CN108959842A - Method is ranked for high-precision energy in organic molecular crystals structure prediction - Google Patents
Method is ranked for high-precision energy in organic molecular crystals structure prediction Download PDFInfo
- Publication number
- CN108959842A CN108959842A CN201810420909.0A CN201810420909A CN108959842A CN 108959842 A CN108959842 A CN 108959842A CN 201810420909 A CN201810420909 A CN 201810420909A CN 108959842 A CN108959842 A CN 108959842A
- Authority
- CN
- China
- Prior art keywords
- energy
- molecule
- single cell
- central single
- radius
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The invention belongs to organic molecular crystals technical field of structures, and in particular to rank method for high-precision energy in organic molecular crystals structure prediction, the quantum mechanics operating radius including determining central single cell；In central single cell, made energy calculation using density piecemeal interaction algorithm；In the range of radius R, the molecule outside central single cell can calculate the effect of molecule in central single cell under quantum mechanics precision；Under classical molecular mechanics precision conditions, calculate except radius R range, effect energy of the cyclic extension unit cell to molecule in central single cell；Crystal gross energy is calculated, crystal gross energy includes the energy between energy and central single cell and cyclic extension unit cell in central single cell.The present invention improves energy balane efficiency in drug molecule crystal, accurately energy back, predicts process in crystal, and guidance CSP is found real low energy advantage crystal form in the proper direction.
Description
Technical field
The invention belongs to organic molecular crystals technical field of structures, and in particular in organic molecular crystals structure prediction
High-precision energy ranks method.
Background technique
Compound has the characteristics that form different crystal structures referred to as polymorphism.And the key of compound itself
Physicochemical property such as density, form, solubility, the strong influence by its crystal form such as rate of dissolution.For drug, crystal form can be strong
The strong bioavilability for influencing drug, the final treatment performance for influencing drug.Experiment polymorph medicine screening has become standard medicine
An essential link in object R&D process.In experiment, crucial crystallization ginseng is arranged manually or by robot hand in people
Number, but correctly crystallization condition is difficult to obtain in a short time by experiment.A kind of selectable scheme is to pass through computer mould
It is quasi- that crystal structure prediction (CSP) is carried out to drug molecule, potential plurality of stable crystal form is found, then for a small number of with clearly defined objective
Potential crystal form is tested.
In the past ten years, inorganic and organic crystal prediction (CSP) all achieves huge progress.Although there are many phases
Like place, but the prediction of inorganic and organic crystal is needed in face of completely different challenge.In inorganic CSP, there is a concern that bonding
And electronic property, and organic C SP changes more concerned with structure and phase transformation.Medicament research and development is then related with the CSP of organic molecule.This neck
At present there are two significant challenge, one is the completeness sampled to crystal space in domain, the other is the ranking of crystal structure energy
Accuracy.First challenge has made great progress, but second point, and the precision energy ranking of structure is still to restrict CSP
The bottleneck of large-scale application in pharmaceutical industries.
The de minimis energy difference accurately calculated between different low energy crystal structures needs high-precision Quantum mechanical calculation, and high-precision
The time complexity of degree Quantum mechanical calculation is O (N with the growth of system electronic number N3)~O (N4), when independent in asymmetric unit
Molecules increase, and when space group operand increases, and carry out high-precision energy to a large amount of crystal structures generated during CSP
Calculate the bottleneck for just becoming CSP.
Summary of the invention
In view of the above technical problems, the present invention provide generate during a kind of pair of organic molecular crystals structure prediction it is a large amount of
The method that crystal structure carries out quick precision energy calculating, to improve the efficiency and precision of the ranking of crystal structure energy.In order to reach
To object above, density piecemeal interaction algorithm and period used in the calculating of isolated macromolecular cluster maximum system energy are used
Property system the method that combines of QM/MM, devise the efficient accurate energy meter for being suitable for periodical organic molecular crystals system
Calculation method.Designed frame, is applicable to the first-principles calculations method of various quantum mechanics precision in this method calculating,
Functional and base group.
Used technical solution are as follows:
Method is ranked for high-precision energy in organic molecular crystals structure prediction, comprising the following steps:
(1) the quantum mechanics operating radius of central single cell is determined
Calculate central single cell in all atoms of intramolecular and surrounding extension unit cell molecule in atom van der Waals radius and,
Using this radius and add 1.5 angstroms as truncation, the nearest circle molecule of distance center unit cell is found, with the geometric center of central single cell
Into this circle molecule, farthest distance is as quantum mechanics operating radius R；
(2) it in central single cell, is made energy calculation using density piecemeal interaction algorithm
Under quantum mechanics precision, the energy of each molecule is calculated separately, the electrostatic potential generated at this time comprising remaining molecule
Effect, electrostatic potential are generated by the atom nuclear potential and electron density distribution of remaining each molecule, and the electronics of electrostatic potential and each molecule is close
Degree distribution is by iteration each other until convergence obtains；Energy after iteration convergence are as follows:
ENERGY E in central single celliso[ρ], wherein ρ represents electron density, this energy is made of three parts: first item
It is the energy summation of each isolated molecule, whereinIt is the energy of molecule i, ρiIt is the electron density of molecule i；Section 2 generation
Electrostatic interaction between table molecule, the electrostatic interaction energy between the atomic nucleus including molecule i and molecule jAnd electricity
Electrostatic interaction energy between sub- densitySection 3 is general respectively from the exchange correlation between each molecule
The Nonlinear Superposition error delta E of letterxcWith the Nonlinear Superposition error delta T of kinetic energy functionals；The parameter meaning of following formula is same should
Formula；
(3) in the range of radius R, the molecule outside central single cell can be in quantum mechanics essence to the effect of molecule in central single cell
Degree is lower to be calculated, and still using the process of the iterative solution of step 2, but only calculates effect of this moieties to central single cell molecule
Can:
Effect energy E of the molecule to central single cell moleculeinter_ij_QMIt include: first item is electrostatic phase interaction between atomic nucleus
With Section 2 is the interaction between electronics；
(4) it under classical molecular mechanics precision conditions, calculates except radius R range, cyclic extension unit cell is to center list
The effect energy of molecule intracellular；The long-range electrostatic that this portion of energy is generated by the molecule in cyclic extension structure cell in central single cell
Gesture integrates to obtain to electron density total in central single cell, its calculation formula is:
WhereinIt represents outside radius R, what molecule j and its all periodical mirror image molecules generated at central single cell
Electrostatic potential, ρiIt is the electron density of molecule i in central single cell；
(5) crystal gross energy is calculated, crystal gross energy includes energy in central single cell and central single cell and periodicity
Extend the energy between unit cell:
E=Eiso+Eperiodic
Wherein, the energy that the energy of central single cell uses (2) step to calculate, and cyclic extension unit cell is to central single cell
Effect can include the energy that calculates of (3) step and (4) step with:
Wherein, Eitner_ij_QMThat is calculated energy in (3) step, condition for peaceIt indicates
Distance is less than the radius R that (1) step determines between molecule i and molecule j, and molecule j is in central single cell, Einter_ij_MMIt is
(4) energy calculated in step.
Method is ranked provided by the present invention for high-precision energy in organic molecular crystals structure prediction, the technology effect having
Fruit has:
(1) energy balane efficiency in drug molecule crystal is improved, when by the calculating of crystal structure quantum mechanics levels of precision
Between complexity from O (N3)~O (N4) drop to O (N)；
(2) it has ensured the correctness of the crystal direction of search, accurately energy back, has predicted process in crystal, CSP will be guided
Real low energy advantage crystal form is found in the proper direction.
Detailed description of the invention
Fig. 1 is that embodiment determines quantum mechanics operating radius schematic diagram；
Fig. 2 is the density piecemeal interaction algorithm flow chart of embodiment.
Specific embodiment
It is described in conjunction with the embodiments the specific technical solution of the present invention.
Method is ranked for high-precision energy in organic molecular crystals structure prediction, comprising the following steps:
(1) the quantum mechanics operating radius of central single cell is determined
Calculate central single cell in all atoms of intramolecular and surrounding extension unit cell molecule in atom van der Waals radius and,
Using this radius and add 1.5 angstroms as truncation, the nearest circle molecule of distance center unit cell is found, with the geometric center of central single cell
Into this circle molecule, farthest distance is as quantum mechanics operating radius R；As shown in Figure 1；
(2) it in central single cell, is made energy calculation using density piecemeal interaction algorithm
Under quantum mechanics precision, the energy of each molecule is calculated separately, the electrostatic potential generated at this time comprising remaining molecule
Effect, electrostatic potential are generated by the atom nuclear potential and electron density distribution of remaining each molecule, and the electronics of electrostatic potential and each molecule is close
Degree distribution is by iteration each other until convergence obtains；As shown in Figure 2；Energy after iteration convergence are as follows:
ENERGY E in central single celliso[ρ], wherein ρ represents electron density, this energy is made of three parts: first item
It is the energy summation of each isolated molecule, whereinIt is the energy of molecule i, ρiIt is the electron density of molecule i；Section 2 generation
Electrostatic interaction between table molecule, the electrostatic interaction energy between the atomic nucleus including molecule i and molecule jAnd electricity
Electrostatic interaction energy between sub- densitySection 3 is general respectively from the exchange correlation between each molecule
The Nonlinear Superposition error delta E of letterxcWith the Nonlinear Superposition error delta T of kinetic energy functionals；The parameter meaning of following formula is same should
Formula；
(3) in the range of radius R, the molecule outside central single cell can be in quantum mechanics essence to the effect of molecule in central single cell
Degree is lower to be calculated, as shown in Figure 1, still using the process of the iterative solution of step 2, but this moieties is only calculated to central single cell point
The effect energy of son:
Wherein effect energy E of the molecule to central single cell moleculeinter_ij_QMIt include: first item is electrostatic phase between atomic nucleus
Interaction, Section 2 are the interactions between electronics；
(4) it under classical molecular mechanics precision conditions, calculates except radius R range, cyclic extension unit cell is to center list
The effect energy of molecule intracellular；The long-range electrostatic that this portion of energy is generated by the molecule in cyclic extension structure cell in central single cell
Gesture integrates to obtain to electron density total in central single cell, its calculation formula is:
WhereinIt represents outside radius R, what molecule j and its all periodical mirror image molecules generated at central single cell
Electrostatic potential, ρiIt is the electron density of molecule i in central single cell；
(5) crystal gross energy is calculated, crystal gross energy includes energy in central single cell and central single cell and periodicity
Extend the energy between unit cell:
E=Eiso+Eperiodic
Wherein, the energy that the energy of central single cell uses (2) step to calculate, and cyclic extension unit cell is to central single cell
Effect can include the energy that calculates of (3) step and (4) step with:
Wherein, Eitner_ij_QMThat is calculated energy in (3) step, condition for peaceIt indicates
Distance is less than the radius R that (1) step determines between molecule i and molecule j, and molecule j is in central single cell, Einter_ij_MMIt is
(4) energy calculated in step.
Claims (1)
1. ranking method for high-precision energy in organic molecular crystals structure prediction, which comprises the following steps:
(1) the quantum mechanics operating radius of central single cell is determined
Calculate central single cell in all atoms of intramolecular and surrounding extension unit cell molecule in atom van der Waals radius and, with this
Radius and plus 1.5 angstroms as truncation, find the nearest circle molecule of distance center unit cell, this arrived with the geometric center of central single cell
Farthest distance is as quantum mechanics operating radius R in one circle molecule；
(2) it in central single cell, is made energy calculation using density piecemeal interaction algorithm
Under quantum mechanics precision, the energy of each molecule is calculated separately, the electrostatic potential effect generated at this time comprising remaining molecule,
Electrostatic potential is generated by the atom nuclear potential and electron density distribution of remaining each molecule, the electron density distribution of electrostatic potential and each molecule
By iteration each other until convergence obtains；Energy after iteration convergence are as follows:
Energy in central single cell, whereinRepresent electron density, this energy is made of three parts: first item is every
The energy of a isolated molecule is summed, whereinIt is the energy of molecule i,It is the electron density of molecule i；Section 2, which represents, divides
Electrostatic interaction between son, the electrostatic interaction energy between the atomic nucleus including molecule i and molecule jIt is close with electronics
Electrostatic interaction energy between degree；Section 3 is respectively from the exchange correlation functional between each molecule
Nonlinear Superposition errorWith the Nonlinear Superposition error of kinetic energy functional；The parameter meaning of following formula is the same as the formula；
(3) in the range of radius R, the molecule outside central single cell can be under quantum mechanics precision to the effect of molecule in central single cell
It calculates, still using the process of the iterative solution of step 2, but only calculates this moieties to the effect energy of central single cell molecule:
Effect energy of the molecule to central single cell moleculeIt include: first item is that electrostatic between atomic nucleus is mutual
Effect, Section 2 is the interaction between electronics；
(4) it under classical molecular mechanics precision conditions, calculates except radius R range, cyclic extension unit cell is in central single cell
The effect energy of molecule；The long-range electrostatic potential pair that this portion of energy is generated by the molecule in cyclic extension structure cell in central single cell
Total electron density integrates to obtain in central single cell, its calculation formula is:
WhereinIt represents outside radius R, the electrostatic potential that molecule j and its all periodical mirror image molecules generate at central single cell
Distribution,It is the electron density of molecule i in central single cell；
(5) crystal gross energy is calculated, crystal gross energy includes the energy in central single cellAnd central single cell and periodicity
Extend the energy between unit cell:
Wherein, the energy that the energy of central single cell uses (2) step to calculate, and effect of the cyclic extension unit cell to central single cell
Can include energy that (3) step and (4) step calculate and:
；
Wherein,That is calculated energy in (3) step, condition for peaceIt indicates to divide
Distance is less than the radius R that (1) step determines between sub- i and molecule j, and molecule j is in central single cell,It is
(4) energy calculated in step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810420909.0A CN108959842B (en) | 2018-05-04 | 2018-05-04 | High-precision energy ranking method for organic molecular crystal structure prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810420909.0A CN108959842B (en) | 2018-05-04 | 2018-05-04 | High-precision energy ranking method for organic molecular crystal structure prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108959842A true CN108959842A (en) | 2018-12-07 |
| CN108959842B CN108959842B (en) | 2021-07-02 |
Family
ID=64498974
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810420909.0A Active CN108959842B (en) | 2018-05-04 | 2018-05-04 | High-precision energy ranking method for organic molecular crystal structure prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108959842B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109935278A (en) * | 2019-02-28 | 2019-06-25 | 深圳晶泰科技有限公司 | Crystal structure collides rapid detection method |
| CN110634537A (en) * | 2019-07-24 | 2019-12-31 | 深圳晶泰科技有限公司 | Double-layer neural network algorithm for high-precision energy calculation of organic molecular crystal structure |
| WO2023115639A1 (en) * | 2021-12-24 | 2023-06-29 | 深圳晶泰科技有限公司 | Molecular crystal energy calculation method and apparatus and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1577822A1 (en) * | 2004-03-15 | 2005-09-21 | Avant-Garde Material Simulation SARL | Method for energy ranking of molecular crystals using DFT calculations and empirical van der Waals potentials |
| WO2008071540A1 (en) * | 2006-12-11 | 2008-06-19 | Avant-Garde Materials Simulation Sarl | Tailor-made force fields for crystal structure prediction |
| EP3040890A1 (en) * | 2014-12-31 | 2016-07-06 | Rafal Michalski | System for the optimization of a method for determining material properties at finding materials having defined properties and the optimization of the said method |
| CN107366018A (en) * | 2017-07-12 | 2017-11-21 | 宁波大学 | A kind of rare earth halide mixing scintillation crystal and preparation method thereof |
| CN107942270A (en) * | 2017-10-20 | 2018-04-20 | 昆明理工大学 | A kind of method using computer determination hexagonal crystal system metal oxide magnetic type |
-
2018
- 2018-05-04 CN CN201810420909.0A patent/CN108959842B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1577822A1 (en) * | 2004-03-15 | 2005-09-21 | Avant-Garde Material Simulation SARL | Method for energy ranking of molecular crystals using DFT calculations and empirical van der Waals potentials |
| WO2008071540A1 (en) * | 2006-12-11 | 2008-06-19 | Avant-Garde Materials Simulation Sarl | Tailor-made force fields for crystal structure prediction |
| EP3040890A1 (en) * | 2014-12-31 | 2016-07-06 | Rafal Michalski | System for the optimization of a method for determining material properties at finding materials having defined properties and the optimization of the said method |
| CN107366018A (en) * | 2017-07-12 | 2017-11-21 | 宁波大学 | A kind of rare earth halide mixing scintillation crystal and preparation method thereof |
| CN107942270A (en) * | 2017-10-20 | 2018-04-20 | 昆明理工大学 | A kind of method using computer determination hexagonal crystal system metal oxide magnetic type |
Non-Patent Citations (2)
| Title |
|---|
| MARCUS A. NEUMANN.ET AL: "Energy Ranking of Molecular Crystals Using Density Functional Theory Calculations and an Empirical van der Waals Correction", 《ACS》 * |
| 姬广富等: "TABA晶体结构的周期性密度泛函理论研究员", 《化学学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109935278A (en) * | 2019-02-28 | 2019-06-25 | 深圳晶泰科技有限公司 | Crystal structure collides rapid detection method |
| CN109935278B (en) * | 2019-02-28 | 2023-04-07 | 深圳晶泰科技有限公司 | Method for rapidly detecting crystal structure collision |
| CN110634537A (en) * | 2019-07-24 | 2019-12-31 | 深圳晶泰科技有限公司 | Double-layer neural network algorithm for high-precision energy calculation of organic molecular crystal structure |
| WO2020164239A1 (en) * | 2019-07-24 | 2020-08-20 | 深圳晶泰科技有限公司 | High-precision dual layer neural network algorithm used for calculating energy of organic molecular crystal structure |
| CN110634537B (en) * | 2019-07-24 | 2022-03-18 | 深圳晶泰科技有限公司 | Double-layer neural net method for high-precision energy calculation of organic molecular crystal structure |
| WO2023115639A1 (en) * | 2021-12-24 | 2023-06-29 | 深圳晶泰科技有限公司 | Molecular crystal energy calculation method and apparatus and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108959842B (en) | 2021-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Matsubara | Nonlinear perturbation theory with halo bias and redshift-space distortions via the Lagrangian picture | |
| Matsika | Electronic structure methods for the description of nonadiabatic effects and conical intersections | |
| CN108959842A (en) | Method is ranked for high-precision energy in organic molecular crystals structure prediction | |
| CN110634537B (en) | Double-layer neural net method for high-precision energy calculation of organic molecular crystal structure | |
| Asenjo et al. | Visualizing basins of attraction for different minimization algorithms | |
| Hsin et al. | Energy transfer dynamics in an RC–LH1–PufX tubular photosynthetic membrane | |
| Qi et al. | United polarizable multipole water model for molecular mechanics simulation | |
| Burn et al. | Creating Gaussian process regression models for molecular simulations using adaptive sampling | |
| Bircher et al. | Improved description of atomic environments using low-cost polynomial functions with compact support | |
| Krog et al. | Bayesian model selection with fractional Brownian motion | |
| Phipps et al. | Energy decomposition analysis based on absolutely localized molecular orbitals for large-scale density functional theory calculations in drug design | |
| Izairi et al. | Comparison study of polar and nonpolar contributions to solvation free energy | |
| Pérez de Alba Ortíz et al. | The adaptive path collective variable: a versatile biasing approach to compute the average transition path and free energy of molecular transitions | |
| Cao et al. | Recent developments in using molecular dynamics simulation techniques to study biomolecules | |
| Dupuis et al. | A polarizable mixed Hamiltonian model of electronic structure for micro-solvated excited states. I. Energy and gradients formulation and application to formaldehyde (1 A 2) | |
| Guan et al. | Enabling a unified description of both internal conversion and intersystem crossing in formaldehyde: A global coupled Quasi-Diabatic hamiltonian for its S0, S1, and T1 states | |
| Fischer et al. | Measurement of observables sensitive to coherence effects in hadronic Z decays with the OPAL detector at LEP | |
| Zhu et al. | Graph theoretic molecular fragmentation for multidimensional potential energy surfaces yield an adaptive and general transfer machine learning protocol | |
| Korth et al. | The Lithium–Thiophene Riddle Revisited | |
| Hofstetter et al. | Graph-convolutional neural networks for (QM) ML/MM molecular dynamics simulations | |
| Matin et al. | Machine Learning Potentials with the Iterative Boltzmann Inversion: Training to Experiment | |
| Comitani et al. | Mapping the conformational free energy of aspartic acid in the gas phase and in aqueous solution | |
| Cao et al. | Artificial neural network potential for gold clusters | |
| Lenner et al. | Continuous tempering molecular dynamics: a deterministic approach to simulated tempering | |
| Li et al. | DeepNCI: DFT noncovalent interaction correction with transferable multimodal three-dimensional convolutional neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 3 / F, Shunfeng industrial building, No.2 Hongliu Road, Fubao community, Fubao street, Futian District, Shenzhen City, Guangdong Province Applicant after: Shenzhen Jingtai Technology Co.,Ltd. Address before: 518000 workshop, 4th floor, building 9, Hualian Industrial Zone, 91 Huaning Road, Dalang street, Longhua District, Shenzhen City, Guangdong Province Applicant before: Shenzhen Jingtai Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |