Chemical Structure Generation via Vector Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The process of identifying and designing new chemical structures with desired properties is time-consuming and often relies on trial and error, limited by past experiences and incomplete knowledge of existing data, leading to unfruitful research paths.
Innovation Solution
A computer-implemented method and system that generates new chemical compounds by preparing feature vectors, compressing them into relational vectors, mapping these vectors to a 2-dimensional space, allowing user selection, and decompressing the selected vector to generate a new chemical structure, thereby reducing reliance on human intuition and trial and error.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional trial and error methods are used for chemical structure design, then researchers can explore new chemical compounds, but the process becomes time-consuming and expensive
Solution Approach 1:
The patent replaces the mechanical trial-and-error process with an automated computer-based system that uses machine learning models, feature vectors, and optimization algorithms to generate and evaluate chemical structures, eliminating manual experimentation cycles
Solution Approach 2:
The system enables self-service chemical structure design by allowing researchers to input desired properties and receive optimized molecular structures automatically generated by the computer system, reducing reliance on manual iterative design
2Adaptability or versatility
If researchers rely on past learning and experiences for chemical structure design, then they can use intuition to generate candidates, but the direction becomes biased and limited by incomplete knowledge
Solution Approach 1:
The patent introduces a computer-based optimization system as an intermediary between the researcher's requirements and the chemical structure design, using comprehensive databases and algorithms to bridge the gap between incomplete human knowledge and optimal molecular design
Solution Approach 2:
The system transforms the design process by changing from subjective intuition-based parameter selection to objective data-driven parameter optimization, using feature vectors and machine learning to systematically explore the chemical space beyond human experience limitations
3Measurement precision
If comprehensive data analysis is performed on known chemical structures, then better structure/property relationships can be understood, but the processing complexity increases
Solution Approach 1:
The patent segments the complex chemical structure data into discrete feature vectors representing specific molecular properties and characteristics, making the data manageable and suitable for computational processing and analysis
Solution Approach 2:
The system changes the state of chemical structure data from complex unstructured information to structured numerical feature vectors, enabling efficient computational processing while maintaining the ability to accurately represent structure-property relationships
Data Source
AI summary
A computer implemented method of generating new chemical compounds is provided. The method includes preparing a feature vector for each of a plurality of chemical compounds for which a chemical or physical property is known. The method further includes compressing each of the feature vectors into a relational vector, and mapping each of the relational vectors to a map having at least two dimensions. The method further includes presenting the map on a display device. The method further includes receiving a selection of a position on the map, wherein the position is converted to a new relational vector, and decompressing the new relational vector to a candidate feature vector. The method further includes generating a new chemical structure from the candidate feature vector.