Dihedral Angle Parameter Fitting via Quantum Mechanics Screening

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional methods for testing and fitting dihedral angle parameters in molecular force fields are computationally intensive and inefficient for large molecules, requiring extensive high-precision quantum mechanics calculations due to the complexity and flexibility of dihedral angles.

Innovation Solution

A method that generates representative conformations of large molecules, compares force field results with quantum mechanics, and iteratively refines dihedral angle parameters by fragmenting the molecule and applying traditional dihedral scanning only to poorly performing angles, using tools like rdkit and high-precision quantum mechanics methods to reduce computational resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional dihedral scanning method is applied to large molecules, then comprehensive evaluation of dihedral angle parameters is achieved, but computational cost increases significantly

Engineering Contradiction:
Improveevaluation accuracyVSAvoidcomputational cost
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent segments the evaluation process into two distinct phases: (1) rapid screening phase using molecular mechanics on representative conformations to identify problematic dihedral angles, and (2) detailed evaluation phase using quantum mechanics only on identified problematic angles. This segmentation allows comprehensive evaluation while minimizing computational cost by avoiding QM calculations on well-performing parameters.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by performing QM calculations only on a subset of dihedral angles that are identified as problematic through the rapid MM-based screening phase, rather than performing exhaustive QM calculations on all dihedral angles. This partial evaluation approach maintains measurement precision for critical parameters while significantly reducing overall computational cost.

Inventive Principle:
Principle #16Partial or excessive action

2Measurement precision

If comprehensive dihedral scanning is performed on large molecules, then all dihedral angle parameters are evaluated, but calculation time increases

Engineering Contradiction:
Improveparameter evaluation completenessVSAvoidcalculation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary action by conducting rapid molecular mechanics calculations on representative conformations before the main evaluation phase. This preliminary screening identifies which dihedral angles require detailed QM evaluation, allowing the system to prepare an optimized evaluation plan that minimizes calculation time while ensuring comprehensive assessment of critical parameters.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent extracts and isolates only the problematic dihedral angles identified during the screening phase for detailed QM evaluation. By taking out only the necessary subset of parameters requiring comprehensive analysis, the method achieves complete evaluation of critical parameters without the time penalty of evaluating all dihedral angles exhaustively.

Inventive Principle:
Principle #2Taking out (Extraction)

3Manufacturing precision

If quantum mechanics calculations are performed on all dihedral angles, then accurate parameter fitting is achieved, but computational resources are wasted

Engineering Contradiction:
Improveparameter fitting accuracyVSAvoidcomputational resource waste
Core Design Contradiction:
Manufacturing precisionVSLoss of energy

Solution Approach 1:

The patent implements feedback by using the results from rapid MM-based screening to guide subsequent QM calculations. The screening phase provides feedback on which dihedral angles exhibit poor performance, and this feedback directs the allocation of computational resources to only those specific angles requiring accurate QM-based parameter fitting, thereby eliminating waste on already-adequate parameters.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent changes the evaluation parameters dynamically based on performance thresholds. Dihedral angles are categorized into different evaluation tiers based on their initial MM assessment results, with only those exceeding performance thresholds undergoing expensive QM calculations. This parameter-based stratification ensures accurate fitting for problematic angles while conserving computational resources.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS20220310210A1Method of testing and fitting the dihedral angle parameters in force field
Publication Date: 2022.09.29 SHENZHEN JINGTAI TECH CO LTD
  • US20220310210A1 patent drawing
  • US20220310210A1 patent drawing

AI summary

The present invention provides a method for testing and fitting the dihedral angle parameters in force field. The method first generates some representative conformations, and then compares the results of force field and quantum mechanics methods using these structures. If the results meet the predefined standards, the process ends; otherwise the molecule will be cut into small-size molecular fragments with only one flexible dihedral angle in each fragment. The dihedral angles will be scanned. And results of force field and quantum mechanics will be compared for each scanned flexible dihedral angle to find out those that do not meet the standards, and their parameters will be selected for further fitting. After new dihedral angle parameters are obtained, apply them to the original series of conformers of the whole molecule for validation. If results meet the standards, complete the whole process of testing and fitting poorly performing dihedral angle parameters.