Search

CN-121583389-B - Three-dimensional modeling and optimizing processing method for medicine molecular structure

CN121583389BCN 121583389 BCN121583389 BCN 121583389BCN-121583389-B

Abstract

The invention relates to the technical field of pharmaceutical chemistry and bioinformatics, in particular to a three-dimensional modeling and optimizing processing method of a pharmaceutical molecular structure, which comprises the steps of constructing a three-dimensional virtual space containing pathogens and pharmaceutical agents; the method comprises the steps of setting connection logic based on distance and energy, enabling a drug intelligent agent to self-assemble to form a molecular cluster when conditions are met, simulating operation dynamics, driving the molecular cluster to form a space surrounding on the surface of a pathogen, calculating the surface coverage rate of the cluster on key sites of the pathogen, generating fitness scores, updating the topological structure of the drug molecule by utilizing an evolution algorithm based on the scores, and iteratively generating an optimal structure. According to the invention, a multi-agent cooperative mechanism is introduced into a calculation model, and the drug self-assembly and virus space are enclosed and brought into screening indexes, so that the method can find a novel drug structure for physically blocking virus infection by forming a large-volume cluster, and the problem of screening omission in the traditional method is effectively solved.

Inventors

  • LAI HUIFANG

Assignees

  • 漳州卫生职业学院

Dates

Publication Date
20260508
Application Date
20260127

Claims (7)

  1. 1. A three-dimensional modeling and optimizing processing method for a medicine molecular structure is characterized by comprising the following steps: s1, constructing a three-dimensional virtual space in computing equipment, mapping a target pathogen into a target object, and mapping an initial candidate drug molecule into a drug agent object with autonomous motion attribute; s2, setting a connection logic between drug molecules, wherein the connection logic is used for monitoring the relative states between drug intelligent agent objects in real time, establishing virtual rigidity constraint when preset combination conditions are met, and locking a plurality of drug intelligent agent objects into molecule cluster objects which move together; S3, performing kinetic simulation in the three-dimensional virtual space, controlling the drug intelligent object or the molecular cluster object to update the position based on a physical potential energy field, and driving the drug intelligent object or the molecular cluster object to form a space surrounding on the surface of the target object; S4, calculating the surface coverage rate of the molecular cluster object to the key site of the target object, and generating an adaptability score of the current drug molecule based on the surface coverage rate; S5, updating the topological structure of the drug molecules by using an evolution algorithm based on the fitness score, generating a new generation drug intelligent object, and returning to the step S2 until a preset convergence condition is met, and outputting an optimal drug molecular structure; In the step S3, the operation dynamics simulation specifically includes: calculating a resultant force vector born by each object for each simulation frame, wherein the resultant force vector at least comprises an electrostatic attraction component pointing to the target object and a random disturbance component simulating thermal motion; updating the space coordinates and the gesture of the object according to the resultant force vector by utilizing a numerical integration algorithm; In the step S3, controlling the pharmaceutical agent object or the molecular cluster object to update the position further includes posture adjustment logic: When detecting that the distance between the drug intelligent object or the molecular cluster object and the target object is smaller than a preset sensing threshold value, calculating the electrostatic moment received by the drug intelligent object or the molecular cluster object based on the electric dipole moment of the drug intelligent object or the molecular cluster object and the electric potential gradient of the surface of the target object; and applying the torque to perform rotation transformation on the drug intelligent object or the molecular cluster object so as to enable the binding site to face the target object.
  2. 2. The method for three-dimensional modeling and optimizing pharmaceutical molecular structure according to claim 1, wherein in the step S2, the preset binding conditions specifically include: extracting charge distribution data of a contact surface, and calculating electrostatic potential products of contact areas; and when the Euclidean distance between two drug agent objects is smaller than the Van der Waals contact threshold, and the electrostatic potential product is negative and meets the energy threshold requirement, judging that the preset combination condition is met.
  3. 3. The method for three-dimensional modeling and optimizing pharmaceutical molecular structure according to claim 1, wherein in the step S2, a virtual rigidity constraint is established, specifically comprising: The first and second drug agent objects are defined in the physical engine as a rigid body combination, and the relative positions and relative angles of the two with respect to a common centroid are locked such that forces applied to the rigid body combination in subsequent simulations produce an overall translational or rotational motion.
  4. 4. The method according to claim 1, wherein in the step S2, the connection logic further comprises a fracture determination: in each simulation frame, calculating the external shear stress born by the virtual rigid constraint in the molecular cluster object, and calculating the equivalent deformation energy corresponding to the virtual rigid constraint in the molecular cluster object by combining the geometric parameters of the constraint action domain; and when the equivalent deformation energy is larger than the binding energy threshold, releasing the virtual rigidity constraint, and decomposing the molecular cluster object into independent drug intelligent object.
  5. 5. The method for three-dimensional modeling and optimizing a molecular structure of a drug according to claim 1, wherein in the step S1, constructing an initialization simulation space further comprises: Generating a plurality of groups of environment interference particles in the three-dimensional virtual space, wherein the environment interference particles have preset quality attributes and initial velocity vectors; In the simulation process of step S3, when the coordinates of the environmental interference particles coincide with the coordinates of the drug agent or the molecular cluster object, a speed vector after collision is calculated according to a law of conservation of momentum, so that the motion trail of the drug agent or the molecular cluster object is changed.
  6. 6. The method for three-dimensional modeling and optimizing pharmaceutical molecular structure according to claim 1, wherein in the step S4, the fitness score generating logic is as follows: calculating the fitness score by adopting a weighted linear combination mode; The variables of the weighted linear function include at least the surface coverage, the sum of bond energies of all virtual rigid constraints within the molecular cluster object, and the number of drug agents that make up the molecular cluster object.
  7. 7. The method for three-dimensional modeling and optimizing a molecular structure of a drug according to claim 1, wherein in step S5, the topology of the drug molecule is updated by using an evolution algorithm, and the method specifically comprises: Converting the topological structure of the high-fitness scoring molecules into a molecular diagram; Identifying non-framework nodes in the molecular diagram, the non-framework nodes being defined as side chain atoms or terminal atoms that do not belong to the molecular pharmacophore framework; Randomly selecting one non-skeleton node, replacing the non-skeleton node with an atom or a functional group fragment with higher hydrophobicity parameters so as to enhance the intermolecular binding capacity and generate topological structure data of the child molecules.

Description

Three-dimensional modeling and optimizing processing method for medicine molecular structure Technical Field The invention relates to the technical field of pharmaceutical chemistry and bioinformatics, in particular to a three-dimensional modeling and optimizing processing method for a pharmaceutical molecular structure. Background In modern drug development, computer Aided Drug Design (CADD) has become a key technology for shortening the development cycle and reducing the cost. The current mainstream technology mainly includes structure-based drug design (SBDD), which is typically represented by molecular docking (Molecular Docking) and molecular dynamics simulation (MD). The prior art scheme generally comprises the steps of obtaining a static crystal structure of a virus target protein, calculating the binding free energy of a single drug small molecule ligand and a target active pocket by using a scoring function, and screening potential drugs by using software such as Autodock or Schr dinger and the like to screen by using a 'key locking model'. However, the prior art suffers from the following significant drawbacks: First, the population synergy is ignored (SWARMEFFECT), and existing methods are mostly based on a "one-to-one" binding pattern for prediction. However, in the case of high concentration local administration in a real physiological environment, drug molecules tend to interact poorly with each other (e.g., van der Waals forces, Stacking) to form dimers or polymers (Cluster), which may wrap the virus surface through a larger steric hindrance volume, resulting in an inhibition effect that the monomer does not possess, and the prior art cannot simulate such a dynamic assembly process, resulting in the leakage of drugs that have weak monomer binding force but strong group blocking ability; Secondly, the static optimization path is single, the traditional structure-activity relationship (QSAR) analysis often relies on manual experience to modify functional groups, and a closed-loop mechanism capable of automatically reversely evolving a molecular structure according to space surrounding capacity is lacked, so that the structure optimization efficiency is low. Therefore, a three-dimensional modeling and optimizing processing method for the molecular structure of the medicine is provided. Disclosure of Invention The invention aims to provide a three-dimensional modeling and optimizing processing method for a medicine molecular structure, which aims to solve the technical problems that the existing medicine design method ignores a cooperative surrounding effect after medicine molecules form clusters in a real environment and lacks an automatic structure evolution mechanism based on a steric hindrance sealing effect. In order to solve the technical problems, the invention aims to provide a three-dimensional modeling and optimizing processing method for a medicine molecular structure, which comprises the following steps: s1, constructing a three-dimensional virtual space in computing equipment, mapping a target pathogen into a target object, and mapping an initial candidate drug molecule into a drug agent object with autonomous motion attribute; s2, setting a connection logic between drug molecules, wherein the connection logic is used for monitoring the relative states between drug intelligent agent objects in real time, establishing virtual rigidity constraint when preset combination conditions are met, and locking a plurality of drug intelligent agent objects into molecule cluster objects which move together; S3, performing kinetic simulation in the three-dimensional virtual space, controlling the drug intelligent object or the molecular cluster object to update the position based on a physical potential energy field, and driving the drug intelligent object or the molecular cluster object to form a space surrounding on the surface of the target object; S4, calculating the surface coverage rate of the molecular cluster object to the key site of the target object, and generating an adaptability score of the current drug molecule based on the surface coverage rate; And S5, updating the topological structure of the medicine molecules by using an evolution algorithm based on the fitness score, generating a new generation medicine intelligent object, and returning to the execution step S2 until a preset convergence condition is met, and outputting the optimal medicine molecular structure. As a further improvement of the present technical solution, in step S2, the preset bonding conditions specifically include: extracting charge distribution data of a contact surface, and calculating electrostatic potential products of contact areas; and when the Euclidean distance between two drug agent objects is smaller than the Van der Waals contact threshold, and the electrostatic potential product is negative and meets the energy threshold requirement, judging that the preset combination condition is met. As a furth