CN103761431B - Method for predicting fish bio-concentration factors of organic chemicals by quantitative structure-activity relationship - Google Patents
Method for predicting fish bio-concentration factors of organic chemicals by quantitative structure-activity relationship Download PDFInfo
- Publication number
- CN103761431B CN103761431B CN201410013562.XA CN201410013562A CN103761431B CN 103761431 B CN103761431 B CN 103761431B CN 201410013562 A CN201410013562 A CN 201410013562A CN 103761431 B CN103761431 B CN 103761431B
- Authority
- CN
- China
- Prior art keywords
- model
- organic compounds
- kinds
- descriptor
- organic
- 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.)
- Active
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention discloses a method for predicting fish bio-concentration factors of organic chemicals by the quantitative structure-activity relationship, and belongs to the field of ecological risk assessment and test strategies. According to the method, bio-concentration factor data of 780 types of organic compounds are collected from public databases or published papers; molecular structures of the organic compounds are optimized according to the density functional theory, and 4885 types of molecule descriptors of the organic compounds are preliminarily screened on the basis of the optimized molecular structures to acquire 3480 molecule descriptors; the organic compounds are divided into a training set and a verification set according to a ratio of 4:1, the training set is used for creating a predication model, and the verification set is used for external verification after model creation. The method has the advantages that the model is clear in application field and covers new pollutants, has good imitative effect, robustness and predication capability, and can effectively predict bio-concentration factors of different types of organic compounds; predication results of the method can provide important data support for risk assessment and management of the organic chemicals and are of great significance in ecological risk assessment.
Description
Technical field
The present invention relates to one kind sets up QSAR (QSAR) to have predicted using arithmetic of linearity regression
The technology of chemical machine product living being concentration ratio, belongs to ecological risk assessment Test Strategy field.
Background technology
The material of some toxic effects can be bioaccumulation, and the transmission for passing through food chain, to mankind itself's health structure
Into potential threat.Biological concentration is a kind of equilibrium assignmen process of the chemical substance between organism and surrounding medium, is referred to
Biology accumulates certain element or refractory organicses material so as to dense in body by way of non-eating from surrounding environment
Phenomenon of the degree more than concentration in surrounding environment.Living being concentration ratio (BCF) can effectively assess polluter and potentially be enriched with
Ability, when it is defined as pollutant and reaches poised state, the ratio of its concentration and concentration in surrounding medium in vivo.
Chemical substance enrichment discipline in vivo is solved and determined, people can be helped to propose to potentially harmful pollution problem earlier
Query, more reasonably formulate the environmental management standard of noxious pollutant, therefore obtain living being concentration ratio (BCF) data to changing
Product risk assessment and management, safeguards that Environmental security and human health etc. all have great importance.
Measuring be at present obtain compound living being concentration ratio (BCF) data an approach, economic cooperation with send out
Continuous-flow type piscine organism enrichment Testing Guideline (OECD guides 305) has been issued in exhibition tissue (OECD) in 1996.But experimental technique
Cycle length (usual 28-60 days), it is costly (EU REACH Legislation formulates basic about 8.5 ten thousand Euros of the testing cost of chemicals, its
Middle bioconcentration belongs to one of very important Testing index in basic detection), violate animal protection principle and (once test about
Need the tail of Experimental fish 100), so cannot meet that more than existing 14 ten thousand kinds of commodity chemicals are carried out with risk assessment and data management
Demand.Therefore, it is necessary to developing experiment substitute technology obtains BCF data, such as Study on Quantitative Structure-Activity Relationship/property relation (QSAR) is pre-
Survey chemicals living being concentration ratio.
The theoretical foundation of QSAR researchs is the determining of structural information such as molecular structure of compounds and intermolecular combination
The property of compound.Material with similar molecular structure, physicochemical property, environmental behaviour, the ecological poison may with similar substance
Effect of science etc..It is that QSAR is established early in the linear free energy relationship (LFER) that earlier 1900s, Hammett et al. are proposed
Thermodynamic Basis, open the theoretical basis research of QSAR.Hansch on the basis of linear free energy relationship (LFER), QSAR
Research range expands to biologically active field.Afterwards Kamlet et al. is expanded out on the basis of linear free energy relationship (LFER) again
Linear solvation energy relationship model (LSER).Used as a kind of believable technical tool, QSAR methods have been successfully applied to pollutant ring
The prediction of border behavioral parameters and toxicologic index.
The forecast model that researcher establishes some organic compound living being concentration ratios using QSAR methods is had at present.
Because material is two similar assigning process from water into fish body to pollutant from water into lipid phase, so earliest
Biological concentration model is often by logKOWSet up.Such as document " J Fish Res.Board Can., l979,36,1040-
1048. " propose earliest based on octanol (KOW) and comprising the linear prediction model of 55 compounds, the mould
The multiple regression coefficient R of type2For 0.90, but model robustness and predictive ability aspect are not estimated.Document
" SAR QSAR Environ.Res., 1993,1,29-39. " establishes first the logBCF-logK comprising 154 compoundsOW
Nonlinear model.As a result show that bilinear model has more excellent degree of fitting than parabola model.Document
" Environ.Toxicol.Chem., 1999,18 (4), 664-672. " then establishes one and covers multiple compounds
logBCF-logKOWModel, though covering 694 compounds, model contains 4 equations, is not easy to practical application.With KOW
The BCF models set up by physical and chemical parameter are more succinct, but KOWThe availability of experimental data itself limits the class model
Using.At the same time, though biological concentration passes through logK with n-octyl alcohol moisture with being two similar process, onlyOWModel is built, no
Be enough to fully characterize biological concentration process.
With the continuous renewal and development of QSAR technologies, model descriptor acquisition and modeling method etc. are all further sent out
Exhibition.If document " Chemosphere, 2000,41,1675-1688. " is using molecular connectivity index and the polarity correction factor, build
The QSAR models of 239 kinds of BCF are found.But there are some researches show may deposit between the molecular connectivity index descriptor included in model
In stronger synteny.Document " SAR QSAR Environ.Res., 2010,7-8, (21), 671-680. " selects hydrophobicity to retouch
Totally 7 descriptors such as symbol, hydrogen bond, molecular topology sex index are stated, with 624 compounds, by artificial neural network (ANN) method
QSAR models are set up, the model does not have clear and definite expression formula, be not easy to mechanism explanation.Document " SAR QSAR
Environ.Res., 2010,21, (7-8), 711-729. " pass through optimal multiple linear regression (best multi-linear
Regression) method, with 473 compounds model is built, and carries out external certificate, but the result with 161 compounds
It is undesirable.
In having studied, some models build for certain particular types compound.Such as document
" Bioorgan.Med.Chem., 2003,11,5045-5050. " proposes PI state indices based on molecular graph theory, establishes one
Living being concentration ratio QSAR models comprising 16 kinds of Polychlorinated biphenyls.Same logKOWInstitute's established model is compared, although what PI indexes were set up
QSAR models are better than logKOWThe model of foundation, but model coefficient correlation not high (r=0.753).Document " Mol.Divers.,
2010,14:67-80. " counted according to relatively electrical and relative bond distance of non-hydrogen atom type, atom belonging, non-hydrogen atom etc.
Calculate, propose molecular electronegativity distance vector (MEDV) descriptor, the QSAR models of nonpolar organic matter are established using the descriptor.
But this kind of method is not easy enough in terms of descriptor computation.Document " Chemosphere, 2012,89,433-444. " passes through 21
PBDE (PBDEs) molecule sets up multiple linear regression model, and model is only more succinct comprising 2 descriptors.Document
" Ecotox.Environ.Safe., 2012,75,213-222. " with 58 Polychlorinated biphenyls (PCBs), by 5 kinds of descriptors, with
Offset minimum binary (PLS) sets up QSAR models, has carried out internal-external checking, and models fitting ability is preferable.But these local modes
Type covers compound and is particular types, and the applicable compound of model is also more single, in Risk Assessment of Chemicals and management
There is certain limitation using in.
In sum, current QSAR models mainly include two big class, Local Model and universal model.The fitting of Local Model
Effect and predictive ability are of a relatively high, but because it builds for particular kind of compound, so cannot meet to big
The acquisition demand of amount variety classes chemicals data.Although universal model meets chemical balance motion quick obtaining variety classes chemical combination
The demand of thing, but still lack that the classes of compounds for covering is more at present, model simple, prediction rule are transparent, be easy to mechanism explains, tool
There are good degree of fitting, robustness, predictive ability, and carried out the BCF-QSAR models of application domain sign.So, having very much must
The QSAR model constructions to propose according to OECD and usage criteria：(1) with clearly defined environmental index, (2) are with clear and definite
Algorithm, (3) define the application domain of model, and (4) model has appropriate degree of fitting, stability and predictive ability, and (5) are best
Mechanism explanation can be carried out, a BCF-QSAR model for covering multiple compounds is set up so as to can comment in chemicals risk
Play a role in valency and the practical application of management.
The content of the invention
The present invention seeks to develop a kind of method of fast prediction organic chemicals living being concentration ratio.The method can be straight
Connect and living being concentration ratio is predicted according to molecular structure of compounds, can make up the deficiency of experimental technique, and then for chemicals risk
Evaluate and manage and necessary basic data is provided.
The QSAR prediction organic chemicals piscine organism enrichment factor of the present invention, the present invention is collected altogether
The living being concentration ratio data of 780 kinds of organic compounds, during collected data are all from disclosed database or publish thesis,
Organic compound includes Polychlorinated biphenyls, PBBs, perfluoroalkyl acid, dioxin, incretion interferent, agricultural chemicals.Fish body kind
Class includes carp, rainbow trout, medaka, fathead minnow, zebra fish.It is the biology for covering compound amounts and most species at present
Enrichment factor QSAR models.
Optimize the molecular structure of 780 kinds of organic compounds initially with density functional theory, tied based on the molecule after optimization
Structure, from 4885 kinds of molecular descriptors of 780 kinds of organic compounds, preliminary screening obtains 3480 descriptors, wherein preliminary screening
Remove the descriptor of constant and approximate constant, remove the descriptor of at least one missing values.
Then by 780 kinds of organic compounds according to 4:1 ratio is divided into training set and checking collection.Training set is pre- for building
Model is surveyed, checking collection is for the external certificate after modeling.Multiple linear regression model is built using stepwise regression method.
According to the characterization parameter of model, descriptor number and variable expansion factor, finally choose with reference to regression result
Model is as follows：
logBCF=2.137+0.061×MLOGP2+0.034×F02[C-Cl]-0.312×nROH-1.282×P-117+
0.323×Mor25m-0.052×N%+0.080×X4v-0.289×O–058-1.137×LLS_01-1.387×H4v+
0.071×SM12_AEA(dm)-0.269×O-057
Wherein, logBCF is the logarithm value of living being concentration ratio, and MLOGP2 represents logKOWSquare, F02 [C-Cl] represent
The frequency that atom pair C and Cl topology distances occur when being 2, the number of nROH representation hydroxies, P-117 represents " X3-P=X " in molecule
The quantity of structure, Mor25m represents the scattering function value that weighted quality is calculated, and N% represents the percentage of N atoms, and X4v is tetravalence
Connectivity Index of Electronic Density, O-058 represents the quantity of "=O " structure in molecule, and LLS_01 is a class medicine index, and H4v represents hydrogen weighting model
The auto-correlation function value of moral China volume delayed 4, SM12_AEA (dm) represents the adjacent square in 12 spectrum distance augmentation edges of weighted dipole moment
Battle array value, O-057 represents the quantity of " OH " in phenol, alcohol molecule.
Gained model training integrates compound number as n=624, and the variable expansion factor (VIF) of each descriptor is respectively less than 5,
The matrix M that independent variable is constituted with dependent variableYXAnd independent variable matrix MXK index of correlation KXYAnd KXXMeet KXX(0.358)<KXY
(0.397), all show that model does not have multiple correlation；(the R that the capability of fitting of model is adjusted by the Jing frees degree2Adj) and
Square error (RMSE) is characterized, R2Adj=0.809, RMSE=0.612, show that the model has good capability of fitting;Model is steady
Strong property by internal verification cross validation coefficient (Q2 LOO) and Bootstrapping methods gained Q2 BOOTEvaluate, Q2 LOO=0.871,
Q2 BOOT=0.797, R2And Q2Difference be much smaller than 0.3, it is believed that there are no Expired Drugs in the model, with good robustness；
In the external authentication process of model, checking collection data amount check next=156, external prediction coefficient R2 ext=0.738,Q2 ext=
0.732, RMSEext=0.736, show that the model has good external prediction ability, can effectively predictive compound biology it is rich
The collection factor.
Model application domain characterizes the Euclidean distance method that adopts, the Europe of a certain compound characteristic vector (descriptor parameters)
A few Reed distances be its arrive descriptor space center point characteristic vector distance, k-th parameter of central point characteristic vector (retouch
State symbol) XkFor the mean value of training set k-th descriptor of all compounds, i.e.,：
Wherein XjkFor k-th descriptor value of compound j, n is training set compound sum.Then a certain compound i is tieed up in M
The Euclidean distance d of the characteristic vector in space (M is adopted Molecular structure descriptor number)iIt is represented by：
X in formulaikFor k-th descriptor value of compound i.Training set compound characteristic vector is calculated by formula (2)
Euclidean distance scope to central point characteristic vector is 0.192～1.438, therefore characteristic vector Euclidean distance is little
Compound in 1.438 is applied to this model.
The invention has the beneficial effects as follows：Model covers 780 kinds of different types of organic compounds, such as emerging pollutant bis-phenol
A, it is organic matter dioxin-like chemical that priority pollutants Polychlorinated biphenyls, PBBs and environmental area are paid close attention to, organic
Agricultural chemicals.Model application domain is wide, and has carried out application domain sign, can be used to predict the living being concentration ratio of variety classes compound,
Can be that Risk Assessment of Chemicals and management work provide basic data；Using multiple linear regression analysis method, 12 kinds of descriptors are chosen
Forecast model is constructed, model simple, prediction rule are transparent, are easy to analysis and understanding and practical application；According to OECD with regard to QSAR
Model construction and it is modeled using directive/guide, institute's established model has good fitting effect, robustness and predictive ability.Using this
The method of invention can quickly and efficiently predict organic chemicals living being concentration ratio.The method is with low cost, easy to be quick,
Manpower, expense and the time that can be saved needed for experiment test.Institute's established model is covering in current all BCF-QSAR models
The most model of compound quantity, and model has good degree of fitting, robustness and predictive ability, while model is answered
With the sign in domain, fully meet OECD model constructions and use directive/guide.Therefore the living being concentration ratio using patent of the present invention is pre-
Result is surveyed, efficiently important data can be provided for organic chemicals risk assessment and management and be supported, to ecological risk assessment
Have great importance.
Description of the drawings
Fig. 1 is the measured value and the fitted figure of predicted value of training set and checking collection.
The Euclidean distance map of Fig. 2 characterization model application domains.
Specific embodiment
Embodiment 1
At random the given chlordene dibenzo of compound 1,2,3,6,7,8--p- dioxin (CAS 57653-85-7), predicts
Its living being concentration ratio.Optimize the molecular structure of 1,2,3,6,7,8- chlordene dibenzo-p- dioxin first, be then based on excellent
The molecular structure of change, calculates 12 kinds of descriptor MLOGP2, FO2 [C-Cl], nROH, P-117, Mor25m, N%, X4v, O-058,
LLS_01, H4v, SM12_AEA (dm), the numerical value of O-057, respectively 20.902,12,0,0,1.239,0,2.907,0,0.67,
0.24,8.004,0.According to formula (2) be calculated characteristic vector Euclidean distance be 0.514 (<1.438), in model
In the range of application domain, can use the chlordene dibenzo of this model prediction 1,2,3,6,7,8--p- dioxin biological concentration because
Son, the value that descriptor value is substituted into into institute's established model logBCF is 3.926, and wherein experiment value is 3.927, is predicted the outcome good.
Embodiment 2
At random given compound bisphenol-A (CAS 80-05-7), predicts its living being concentration ratio.Optimize bisphenol-A first
Molecular structure, is then based on the molecular structure for optimizing, and calculates 12 kinds of descriptor MLOGP2, FO2 [C-Cl], nROH, P-117,
Mor25m, N%, X4v, O-058, LLS_01, H4v, SM12_AEA (dm), the numerical value of O-057, respectively 10.928,0,0,0,
0.375,0,1.923,0,0.83,0.285,6.164,2.The Euclidean distance of characteristic vector is calculated according to formula (2)
For 0.645 (<1.438), in the range of model application domain, the living being concentration ratio of this model prediction bisphenol-A can be used, will be retouched
The value that stating identifier value substitution institute established model has logBCF is 1.639, and wherein experiment value is 1.641, is predicted the outcome good.
Embodiment 3
At random the given trichloroaniline of compound 2,4,6- (CAS 634-93-5), predicts its living being concentration ratio.It is excellent first
Change the molecular structure of 2,4,6- trichloroanilines, be then based on the molecular structure for optimizing, calculate 12 kinds of descriptor MLOGP2, FO2
[C-Cl], nROH, P-117, Mor25m, N%, X4v, O-058, LLS_01, H4v, SM12_AEA (dm), the numerical value of O-057, point
Wei 10.982,6,0,0,0.191,7.1,1.37,0,0.83,0.077,5.225,0.Feature is calculated according to formula (2)
Vector Euclidean distance be 0.267 (<1.438), in the range of model application domain, this model prediction 2,4,6- can be used
The living being concentration ratio of trichloroaniline, it is 2.133 that descriptor value substitute into institute's established model to be had the value of logBCF, and wherein experiment value is
2.001, predict the outcome good.
Embodiment 4
Compound 2,3,3' is given at random, and 4,4', 5- chlordene biphenyl (CAS 38380-08-4) predicts its biological concentration
The factor, optimizes first 2,3,3', and the molecular structure of 4,4', 5- chlordene biphenyl is then based on the molecular structure for optimizing, and calculates 12
Plant descriptor MLOGP2, FO2 [C-Cl], nROH, P-117, Mor25m, N%, X4v, O-058, LLS_01, H4v, SM12_AEA
(dm), the numerical value of O-057, respectively 41.857,12,0,0,0.485,0,2.656,0,0.67,0.08,7.763,0.According to public affairs
Formula (2) be calculated characteristic vector Euclidean distance be 0.695 (<1.438), in the range of model application domain, Ke Yiyun
With this model prediction 2,3,3', the living being concentration ratio of 4,4', 5- chlordene biphenyl, descriptor value is substituted into into institute's established model has
The value of logBCF is 5.146, and wherein experiment value is 5.390, is predicted the outcome good.
Embodiment 5
At random given compound pentachloronitrobenzene (CAS 82-68-8), predicts its living being concentration ratio, and five are optimized first
The molecular structure of chloronitrobenzene, is then based on the molecular structure for optimizing, and calculates 12 kinds of descriptor MLOGP2, FO2 [C-Cl],
NROH, P-117, Mor25m, N%, X4v, O-058, LLS_01, H4v, SM12_AEA (dm), the numerical value of O-057, respectively
18.405,10,0,0, -0.048,7.100,2.020,0,0.830,0.206,7.292,0.Spy is calculated according to formula (2)
Levy vector Euclidean distance be 0.398 (<1.438), in the range of model application domain, this model prediction pentachloro- can be used
The living being concentration ratio of nitrobenzene, it is 2.665 that descriptor value substitute into institute's established model to be had the value of logBCF, and wherein experiment value is
2.738, predict the outcome good.
Embodiment 6
At random the given bromo biphenyl (CAS 59080-33-0) of compound 2,4,6- tri-, predicts its living being concentration ratio, first
The molecular structure of the optimization bromo biphenyls of 2,4,6- tri-, is then based on the molecular structure for optimizing, and calculates 12 kinds of descriptor MLOGP2,
FO2 [C-Cl], nROH, P-117, Mor25m, N%, X4v, O-058, LLS_01, H4v, SM12_AEA (dm), the numerical value of O-057,
Respectively 34.143,0,0,0,0.742,0,4.139,0,0.67,0.295,5.225,0.Feature is calculated according to formula (2)
Vector Euclidean distance be 0.446 (<1.438), in the range of model application domain, this model prediction 2,4,6- can be used
The living being concentration ratio of three bromo biphenyls, it is 3.991 that descriptor value substitute into institute's established model to be had the value of logBCF, and wherein experiment value is
3.873, predict the outcome good.
Embodiment 7
At random given compound perfluoro hexyl sulfonic acid (CAS 355-46-4), predicts its living being concentration ratio, optimizes first
The molecular structure of perfluoro hexyl sulfonic acid, is then based on the molecular structure for optimizing, and calculates 12 kinds of descriptor MLOGP2, FO2 [C-
Cl], nROH, P-117, Mor25m, N%, X4v, O-058, LLS_01, H4v, SM12_AEA (dm), the numerical value of O-057, respectively
8.139,0,1,0,1.144,0,1.369,2,0.5,0.266,13.351,0.Characteristic vector is calculated according to formula (2)
Euclidean distance be 0.663 (<1.438), in the range of model application domain, this model prediction perfluoro hexyl sulfonic acid can be used
Living being concentration ratio, the value that descriptor value is substituted into into institute established model logBCF is 2.233, and wherein experiment value is 2.000, in advance
Survey result good.
Claims (1)
1. a kind of method that employing QSAR predicts the piscine organism enrichment factor of organic chemicals, its feature exists
In the living being concentration ratio data of 780 kinds of organic compounds being collected, using density functional theory to 780 kinds of organic compounds
Molecular structure is optimized, based on the molecular structure after optimization, from 4885 kinds of molecular descriptors of 780 kinds of organic compounds just
Step screening obtains 3480 descriptors；Wherein preliminary screening is to remove the descriptor of constant and approximate constant, removes at least one
The descriptor of individual missing values；Described organic compound includes Polychlorinated biphenyls, PBBs, perfluoroalkyl acid, dioxin, double
Phenol A, agricultural chemicals；
Then by 780 kinds of above-mentioned organic compounds according to 4:1 ratio is divided into training set and checking collection, and training set is used to build
Forecast model, checking collection is for the external certificate after modeling；Multiple linear regression model, the mould are built using stepwise regression method
Type is as follows：
LogBCF=2.137+0.061 × MLOGP2+0.034 × F02 [C-Cl] -0.312 × nROH-1.282 × P-117+
0.323 × Mor25m-0.052 × N%+0.080 × X4v-0.289 × O -058-1.137 × LLS_01-1.387 × H4v+
0.071×SM12_AEA(dm)-0.269×O-057
Wherein, logBCF represents the logarithm of living being concentration ratio, and MLOGP2 represents logKOWSquare, F02 [C-Cl] represents atom
The frequency occurred when to C and Cl topology distances being 2, the number of nROH representation hydroxies, P-117 represents " X3-P=X " knot in molecule
The quantity of structure, Mor25m represents the scattering function value that weighted quality is calculated, and N% represents the percentage of N atoms, and X4v is tetravalence
Connectivity Index of Electronic Density, O-058 represents the quantity of "=O " structure in molecule, and LLS_01 is a class medicine index, and H4v represents hydrogen weighting
The auto-correlation function value of Van der waals volumes delayed 4, SM12_AEA (dm) represents 12 spectrum distance augmentation edges of weighted dipole moment and adjoins
Matrix value, O-057 represents the quantity of " OH " in phenol, alcohol molecule；
Characteristic vector Euclidean distance is not more than 1.438 compound and is applied to this model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410013562.XA CN103761431B (en) | 2014-01-10 | 2014-01-10 | Method for predicting fish bio-concentration factors of organic chemicals by quantitative structure-activity relationship |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410013562.XA CN103761431B (en) | 2014-01-10 | 2014-01-10 | Method for predicting fish bio-concentration factors of organic chemicals by quantitative structure-activity relationship |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103761431A CN103761431A (en) | 2014-04-30 |
| CN103761431B true CN103761431B (en) | 2017-04-12 |
Family
ID=50528667
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410013562.XA Active CN103761431B (en) | 2014-01-10 | 2014-01-10 | Method for predicting fish bio-concentration factors of organic chemicals by quantitative structure-activity relationship |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103761431B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104376221B (en) * | 2014-11-21 | 2018-06-15 | 环境保护部南京环境科学研究所 | A kind of method for the skin permeability coefficient for predicting organic chemicals |
| CN105468914A (en) * | 2015-11-26 | 2016-04-06 | 昆明理工大学 | Method for predicting hydrolysis rate of sulfur-containing organic compounds in atmosphere |
| CN105548463B (en) * | 2015-11-26 | 2017-11-10 | 昆明理工大学 | A kind of method of the sulfur-containing compound rate of adsorption in prediction air |
| CN105893783B (en) * | 2016-06-03 | 2018-05-01 | 中国农业大学 | A kind of novel agrochemical screening technique |
| CN108627819B (en) * | 2018-05-11 | 2020-09-25 | 清华大学 | Radar observation-based distance extension target detection method and system |
| CN111312340A (en) * | 2018-12-12 | 2020-06-19 | 深圳市云网拜特科技有限公司 | SMILES-based quantitative structure effect method and device |
| CN110146695B (en) * | 2019-05-08 | 2021-12-10 | 南京理工大学 | Method for screening human transthyretin interferent by adopting k nearest neighbor algorithm |
| CN110993019B (en) * | 2019-10-25 | 2020-11-27 | 北京师范大学 | Biological enrichment factor prediction method for establishing water quality benchmark and risk assessment |
| CN110853701A (en) * | 2019-11-07 | 2020-02-28 | 大连理工大学 | Method for predicting fish biological enrichment factor of organic compound by adopting multi-parameter linear free energy relation model |
| CN110910970B (en) * | 2019-11-20 | 2022-05-13 | 大连理工大学 | Method for predicting toxicity of chemicals by taking zebra fish embryos as receptors through building QSAR model |
| CN111261238A (en) * | 2020-01-15 | 2020-06-09 | 生态环境部南京环境科学研究所 | Construction method of PPCPs organic chemical mesophilic anaerobic digestion removal rate prediction model |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101673321A (en) * | 2009-10-17 | 2010-03-17 | 大连理工大学 | Method for fast predicting organic pollutant n-caprylic alcohol/air distribution coefficient based on molecular structure |
| CN103345544A (en) * | 2013-06-11 | 2013-10-09 | 大连理工大学 | Predicting organic chemical biodegradability according to logistic regression method |
-
2014
- 2014-01-10 CN CN201410013562.XA patent/CN103761431B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101673321A (en) * | 2009-10-17 | 2010-03-17 | 大连理工大学 | Method for fast predicting organic pollutant n-caprylic alcohol/air distribution coefficient based on molecular structure |
| CN103345544A (en) * | 2013-06-11 | 2013-10-09 | 大连理工大学 | Predicting organic chemical biodegradability according to logistic regression method |
Non-Patent Citations (1)
| Title |
|---|
| 卤代有机化合物生物富集因子的定量结构-活性关系模型;郑玉婷 等;《生态毒理学报》;20131015;第8卷(第5期);第772-777页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103761431A (en) | 2014-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103761431B (en) | Method for predicting fish bio-concentration factors of organic chemicals by quantitative structure-activity relationship | |
| Grech et al. | Toxicokinetic models and related tools in environmental risk assessment of chemicals | |
| Basant et al. | Linear and nonlinear modeling for simultaneous prediction of dissolved oxygen and biochemical oxygen demand of the surface water—a case study | |
| Karaouzas et al. | Assessing the ecological effects of water stress and pollution in a temporary river-Implications for water management | |
| Brucet et al. | Fish diversity in E uropean lakes: geographical factors dominate over anthropogenic pressures | |
| Cunha et al. | A trophic state index for tropical/subtropical reservoirs (TSItsr) | |
| Zhang et al. | Ecological vulnerability assessment based on multi-sources data and SD model in Yinma River Basin, China | |
| Zhang et al. | Research development, current hotspots, and future directions of water research based on MODIS images: A critical review with a bibliometric analysis | |
| Zheng et al. | Quantifying the spatial impact of landscape fragmentation on habitat quality: A multi-temporal dimensional comparison between the Yangtze River Economic Belt and Yellow River Basin of China | |
| Lubetkin et al. | Multi‐source mixing models to quantify food web sources and pathways | |
| Brown et al. | The metabolic theory of ecology and the role of body size in marine and freshwater ecosystems | |
| Mykrä et al. | Variability of lotic macroinvertebrate assemblages and stream habitat characteristics across hierarchical landscape classifications | |
| Yates et al. | Multi‐scaled drivers of rural prairie stream metabolism along human activity gradients | |
| Radomyski et al. | Bioaccumulation of trace metals in aquatic food web. A case study, Liaodong Bay, NE China | |
| Mostafavi et al. | Predicting presence and absence of trout (Salmo trutta) in Iran | |
| Brown et al. | Predicting river ecosystem response to glacial meltwater dynamics: a case study of quantitative water sourcing and glaciality index approaches | |
| Núñez et al. | Critical analysis of life cycle impact assessment methods addressing consequences of freshwater use on ecosystems and recommendations for future method development | |
| Radomyski et al. | Modelling ecological and human exposure to POPs in Venice lagoon–Part II: quantitative uncertainty and sensitivity analysis in coupled exposure models | |
| Kong et al. | High frequency data provide new insights into evaluating and modeling nitrogen retention in reservoirs | |
| Gutiérrez‐Estrada et al. | Estimating fish community diversity from environmental features in the Tagus estuary (Portugal): multiple linear regression and artificial neural network approaches | |
| Li et al. | Ecological exergy as an indicator of land-use impacts on functional guilds in river ecosystems | |
| Davies et al. | Niche shift and resource supplementation facilitate an amphibian range expansion | |
| Kaller et al. | Aquatic vegetation mediates the relationship between hydrologic connectivity and water quality in a managed floodplain | |
| Wang et al. | A SEEC model based on the DPSIR framework approach for watershed ecological security risk assessment: a case study in Northwest China | |
| Yang et al. | In silico toxicity evaluation of dioxins using structure–activity relationship (SAR) and two-dimensional quantitative structure–activity relationship (2D-QSAR) |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |