Exciting Results from Recent Experiments
Published on Jan 6, 2024
Our team has recently achieved a significant breakthrough in utilizing machine learning (ML) for the evaluation of drug candidates. We are thrilled to announce that our findings have been published at USPTO under publication No. US2024/0021277A1, with a publication date of January 18th, 2024. For those eager to delve into the project and explore further, all the comprehensive details are accessible through our repository.Read more about our findings in the USPTO publication linked below.
Read more.
Our team has achieved a breakthrough in improving the estimation of drug efficacy in preclinical evaluations. Traditionally, measuring important physicochemical properties like Log P and Log D is challenging due to low concentrations for highly hydrophobic or hydrophilic species. We overcame these limitations by using machine learning to correlate these properties with liquid chromatography retention time (RT). Our predictive models, incorporating molecular descriptors and RT, achieved a significant improvement in performance (MAE = 0.366, R2 = 0.89). This breakthrough allows for more efficient estimation of drug distribution in the body during preclinical evaluations. Read more about our findings in the latest publication linked below.
Upcoming Conference Presentation
Announced on December 15, 2023
Join us at the ARVO Annual Meeting where we will present our recent work on metabolomic analysis with machine learning.
Event details
Stay Tuned!
More exciting updates are on the way. Keep an eye on this space for the latest news and research findings from our lab.