Microbial Strain Combination Modeling for Reproducible Growth Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for selecting and combining microbial strains for industrial applications are time-consuming, labor-intensive, and lack reproducibility due to sensitivity to experimental conditions, making it difficult to find the optimal strain combination.
Innovation Solution
A reinforcement learning method using genome analysis, metabolic information, and growth index information to determine optimal microbial strain combinations through models like regression and latent factor collaborative filtering algorithms, considering interactions and competition among strains.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If experimental methods are used to evaluate microbial strain combinations through large-scale culture experiments, then growth rates and substance production can be assessed, but the process requires a lot of time and labor and has limited ability to find optimal combinations
Solution Approach 1:
The patent creates virtual copies of microbial strains through genome sequence data and metabolic models, allowing computational simulation of strain combinations without physical culture experiments. This enables parallel evaluation of numerous combinations simultaneously, dramatically reducing time while maintaining assessment accuracy through predictive algorithms that simulate growth rates and metabolite production.
Solution Approach 2:
The patent replaces the mechanical biological experiment system with a computational information processing system. Instead of performing physical culture experiments to evaluate strain interactions, the system uses genome analysis data, metabolic pathway information, and machine learning algorithms to predict combination performance, substituting biological measurement with computational modeling.
2Measurement precision
If experimental methods are used to evaluate microbial strain combinations, then growth rates and substance production can be measured, but the process is labor-intensive and has limited scalability
Solution Approach 1:
The patent uses genome sequence data as digital copies of microbial strains, replacing the need for physical strain maintenance and experimental setup. These computational representations can be stored, manipulated, and combined virtually without the complexity of managing actual biological materials, enabling scalable evaluation of numerous strain combinations.
Solution Approach 2:
The patent introduces intermediate computational layers including metabolic models and machine learning algorithms that mediate between raw genome data and final combination assessment. These intermediaries process and interpret complex biological information, simplifying the overall system architecture while maintaining assessment accuracy.
3Measurement precision
If experimental methods are used to select microbial strains, then growth rates and interactions can be evaluated, but subtle differences in experimental conditions have large impacts on results, reducing reproducibility
Solution Approach 1:
The patent uses standardized genome sequence data as immutable digital copies that are identical across all evaluations. This eliminates the variability introduced by experimental conditions, as the same genomic information can be input into the computational model repeatedly without variation, ensuring consistent and reproducible results.
Solution Approach 2:
The patent transforms variable experimental parameters (temperature, pH, incubation time) into fixed computational inputs based on genome data and metabolic models. The evaluation system uses standardized algorithmic parameters that consistently apply the same logical rules across different assessments, eliminating condition-dependent variability and improving reproducibility.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of the present application relate to methods of optimizing a combination of microbial strains and predicting the growth status of the corresponding strains. According to an embodiment of the present application for achieving the aforementioned task, a method of determining a combination of microbial strains may include: obtaining genome analysis information related to a target microorganism; obtaining first metabolic information related to each of a plurality of first microbial colonies including the target microorganism; estimating first growth index information related to each of the plurality of first microbial colonies by using the genome analysis information and the metabolic information as inputs of a first model; and determining a strain combination, based on at least one of the metabolic information and the growth index information.