CN107516012A - A kind of structured descriptor calculated based on organic compound molecule three-dimensional structure - Google Patents
A kind of structured descriptor calculated based on organic compound molecule three-dimensional structure Download PDFInfo
- Publication number
- CN107516012A CN107516012A CN201710718151.4A CN201710718151A CN107516012A CN 107516012 A CN107516012 A CN 107516012A CN 201710718151 A CN201710718151 A CN 201710718151A CN 107516012 A CN107516012 A CN 107516012A
- Authority
- CN
- China
- Prior art keywords
- hydrogen atom
- molecule
- organic compound
- atom
- compound
- 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.)
- Withdrawn
Links
Classifications
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of structured descriptor calculated based on organic compound molecule 3-D solid structure, belong to compound Quantitative Study of Structure Property relation research method technical field.Purpose is by being training set from part known compound, structural characterization is carried out to training set sample using the descriptor, then such compound structure property relation (QSPR/QSAR) model is built with appropriate mathematical method (multiple linear regression (MLR), PLS (PLS)), for a certain property of the similar unknown compound of simulation and forecast.Method comprises the following steps：Skeleton non-hydrogen atom classification in step 1 organic compound molecule；Step 2 carries out parametrization dyeing to different non-hydrogen atoms；Step 3 builds the relation between different types of non-hydrogen atom by reciprocal function；Organic compound molecule structure is optimized minimum energy state by step 4, obtains the space coordinates of non-hydrogen atom, and structured descriptor is calculated in application program.By establishing the relational model between compound structure descriptor and certain property, can the accurately similar organic compound of simulation and forecast property, the QSPR/QSAR researchs for organic compound have very high reference value.
Description
Technical field
Present invention relates particularly to a kind of structured descriptor calculated based on organic compound molecule three-dimensional structure, belong to chemical combination
Thing Quantitative Structure Property Relationship (QSPR/QSAR) research method technical field.
Background technology
Compound structure determines property, and property is the reflection of compound structure.Quantitative pass between molecular structure and property
The structure of system is, it is necessary to introduce corresponding structured descriptor.For a long time, researchers have done many significant in this respect
Work.Structure description is carried out by using the geometry of molecule, topological property and connection features and various physico-chemical parameters, so
The various properties that QSPR/QSAR models carry out predictive compound are established afterwards, see paper：It is dirty that Generalized Correlative Index is used for persistent environmental
Contaminate D-M (Determiner-Measure) construction-Retention Relationship research [J] analytical chemistry of thing, 2006,34 (8)：1096-1100.But the above method is all
It is two-dimentional (2D) structured descriptor, it is difficult to reproduce molecule real space stereochemical structure, the structures such as cis-trans isomerism are difficult to differentiate between.Three
Dimension (3D) descriptor develops rapidly, and turns into the main flow of QSPR/QSAR molecular structure characterizations, mainly there is WHIM indexes and CoMFA,
See paper：MS-WHIM, new 3D theoretical descriptors derived from molecular surface
properties：a comparative 3D QSAR study in a series of steroids.J Comput-Aided
Mol Des, 1997,11：79-92；Investigation of structural requirements for inhibitory
activity at the rat and housefly picrotoxinin binding sites in ionotropic
GABA receptors using DISCOtech and CoMFA, Chemosphere, 2007,69：864-871.WHIM indexes
It is to be weighted conversion to atomic space coordinate as different physical quantities to produce to obtained from rotation and translation invariant, calculating
Process is considerably complicated and is difficult to be used widely.And the drawbacks of CoMFA is to be first had to when carrying out one group of molecular studies to sample
This molecule carries out space structure overlapping, and conformation is overlapping, in addition space lattice division, variables number control and the selection of potential field probe etc.
The problem of process complexity is hard to understand, workload is big, and has many uncertain factors, and these are all very important.Therefore build
Easy, understandable three-dimensional (3D) descriptor based on compound molecule stereoeffect is significant, but at present also
There is no highly effective, easy method to occur.
The content of the invention
Therefore, for the above-mentioned deficiency of prior art, the present invention seeks to for QSPR/QSAR researchs provide it is simple, understandable,
Compounds effective molecular structural parameter characterizing method (structured descriptor).In specific application, from chemical combination known to part
Thing is training set, structural characterization is carried out to training set sample by the descriptor, then using appropriate mathematical method (polynary line
Property return (MLR), PLS (PLS)) build such compound Quantitative Structure Property Relationship (QSPR/QSAR) mould
Type, for predicting similar unknown compound a certain property (such as chromatographic retention, toxicity, migration characteristic, degradability, drug effect, biology
Activity etc.), provide reference to carry out other correlative studys.
The method of the present invention comprises the following steps：
Skeleton non-hydrogen atom classification in step 1 organic compound molecule
Non-hydrogen atom in organic compound forms molecule by different connected modes (chemical bond), ignores non-skeleton hydrogen original
The influence of son, the non-hydrogen atom of intramolecular can be divided into the class of A1, A2, A3, A4 tetra- according to its non-hydrogen atom number connected, respectively table
Show and be connected with 1,2,3,4 non-hydrogen atom.
Step 2 carries out parametrization dyeing to different non-hydrogen atoms
The feature of non-hydrogen atom in the molecule, mainly determined by factors such as its valence electron number, the electronics numbers of plies, thus used down
Formula carries out parametrization dyeing to different non-hydrogen atoms and obtains the parametrization dye number of non-hydrogen atom.
Zi=[mi(ni-1)(XC/Xi)1/2-hi]1/2Formula one
N in formulaiFor the non-hydrogen atom i electronics number of plies, miFor outermost electron number, XiFor the Pauling electronegativity of carbon atom, hi
For the number of hydrogen atoms directly connected；XCFor the Pauling electronegativity of carbon atom.
Step 3 builds the relation between different types of non-hydrogen atom by reciprocal function
Relation in molecule between non-hydrogen atom is not certain specific effect between atom, but to reflect its level of intimate
With non-hydrogen atom ZiThe change trend of value is consistent and the two aspect situations opposite with the change trend of both distances.Usual shape reciprocal
Function can meet this requirement, carry out expressing the relation between different non-hydrogen atoms using following formula.
rijIt is relative distance (i.e. the ratio between the space length and carbon-carbon single bond bond distance's value between the two of non-hydrogen atom i, j in molecule
Represent)；N and l is the affiliated type of non-hydrogen atom.4 class non-hydrogen atoms can produce 10 kinds of different continuous items in compound molecule：
m11、m12、m13、m14、m22、m23、m24、m33、m34、m44, wherein m12Represent first kind non-hydrogen atom and the second class non-hydrogen in molecule
Relation between atom, similarly m23The relation between the second class non-hydrogen atom and the 3rd class non-hydrogen atom in molecule is represented, with this
Analogize.10 kinds of different relational terms are designated as x respectively1、x2、x3、x4、x5、x6、x7、x8、x9And x10, so at most will for research sample
Produce 10 variables related to molecular structure.
Organic compound molecule structure is optimized minimum energy state by step 4, and the space for obtaining non-hydrogen atom is sat
Structured descriptor is calculated in mark, application program.
The initial volumetric structure of organic compound molecule is built using ChemOffice 8.0, is carried with Chem 3D
MOPAC Semi-empirical quantums software final optimization pass in AM1 levels obtains molecular structure (cutoff value 0.001kJmol-1),
And obtain the locus coordinate of each atom.By the locus coordinate of each atom in molecule and parametrization dye number input
Self-editing C language application program is acted upon, and obtains Molecular structure descriptor.
When specifically with the structured descriptor of the present invention, it is also necessary to the variable for being obtained above-mentioned steps, use first
Successive Regression screens according to variable conspicuousness to variable, then dependent variable X is combined as with the variable filtered out, with compound
Property is dependent variable Y, should with appropriate mathematical method (multiple linear regression (MLR), PLS (PLS)) structure
Class compound Quantitative Structure Property Relationship (QSPR/QSAR) model, and then similar unknown compound property simulate pre-
Survey.
The beneficial effects of the present invention are：The present invention provides a kind of knot calculated based on organic compound molecule three-dimensional structure
Structure descriptor, calculates simple, understandable, parametrization can characterize organic compound molecule structure row.With appropriate mathematics side
Method (multiple linear regression (MLR), PLS (PLS)) structure compound Quantitative Structure Property Relationship (QSPR/
QSAR coefficient correlation (the R of) model, model coefficient correlation (R) and cross-verificationCV) ideal, disclose to a certain extent
Influence the structural factor of compound property.Model can accurately predict the relevant nature of similar unknown compound, for
The QSPR/QSAR researchs of organic compound have very high reference value.
Brief description of the drawings
Fig. 1 is ortho-xylene molecular structure in embodiment
Fig. 2 is Variable Selection process MLR model coefficient correlations (R/R in embodimentCV) situation of change
Fig. 3 is Variable Selection process MLR model criteria deviations (SD/SD in embodimentCV) situation of change
Fig. 4 is 37 samples in embodiment in PLS the first two principal component scores space scatter diagram；
Fig. 5 is predicted value figure related to experiment value in embodiment；
Embodiment
The present invention organic compound structure descriptor can be used for a variety of properties of compound (such as chromatographic retention, toxicity,
Migration characteristic, degradability, drug effect, bioactivity etc.) simulation and forecast, below in conjunction with the accompanying drawings to the present invention apply to compound
The simulation and forecast embodiment of chromatographic retention illustrates, can when needing the other properties of simulation and forecast compound
To be implemented using similar approach.
Experiment material
The present embodiment choose 37 wild jasmine flowers fragrance component for research sample, compound gas phase chromatographic retention with
tRRepresent, experiment value is derived from paper：Fragrance component [J] fine chemistry industries of SPME Gc-mss wild jasmine flower,
2007,24 (2)：159-161.Compound and its gas chromatography retention time (tR) it is listed in table 1.
Table 1
Experimental method
1) molecular structure of compounds characterizes
Chromatographic retention (the t of organic compoundR) except having outside the Pass with measurement factor, it is also related to the structure of molecule.
Forming connected mode between compound atomic species, number, atom etc. all can be to tRHave an impact.Non-hydrogen atom presses different companies
Connect mode (chemical bond) and form molecule.Ignore the influence of non-skeleton hydrogen atom, the non-hydrogen atom of intramolecular is connected according to it
Non-hydrogen atom number can be divided into the class of A1, A2, A3, A4 tetra-, respectively represent be connected with 1,2,3,4 non-hydrogen atom, it is such as non-with two
The connected secondary carbon of hydrogen atom belongs to A2 atomic types.The feature of non-hydrogen atom in the molecule, mainly by its valence electron number, electricity
The factors such as sublayer number determine that thus carrying out parametrization dyeing to different non-hydrogen atoms using following formula (1) obtains non-hydrogen atom ginseng
Numberization dye number.
Zi=[mi(ni-1)(XC/Xi)1/2-hi]1/2 (1)
N in formulaiFor the non-hydrogen atom i electronics number of plies, miFor outermost electron number, XiFor the Pauling electronegativity of carbon atom, hi
For the number of hydrogen atoms directly connected；XCFor the Pauling electronegativity of carbon atom.
Relation in molecule between non-hydrogen atom is not certain specific effect between atom, but to reflect its level of intimate
With non-hydrogen atom ZiThe change trend of value is consistent and the two aspect situations opposite with the change trend of both distances.Usual shape reciprocal
Function can meet this requirement, carry out expressing the relation between different non-hydrogen atoms using following formula (2).
rijIt is molecule non-hydrogen atom i, j relative distance (i.e. the ratio between space length and carbon-carbon single bond bond distance's value table between the two
Show)；N and l is the affiliated type of non-hydrogen atom.4 class non-hydrogen atoms can produce 10 kinds of different continuous items in compound molecule：
m11、m12、m13、m14、m22、m23、m24、m33、m34、m44, wherein m12Represent first kind non-hydrogen atom and the second class non-hydrogen in molecule
Relation between atom, similarly m23The relation between the second class non-hydrogen atom and the 3rd class non-hydrogen atom in molecule is represented, with this
Analogize.10 kinds of different relational terms are designated as x respectively1、x2、x3、x4、x5、x6、x7、x8、x9And x10, so at most will for research sample
Produce 10 variables related to molecular structure.
The initial volumetric structure of organic compound is built using ChemOffice 8.0, the MOPAC half carried with Chem 3D
Experience quantum chemistry software final optimization pass in AM1 levels obtains molecular structure (cutoff value 0.001kJmol-1), and obtain every
The locus coordinate of individual atom.The locus coordinate of each atom and parametrization dye number are inputted into self-editing C language should
It is acted upon with program, obtains Molecular structure descriptor.
Here by taking ortho-xylene as an example, the calculating of structured descriptor is illustrated.First to each non-hydrogen in compound molecule
Atom is numbered, and then judges the atomic type of each non-hydrogen atom.For ortho-xylene, intramolecular has 8 non-hydrogen atoms
(see Fig. 1), its atomic type are respectively A3, A3, A2, A2, A2, A2, A1 and A1.Calculate x1, x1It is former for first kind non-hydrogen in molecule
The relation of son and first kind non-hydrogen atom, i.e., the correlation between No. 7 and No. 8 two atoms.Ortho-xylene Optimum configuration (see
When Fig. 1) space coordinates of No. 7 and No. 8 two atoms be respectively (- 0.3524,2.0282, -0.0202) and (2.1627,
0.5775,0.0627), their space length is 0.2905nm, then r7,8=0.2905/0.1540=1.8864.In addition, root
Dye number, Z7=Z8=1.0000, therefore x are calculated according to formula (1)1Numerical computations are as follows：x1=m11=1.0000 × 1.0000/
1.88642=0.2810.Equally, other several structured descriptor values can be calculated.
Parametrization sign is carried out to the research composition of sample using above method, compound structure descriptor is obtained, due to sample
The relation between the 4th class non-hydrogen atom is not present in this, thus finally gives 9 non-complete " 0 " structured descriptors, is listed in table 2.
Table 2
QSPR is modeled and examined
Multiple linear regression (MLR) and PLS (PLS) method are used, respectively, to establish model, using " staying
One method " interacts inspection to model.It is generally believed that modeling coefficient correlation (R) shows model height phase between 0.8-1.0
Close；Cross-verification coefficient RCV>=0.7, show that model has good robustness and predictive ability.Each independent variable in model
Multicollinearity evaluated with variance inflation factor (variance inflation factors, VIF), VIF definition is：
VIF=(1-r2)-1, r is the degree of correlation (criterion of certain independent variable and other variables in formula：VIF values are more than 5.0, between variable
Synteny is serious, and dependent equation is unreliable), such as diagnosis finds that model variable V IF values are excessive, then continues to reduce variable modeling.
Variable is extracted using successive Regression (SMR) successively according to variable conspicuousness size first, the then change to pick out
Amount is combined as independent variable X, with compound gas phase chromatographic retention (tR) it is dependent variable Y, then with multiple linear regression
(MLR) establish model, Variable Selection and the results are shown in Table 3 and Fig. 2, Fig. 3 accordingly.
Table 3
One good forecast model not only has preferable capability of fitting to internal specimen, but also should be to external samples
With stronger predictive ability.Therefore in preference pattern, in the case where ensureing that there is good fit effect to internal specimen,
Cross-verification coefficient correlation (R is selected as far as possibleCV) larger model, to ensure that model has stronger predictive ability and stability.
It can be seen that in table 3, Fig. 2, Fig. 3, it should which the variable that selection is selected by the step of successive Regression (SMR) the 3rd combines (x3、x5、x7) modeling
Gained model, now compound gas phase chromatographic retention (tR) regression equation between structured descriptor is：
tR=0.2095+0.4030x3+0.2251·x5+0.2249·x7 (3)
N=37, R=0.9403, SD=1.5498, F=84.0232；RCV=0.9186, SDCV=1.7994FCV=
59.4872。
Modeling coefficient correlation (R) reaches 0.9403 (being between 0.8-1.0), and cross-verification coefficient correlation (RCV) reach
Maximum 0.9186 (is more than 0.7), illustrates this model height correlation, robustness is good, predictive ability is strong.Standard deviation (SD) is
1.5498, cross-verification standard deviation (SDCV) it is 1.7994, both of which is smaller, illustrates that model prediction accuracy is higher.In variable
x5Variance inflation factor (VIF) be up to 1.3379, the standard much smaller than 5.0, illustrate to there's almost no between model variable altogether
Linearly.x3、x5、x7Standardized regression coefficient be respectively 0.6190,1.0776,0.1581, illustrate x5The chromatogram of compound is protected
Stay time effects maximum, next to that x3, x7The chromatographic retention of compound is influenceed relatively small.x5Corresponding to the 2nd class non-hydrogen
Relation between atom, illustrate that the class non-hydrogen atom more multi-color spectrum retention time of compound the 2nd is bigger.
The step of successive Regression the 3rd is screened into gained Molecular structure descriptor as X variables, when retaining with compound gas phase chromatogram
Between (tR) it is dependent variable Y, training set sample is carried out with Simca-P11.5 to establish PLS (PLS) model, simultaneously
Inspection is interacted to gained PLS models using " leaving-one method ".Final gained PLS models contain 2 principal components (A), now chemical combination
Thing gas chromatography retention time (tR) regression equation is formula (4) between prototype structure descriptor.
tR=0.3275+0.3790x3+0.2243·x5+0.3157·x7 (4)
Each coefficient R of model is 0.9381 (between 0.8-1.0), RCVFor 0.8899 (being more than 0.7).Each standard
Deviation SD, SDCVRespectively 1.5362,1.7436.Models fitting effect described above is good, robustness is good, predictive ability is strong.Fig. 4
It is 37 samples in PLS the first two principal component scores space scatter diagram, Hotelling T2Ellipse is 95% confidence level confidence circle,
It can be seen that most sample points all fall in confidence circle, the results showed that structured descriptor can appropriately show organic compound
Molecular characterization, and correct response is made in statistical model.
Gas chromatography retention time (the t of MLR models and PLS models to sampleR) it is simulated prediction, predicted value difference
It is listed in the Cal1 and Cal2 of table 1.Related figures of the Fig. 5 between model predication value and experiment value, from fig. 5, it can be seen that all
Sample point is all distributed on 45 ° of diagonal or against diagonal, illustrates that predicted value is very close with experiment value, macro-forecast effect
Fruit is good, and prediction result is with a high credibility.
Compared with prior art, the structured descriptor of structure is that the three-dimensional structure based on molecule is calculated three-dimensional (3D) point
Descriptor of substructure, is easily understood, amount of calculation is small, can distinguish the isomers such as cis-trans isomerism, and non-hydrogen atom dye number considers
Principal quantum number, electronegativity, outermost electron number, the abundant structural information such as number of hydrogen atoms of connection.
Non-hydrogen atom in molecule is classified and parameterizes dyeing, the space length between different non-hydrogen atoms is closed
System is used as Molecular structure descriptor, and parametrization has been carried out to part organic compound structure and has been characterized.Using successive Regression (SMR) with
Multiple linear regression (MLR), partial least-square regression method (PLS) construct compound structure and gas chromatography retention time
(tR) relational model, the coefficient correlation (R) of model coefficient correlation (R) and cross-verification is ideal, takes off to a certain extent
Influence compound gas phase chromatographic retention (t is shownR) structural factor.Model can accurately be predicted in plants essential oil and waved
Gas chromatography retention time (the t of hair property organic compoundR), the QSPR/QSAR researchs for organic compound have higher
Reference value.
Described above is the preferred embodiment of the present invention, it should be pointed out that Molecular structure descriptor of the present invention removes can
With applied to gas chromatography retention time (tR) outside simulation and forecast, the toxicity, migration characteristic, degraded of compound can also be applied to
The simulation and forecast of a variety of properties such as property, drug effect, bioactivity, for those skilled in the art, is not taking off
On the premise of from principle of the present invention, some improvements and modifications can also be made, these improvements and modifications also should be regarded as this hair
Bright protection domain.
Claims (1)
1. a kind of structured descriptor calculated based on organic compound molecule three-dimensional structure, it is characterised in that methods described includes
Following steps：
Skeleton non-hydrogen atom classification in step 1 organic compound molecule
Non-hydrogen atom forms molecule by different connected modes (chemical bond) in organic compound, ignores the shadow of non-skeleton hydrogen atom
Ringing, the non-hydrogen atom of intramolecular can be divided into the class of A1, A2, A3, A4 tetra- according to its non-hydrogen atom number connected, represent respectively with 1,
2nd, 3,4 non-hydrogen atoms are connected.
Step 2 carries out parametrization dyeing to different non-hydrogen atoms
The feature of non-hydrogen atom in the molecule, mainly determined by factors such as its valence electron number, the electronics numbers of plies, thus using following formula pair
Different non-hydrogen atoms carries out parametrization dyeing and obtains the parametrization dye number of non-hydrogen atom.
Zi=[mi(ni-1)(XC/Xi)1/2-hi]1/2Formula one
N in formulaiFor the non-hydrogen atom i electronics number of plies, miFor outermost electron number, XiFor the Pauling electronegativity of carbon atom, hiFor with
Its number of hydrogen atoms being directly connected to；XCFor the Pauling electronegativity of carbon atom.
Step 3 builds the relation between different types of non-hydrogen atom by reciprocal function
Relation in molecule between non-hydrogen atom is not certain specific interaction between atom, but to reflect its level of intimate
With non-hydrogen atom ZiThe change trend of value is consistent and the two aspect situations opposite with the change trend of both distances.Usual shape reciprocal
Function can meet this requirement, carry out expressing the relation between different non-hydrogen atoms using following formula.
rijIt is relative distance (i.e. the ratio between space length and carbon-carbon single bond bond distance's value table between the two of non-hydrogen atom i, j in molecule
Show)；N and l is the affiliated type of non-hydrogen atom.4 class non-hydrogen atoms can produce 10 kinds of different relational terms in compound molecule：
m11、m12、m13、m14、m22、m23、m24、m33、m34、m44, wherein m12Represent first kind non-hydrogen atom and the second class non-hydrogen in molecule
Relation between atom, similarly m23The relation between the second class non-hydrogen atom and the 3rd class non-hydrogen atom in molecule is represented, with this
Analogize.10 kinds of different relational terms are designated as x respectively1、x2、x3、x4、x5、x6、x7、x8、x9And x10, so at most will for research sample
Produce 10 structured descriptors related to molecular structure.
Organic compound molecule structure is optimized minimum energy state by step 4, obtains the space coordinates of non-hydrogen atom,
Structured descriptor is calculated in application program.
The initial volumetric structure of organic compound molecule is built using ChemOffice 8.0, the MOPAC half carried with Chem 3D
Experience quantum chemistry software final optimization pass in AM1 levels obtains molecular structure (cutoff value 0.001kJmol-1), and obtains
The locus coordinate of each atom.The locus coordinate of each atom in molecule and parametrization dye number input is self-editing
C language application program is acted upon, and obtains Molecular structure descriptor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710718151.4A CN107516012A (en) | 2017-08-21 | 2017-08-21 | A kind of structured descriptor calculated based on organic compound molecule three-dimensional structure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710718151.4A CN107516012A (en) | 2017-08-21 | 2017-08-21 | A kind of structured descriptor calculated based on organic compound molecule three-dimensional structure |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107516012A true CN107516012A (en) | 2017-12-26 |
Family
ID=60722355
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710718151.4A Withdrawn CN107516012A (en) | 2017-08-21 | 2017-08-21 | A kind of structured descriptor calculated based on organic compound molecule three-dimensional structure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107516012A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111429980A (en) * | 2020-04-14 | 2020-07-17 | 北京迈高材云科技有限公司 | Automatic acquisition method for material crystal structure characteristics |
| CN112185477A (en) * | 2020-09-25 | 2021-01-05 | 北京望石智慧科技有限公司 | Method and device for extracting molecular characteristics and calculating three-dimensional quantitative structure-activity relationship |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107024558A (en) * | 2017-01-10 | 2017-08-08 | 内江师范学院 | A kind of organic compound molecule structure parameterization characterizing method |
-
2017
- 2017-08-21 CN CN201710718151.4A patent/CN107516012A/en not_active Withdrawn
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107024558A (en) * | 2017-01-10 | 2017-08-08 | 内江师范学院 | A kind of organic compound molecule structure parameterization characterizing method |
Non-Patent Citations (2)
| Title |
|---|
| 廖立敏等: "部分有机污染物灰/水分配系数的定量结构性质关系研究", 《南京理工大学学报》 * |
| 李悦等: "新型结构描述法用于芳烃类化合物水溶性模拟", 《内江师范学院学报》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111429980A (en) * | 2020-04-14 | 2020-07-17 | 北京迈高材云科技有限公司 | Automatic acquisition method for material crystal structure characteristics |
| CN112185477A (en) * | 2020-09-25 | 2021-01-05 | 北京望石智慧科技有限公司 | Method and device for extracting molecular characteristics and calculating three-dimensional quantitative structure-activity relationship |
| CN112185477B (en) * | 2020-09-25 | 2024-04-16 | 北京望石智慧科技有限公司 | Method and device for extracting molecular characteristics and calculating three-dimensional quantitative structure-activity relationship |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jalalvand et al. | MATLAB in electrochemistry: A review | |
| Polanski et al. | The comparative molecular surface analysis (COMSA): a novel tool for molecular design | |
| CN104237158B (en) | A kind of Qualitative Analysis of Near Infrared Spectroscopy method with universality | |
| Susto et al. | Multi-step virtual metrology for semiconductor manufacturing: A multilevel and regularization methods-based approach | |
| Qiao et al. | Comparison of common spatial interpolation methods for analyzing pollutant spatial distributions at contaminated sites | |
| Jamialahmadi et al. | Relationship of permeability, porosity and depth using an artificial neural network | |
| Ahmadi et al. | A systematic study on the accuracy of chemical quantitative analysis using soft modeling methods | |
| CN107516012A (en) | A kind of structured descriptor calculated based on organic compound molecule three-dimensional structure | |
| CN108549908A (en) | Chemical process fault detection method based on more sampled probability core principle component models | |
| CN106055827A (en) | Oil deposit numerical value simulation parameter sensibility analysis device and method | |
| CN106153871A (en) | A kind of OIL SOURCE CORRELATION method | |
| Kuzmanovski et al. | Counter-propagation neural networks in Matlab | |
| CN107749627A (en) | Based on the intelligent distribution network Load Flow Jacobian Matrix method of estimation for improving match tracing | |
| Liu et al. | Predictive model for water absorption in sublayers using a machine learning method | |
| CN107560539B (en) | Method and system for making measurements in complex patterned structures | |
| Gao et al. | Process knowledge discovery using sparse principal component analysis | |
| CN108959741A (en) | A kind of parameter optimization method based on marine physics ecologic coupling model | |
| CN103344740B (en) | Based on the glutamic acid production concentration online soft sensor method of multi input Wiener model | |
| CN112554864A (en) | Method for calculating single-well control reserve of water-producing gas well | |
| CN108595782B (en) | Calculation method for mass transfer between matrix and cracks in discrete cracks | |
| CN103390103A (en) | Melt index online detection method based on subspace independent component regression model | |
| CN103076556B (en) | Method for selecting function-maintenance testing points of avionic assembly | |
| CN102323973B (en) | Method for predicting common environment poison property/activity on the basis of intelligent correlation index | |
| CN116933386A (en) | Aircraft pneumatic data fusion method based on MCOK proxy model | |
| CN110544507B (en) | Method, system and application for regulating and controlling solid state fermentation based on histology integration technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20171226 |
|
| WW01 | Invention patent application withdrawn after publication |