to confirm blood-brain barrier permeability, the commonly investigated index is the cerebral blood ratio of the drug, i.e. the logarithmic value of the distribution ratio of the drug in the brain and plasma in steady state is defined as cerebral blood ratio, abbreviated logBB (log BB ═ log (Cbrain/cbiod)). Indeed, experimental determination of logBB is often very difficult, time consuming and expensive, and requires sufficient amounts of pure compound and extensive animal experiments, and is therefore not suitable for providing results in a high throughput manner. Therefore, there is an increasing interest in computational methods for the rapid prediction of the blood-brain barrier permeability of molecules, which are good, reliable and easy to apply. And the prediction model can be widely used in the drug development process, especially in the field of central nervous systems.
The artificial neural network can effectively calculate nonlinear data, can process large-scale data sets, can be used for a plurality of input and output variables, and can be used for constructing a blood brain barrier prediction model. Luxiaquan et al (Chinese patent of invention, CN201010583391.6) invented a method for predicting the pharmacokinetic properties and toxicity of drug molecules based on genetic algorithm and artificial neural network. Khan et al, in developing a mathematical model of the transport of PNP across blood brain barrier endothelial cells, used an Artificial Neural Network (ANN) to process experimental data from in vitro blood brain barrier experiments to determine kinetic rate parameters.
The invention adopts a method of combining a group contribution method with an artificial neural network, adopts the artificial neural network to establish a prediction model, aims to establish a nonlinear model for predicting the permeability of a medicament blood brain barrier, provides a nonlinear method which is more in line with the relationship between the structure and the property of the medicament for the prediction and the evaluation of the permeability of the medicament blood brain barrier, and is a novel method for predicting the permeability of the medicament blood brain barrier.
Disclosure of Invention
The invention aims to overcome the defects of poor prediction accuracy and weak applicability in the prior art, provides a method for predicting the permeability of a blood brain barrier of a medicine based on a group contribution method and an artificial neural network, and improves the accuracy and the applicability of model prediction.
The purpose of the invention can be realized by the following technical scheme:
a method for predicting permeability of a drug blood brain barrier based on a group contribution method and an artificial neural network comprises the following steps:
(1) after obtaining the medicine with known blood brain barrier permeability, introducing the medicine into molecular descriptor calculation software, and calculating to obtain a plurality of molecular descriptors;
(2) transforming the set of molecular descriptor values into a feature dataset for the drug and combining the feature dataset for the drug with the blood-brain barrier permeability of the drug to form a feature matrix for the drug;
(3) randomly dividing the data set into a training set X1 and a testing set X2, wherein the training set X1 is used for neural network training, adjusting model parameters and establishing a prediction model, and the testing set X2 is used for testing and verifying the prediction capability of the developed neural network model;
(4) leading the training set X1 and the testing set X2 into an artificial neural network for training, and determining parameters of an artificial neural network model;
(5) and converting the molecular descriptors of the medicine to be predicted into a characteristic vector form, and predicting according to the artificial neural network model.
Preferably, the acquiring of the molecular descriptor in step (1) includes: a molecular structural group descriptor, a physicochemical property descriptor, and a biological descriptor.
Preferably, the characteristic data set of the drug in step (2) is used as an input parameter of the model, and the blood-brain barrier permeability parameter of the drug is used as an output parameter.
Preferably, the method for dividing the training set X1 and the test set X2 in step (3) is as follows: 90% of the data set was taken as training set X1 and 10% of the data set was taken as test set X2.
Preferably, the artificial neural network architecture in step (4) is a neural network of an input layer, a hidden layer and an output layer.
Preferably, in step (4), the data set is trained by using a gradient descent algorithm or a back propagation algorithm in a feed-forward neural network propagation algorithm.
Preferably, in step (4), the determining parameters of the artificial neural network model includes the number of hidden layers, the number of activation functions of each layer, the number of neurons of each layer, and training parameters of the model.
Preferably, the molecular structural group descriptor is resolved by UNIFAC group contribution to obtain each molecular group.
Preferably, the physical and chemical property descriptors include molecular weight, lipid water partition coefficient, hydrogen bond donor ability, hydrogen bond acceptor ability, polarizability.
Preferably, the biological descriptors include plasma protein binding rate and binding rate of P-glycoprotein high affinity substrate sites.
Preferably, the parameters of the artificial neural network model are as follows: the topological structure of the artificial neural network is 60 multiplied by 6 multiplied by 2 multiplied by 1, the transfer function of the first hidden layer is a Logsig type activation function, the transfer function of the second hidden layer is a Tansig type activation function, the transfer function of the output layer is a purelin type activation function, and the training algorithm adopts a Levenberg-Marquardt algorithm. The training parameters are the minimum error of the training target of 0.1, the learning rate of 0.01, the minimum performance gradient of 1e-6 and the maximum number of times of failure of validation of 50.
The invention has the advantages and beneficial effects that:
(1) the method is simple and easy to operate, can quickly predict the blood brain barrier permeability of the medicine by only providing the molecular structure of the medicine, and provides a basis for improving the blood brain barrier permeability of the medicine and carrying out structural modification in the design of medicine molecules.
(2) The invention can freely add the molecular descriptor data, only needs to add the same descriptor to the same data set, and has universality.
(3) The invention establishes a nonlinear prediction model for predicting the permeability of the blood brain barrier of the drug based on a group contribution method and an artificial neural network, predicts the permeability of the blood brain barrier of the drug by utilizing the molecular structure of the drug, overcomes the defects of large difficulty, time consumption, high cost and the like of the traditional experiment, and provides a prediction method which is more in line with the nonlinear relation between the molecular structure of the drug and the permeability of the blood brain barrier, high in accuracy and strong in applicability for the prediction of the permeability of the blood brain barrier of the drug.
(4) The invention effectively utilizes the statistical learning ability of the artificial neural network and the advantages of processing large-scale data sets, optimizes the molecular descriptors of the drugs and the parameters of the artificial neural network, and obviously improves the prediction efficiency and the prediction ability of the permeability of the blood brain barrier of the drugs.
Drawings
FIG. 1 is a schematic diagram of a neural network model;
FIG. 2 is a cross plot of experimental and predicted values for the model;
FIG. 3 is a flowchart illustrating steps according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples, but the present invention is not limited to the embodiments.
Examples
The invention utilizes a group contribution method and an artificial neural network to establish a nonlinear model of the permeability of the blood brain barrier of the medicine, the model takes a molecular descriptor obtained by calculation software as an input parameter of the model, the blood brain barrier permeability parameter of the medicine as an output parameter, the model is constructed by adopting the artificial neural network, and a prediction result of the blood brain barrier permeability of the medicine is rapidly obtained by inputting the molecular descriptor of the medicine. The technical scheme of the invention comprises the following concrete implementation steps:
(1) obtaining the medicine with known blood brain barrier permeability, constructing a database of the medicine and the blood brain barrier permeability parameters thereof, introducing the database into molecular descriptor calculation software, and calculating to obtain a plurality of molecular descriptors:
and (3) acquiring 300 blood brain barrier permeability parameters of different drugs with the same data source by adopting a data investigation means, and constructing a database required by the model. The method ensures that experimental conditions or data sources are the same, and avoids the influence of the difference between data measured by different researchers on the accuracy of the model. Drawing the 3D molecular structure of the drug in the database and calculating the molecular descriptors: and drawing the 3D molecular structure of the medicament in the constructed database by adopting chemical structure drawing software, optimizing the molecular structure to obtain a more stable 3D molecular structure, introducing the molecular structure into molecular descriptor calculation software, and calculating to obtain various molecular descriptors.
In the examples of the present invention, 70 molecular descriptors (see table one in detail) were constructed in total, and these descriptors include a molecular structural group descriptor, a physicochemical property descriptor, and a biological descriptor. Wherein the molecular structure group descriptor is split by a UNIFAC group contribution method to obtain each molecular group; the physical and chemical property descriptors comprise molecular weight, lipid-water partition coefficient, hydrogen bond donor capacity, hydrogen bond acceptor capacity and polarizability, and the biological descriptors comprise plasma protein binding rate and P-glycoprotein high-affinity substrate site binding rate.
The classification of the molecular descriptors is shown in table one:
watch 1
(2) Transforming the molecular descriptor value set into a characteristic data set of the medicine, and combining the characteristic data set of the medicine with the blood brain barrier permeability of the medicine to form a characteristic matrix of the medicine;
(3) dividing a database:
the obtained database of the molecular descriptor values and the blood brain barrier permeability parameters is randomly divided into a 90% training set X1 and a 10% testing set X2, the training set X1 is used for neural network training, model parameters are adjusted, a prediction model is built, and the testing set X2 is used for testing and verifying the prediction capability of the developed neural network model.
(4) Establishing and training artificial neural network model
And taking the obtained molecular descriptor value of the database as an input parameter, taking the blood brain barrier permeability parameter as an output parameter, training an artificial neural network model by adopting a back propagation algorithm in a feedforward neural network, determining the structure of the artificial neural network, and determining the neuron number and the activation function of an input layer, a hidden layer and an output layer. Training of the artificial neural network is performed using the data of the training set X1. This work is done by the Matlab program.
The topological structure of the artificial neural network is 60 multiplied by 6 multiplied by 2 multiplied by 1, the transfer function of the first hidden layer is a Logsig type activation function, the transfer function of the second hidden layer is a Tansig type activation function, the transfer function of the output layer is a purelin type activation function, and the training algorithm adopts a Levenberg-Marquardt algorithm. The training parameters are the minimum error of the training target of 0.1, the learning rate of 0.01, the minimum performance gradient of 1e-6 and the maximum number of times of failure of validation of 50.
The specific calculation formula for predicting blood brain barrier permeability parameters by the established artificial neural network model is as follows:
The model was then tested and validated: and testing the prediction performance, stability and generalization capability of the trained artificial neural network model by using the data of the test set X2. The relative error RE, the root mean square error RMSE and the correlation coefficient R are calculated according to the following evaluation standard calculation formula:
wherein, PexpIs an experimental value, PpredIs a calculated value and n is the number of compounds in the data set.
And
representing the observed average permeability and the predicted permeability, respectively.
The experimental and predicted values of the blood-brain barrier permeability parameters of the drugs are shown in fig. 2. The test and verification results of the model are shown in table three.
Watch III
(5) The application of the prediction method comprises the following steps: and converting the molecular descriptors of the medicine to be predicted into a characteristic vector form, and predicting according to the artificial neural network model.
After the model with high prediction accuracy is established, the model is not required to be retrained and adjusted, and the prediction result of the blood brain barrier permeability parameter can be calculated and obtained by inputting the corresponding molecular descriptor of the drug. FIG. 3 is a diagram showing the steps of predicting blood-brain barrier permeability of a drug by using a model for predicting blood-brain barrier permeability of a drug according to an embodiment of the present invention.
The embodiment of the invention predicts the medicine through the blood brain barrier permeability prediction model, can accurately predict the blood brain barrier permeability of the medicine, and has wider application range.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. It will be apparent to those skilled in the art that various modifications, substitutions, combinations, or simplifications can be made in the present invention. Therefore, modifications, combinations, and substitutions may be made without departing from the spirit or scope of the invention and these are intended to be within the scope of the claims.
Claims (10)
1. A method for predicting permeability of a blood brain barrier of a drug utilizes a group contribution method and an artificial neural network to establish a nonlinear model for predicting the permeability of the blood brain barrier of the drug, and is characterized by comprising the following steps:
(1) after obtaining the medicine with known blood brain barrier permeability, introducing the medicine into molecular descriptor calculation software, and calculating to obtain a plurality of molecular descriptors;
(2) transforming the set of molecular descriptor values into a feature dataset for the drug and combining the feature dataset for the drug with the blood-brain barrier permeability of the drug to form a feature matrix for the drug;
(3) randomly partitioning the data set into a training set X1And test set X2Training set X1Used for neural network training, adjusting model parameters, establishing a prediction model, and testing set X2Testing and verifying the predictive capability of the developed neural network model;
(4) will train set X1And test set X2Introducing an artificial neural network for training, and determining the structure and parameters of an artificial neural network model;
(5) and converting the molecular descriptor of the drug to be predicted into a characteristic vector form, and predicting by using the constructed artificial neural network model.
2. The method for predicting permeability of a drug blood-brain barrier of claim 1, wherein the obtained molecular descriptors include molecular structural group descriptors, physicochemical property descriptors and biological descriptors.
3. The method for predicting permeability of a drug to the blood-brain barrier of claim 2, wherein the molecular structural group descriptor is resolved by the UNIFAC group contribution method to obtain each molecular group.
4. The method for predicting permeability of a pharmaceutical blood brain barrier according to claim 2, wherein said descriptors of physicochemical properties include molecular weight, lipid-water partition coefficient, hydrogen bond donor ability, hydrogen bond acceptor ability, polarizability.
5. The method for predicting permeability of a drug to the blood-brain barrier of claim 2, wherein said biological descriptors include plasma protein binding rate and binding rate of P-glycoprotein high affinity substrate sites.
6. The method of predicting blood-brain barrier permeability of a drug according to claim 1, wherein the feature data set of the drug is used as an input parameter of an artificial neural network model, and the blood-brain barrier permeability parameter of the drug is used as an output parameter of the artificial neural network model.
7. The method for predicting permeability of a drug to the blood-brain barrier of claim 1, wherein in the step (3), the training set X1 and the testing set X2 are divided into: 90% of the data set was taken as training set X1 and 10% of the data set was taken as test set X2.
8. The method of claim 1, wherein the artificial neural network model is configured as a neural network including an input layer, a hidden layer, and an output layer.
9. The method for predicting permeability of blood brain barrier to drugs according to claim 1, wherein in the step (4), the data set is trained by using a gradient descent algorithm or a back propagation algorithm in a feedforward neural network propagation algorithm.
10. The method for predicting permeability of a drug blood-brain barrier of claim 1, wherein in the step (4), the parameters for determining the model of the artificial neural network include the number of hidden layers, the activation function of each layer, the number of neurons of each layer, and the training parameters for adjusting the model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110060117.9A CN112750510A (en) | 2021-01-18 | 2021-01-18 | Method for predicting permeability of blood brain barrier of medicine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110060117.9A CN112750510A (en) | 2021-01-18 | 2021-01-18 | Method for predicting permeability of blood brain barrier of medicine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112750510A true CN112750510A (en) | 2021-05-04 |
Family
ID=75652219
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110060117.9A Pending CN112750510A (en) | 2021-01-18 | 2021-01-18 | Method for predicting permeability of blood brain barrier of medicine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112750510A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113948156A (en) * | 2021-10-20 | 2022-01-18 | 大连理工大学 | Multitask neural network method for predicting degradation half-life of chemicals in four environment media |
| CN114171111A (en) * | 2021-11-12 | 2022-03-11 | 中检科健(天津)检验检测有限责任公司 | QSAR screening method for chemical substance P-glycoprotein interaction |
| CN114520031A (en) * | 2022-01-24 | 2022-05-20 | 浙江大学 | Method for predicting permeability of compound placental membrane based on machine learning |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109545389A (en) * | 2018-11-08 | 2019-03-29 | 电子科技大学中山学院 | Method for establishing data set in prediction of blood brain barrier permeability of medicine and data model |
| US20200342960A1 (en) * | 2017-07-06 | 2020-10-29 | Pukyong National University Industry-University Cooperation Foundation | Method for screening of target-based drugs through numerical inversion of quantitative structure-(drug)performance relationships and molecular dynamics simulation |
-
2021
- 2021-01-18 CN CN202110060117.9A patent/CN112750510A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200342960A1 (en) * | 2017-07-06 | 2020-10-29 | Pukyong National University Industry-University Cooperation Foundation | Method for screening of target-based drugs through numerical inversion of quantitative structure-(drug)performance relationships and molecular dynamics simulation |
| CN109545389A (en) * | 2018-11-08 | 2019-03-29 | 电子科技大学中山学院 | Method for establishing data set in prediction of blood brain barrier permeability of medicine and data model |
Non-Patent Citations (3)
| Title |
|---|
| YAN CHEN 等: "A prediction model for blood–brain barrier permeation and analysis on its parameter biologically", 《ELSEVIER》 * |
| ZEYU WU 等: "Artificial neural network approach for predicting blood brain barrier permeability based on a group contribution method", 《ELSEVIER》 * |
| 戴康等: "贝叶斯规整化神经网络模型预测化合物分子的血脑屏障通透性", 《华中师范大学学报(自然科学版)》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113948156A (en) * | 2021-10-20 | 2022-01-18 | 大连理工大学 | Multitask neural network method for predicting degradation half-life of chemicals in four environment media |
| CN113948156B (en) * | 2021-10-20 | 2024-05-07 | 大连理工大学 | Multitasking neural network method for predicting degradation half-life of chemicals in four environmental media |
| CN114171111A (en) * | 2021-11-12 | 2022-03-11 | 中检科健(天津)检验检测有限责任公司 | QSAR screening method for chemical substance P-glycoprotein interaction |
| CN114520031A (en) * | 2022-01-24 | 2022-05-20 | 浙江大学 | Method for predicting permeability of compound placental membrane based on machine learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108647489B (en) | Method and system for screening disease drug target and target combination | |
| Peng et al. | Predicting drug response based on multi-omics fusion and graph convolution | |
| CN108877953A (en) | A kind of drug sensitivity prediction method based on more similitude networks | |
| CN108830040B (en) | Drug sensitivity prediction method based on cell line and drug similarity network | |
| CN112750510A (en) | Method for predicting permeability of blood brain barrier of medicine | |
| CN110232978B (en) | Cancer cell line treatment drug prediction method based on multidimensional network | |
| US20150324546A1 (en) | Method for predicting drug-target interactions and uses for drug repositioning | |
| CN106503483B (en) | Myeloma signal path mechanism confirmation method based on modularization factor graph | |
| CN112599187B (en) | Method for predicting drug and target protein binding fraction based on double-flow neural network | |
| Zankov et al. | QSAR modeling based on conformation ensembles using a multi-instance learning approach | |
| WO2024011837A1 (en) | Drug screening method and system | |
| Masoudi-Sobhanzadeh et al. | Synthetic repurposing of drugs against hypertension: a datamining method based on association rules and a novel discrete algorithm | |
| CN114913916A (en) | Drug relocation method for predicting new coronavirus adaptive drugs | |
| WO2014072950A1 (en) | System and method for development of therapeutic solutions | |
| Doshi et al. | A computational approach to drug repurposing using graph neural networks | |
| Wu et al. | Artificial neural network approach for predicting blood brain barrier permeability based on a group contribution method | |
| Badam et al. | A validated generally applicable approach using the systematic assessment of disease modules by GWAS reveals a multi-omic module strongly associated with risk factors in multiple sclerosis | |
| Chen et al. | Identification and analysis of spinal cord injury subtypes using weighted gene co-expression network analysis | |
| Wang et al. | A multitask GNN-based interpretable model for discovery of selective JAK inhibitors | |
| Klein et al. | Association between ALS and retroviruses: evidence from bioinformatics analysis | |
| Wong et al. | The New Answer to Drug Discovery: Quantum Machine Learning in Preclinical Drug Development | |
| CN113889217A (en) | Medicine recommendation method based on twin neural network and depth factorization machine | |
| CN113178234A (en) | Compound function prediction method based on neural network and connection graph algorithm | |
| CN110534153B (en) | Target prediction system and method based on deep learning | |
| KR20210110241A (en) | Prediction system and method of cancer immunotherapy drug Sensitivity using multiclass classification A.I based on HLA Haplotype |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210504 |
|
| RJ01 | Rejection of invention patent application after publication |