ACS Spring 2024 – Many Flavors of Chemistry
New Orleans, LA, USA.
Join Optibrium at the ACS Spring meeting 2024 where we are delighted to be presenting the below talks.
Matt Segall will be presenting ‘Predicting in vivo PK from limited in vitro ADME data using deep learning imputation’. This presentation will:
- Outline a unique deep learning imputation method that can learn from sparse data
- Provide a case study demonstrating that this approach can predict PK parameters with state-of-the-art accuracy using a high-quality data set of rat PK and sparse data for nine in vitro ADME properties
- Demonstrate the application of this method to an ongoing anti-infective drug discovery project
Location: Room 349 (Ernest N. Morial Convention Center)
Time: 08:05am – 08:30am Monday 18 March 2024
Matthew will also be presenting ‘From quantum mechanics to metabolic pathways’. This work has been carried out in with Mario Oeren, Peter Hunt, Hamed Tabatabaei Ghomi and Tamsin Mansley from Optibrium and will showcase:
- Methods to predict isoform-specific metabolism for a broad range of enzymes involved in Phase I and Phase II metabolism
- How these site-of-metabolism models can be combined with models that predict which enzymes and isoforms are likely to metabolise a compound
- The validity of these pathway predictions by comparison with experimentally observed metabolite profiles.
Location: Room 349 (Ernest N. Morial Convention Center)
Time: 09:35am – 10:00am Monday March 18, 2024
We’ll also be presenting a poster on ‘Application of Deep Learning to Agrochemical Development’. This work was carried out in collaboration with FMC, an agricultural sciences company where we will show how:
- Imputation improved performance for all sparse agrochemical bioactivity datasets
- Decision-making confidence was enhanced and number of required experiments were reduced
- Uncertainty estimates allowed accurate predictions to be made
Poster number: 3994877
Location: Hall C (Ernest N. Morial Convention Center)
Time: 8 pm – 10 pm – Tuesday March 19 2024
Join Michael Parker for his presentation on ‘Rapid AI generation of optimised compound designs guided by user interactions’. The presentation will provide an overview of a model that can identify user goals within a multi-dimensional parameter space within a few interactions and successfully generate relevant, optimised compound designs meeting multi-parameter goals. This presentation forms part of a project that included Charlotte Wharrick; Mario Oeren; Peter Hunt; Tamsin Mansley and Matt Segall.
Location: Hall B, Room 3 (Ernest N. Morial Convention Center)
Time: 11:20am – 11:50am Monday 18 March 2024
Join Ann Cleves talk ‘Learning causal models from limited data for lead optimization’, part of the Drug Design: Methods in Drug Discovery symposium. Ann’s talk will focus on:
- The importance of computational prediction of ligand binding affinities
- How QuanSA uses machine learning to induce a physically meaningful model of ligand binding pockets
Location: R02 (Ernest N. Morial Convention Center)
Time: 10:25am – 10:50am Tuesday March 19 2024
Join Ajay Jain in the Machine Learning in Chemistry: Drug Discovery and Ligand Analysis symposium organised by Alex Dickson and David Koes on March 20. Ajay’s presentation is titled ‘Nucleating 3D molecular representations for model induction: Multiple Ligand Alignment’ and will discuss an approach to multiple ligand alignment based on the eSim molecular similarity method.