20240029834. Drug Optimization by Active Learning simplified abstract (EXSCIENTIA AI LIMITED)
Contents
Drug Optimization by Active Learning
Organization Name
Inventor(s)
Emil Nicolae Nichita of Oxford (GB)
Drug Optimization by Active Learning - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240029834 titled 'Drug Optimization by Active Learning
Simplified Explanation
The abstract describes a method for computational drug design using active learning. Here are the bullet points explaining the patent/innovation:
- Defining a population of compounds
- Defining a training set of compounds with known biological properties
- Defining multiple objectives, each representing a desired biological property
- Training a Bayesian statistical model using the training set to approximate the biological properties of compounds based on their structural features
- Determining a subset of compounds from the population that are not in the training set
- Optimizing an acquisition function based on the probability distribution from the trained model and the defined objectives to select compounds for synthesis
Potential applications of this technology:
- Accelerating the drug discovery process by efficiently selecting compounds for synthesis
- Designing new drugs with specific desired biological properties
- Optimizing drug candidates based on multiple objectives
Problems solved by this technology:
- Reducing the time and cost associated with traditional trial-and-error drug discovery methods
- Improving the efficiency of selecting compounds for synthesis by incorporating probabilistic models and defined objectives
Benefits of this technology:
- Faster identification of potential drug candidates
- Increased success rate in finding compounds with desired biological properties
- Cost and time savings in the drug discovery process
Original Abstract Submitted
a method for computational drug design by active learning includes defining a population of compounds, defining a training set of compounds from the population for which a plurality of biological properties are known, and defining a plurality of objectives each defining a desired biological property. the method includes training, using the training set, a bayesian statistical model to output a probability distribution approximating biological properties of compounds in the population as an objective function of structural features of the compounds in the population. the method includes determining, from the population, a subset of compounds that are not in the training set. the subset is determined according to an optimization of an acquisition function based on the probability distribution from the trained bayesian statistical model and based on the defined objectives. the method includes selecting at least some of the compounds in the determined subset for synthesis.
