AI for Reaction Outcome and Synthetic Route Prediction

AI for Reaction Outcome and Synthetic Route Prediction

When: 9^th-11^thMarch 2020

Where: DeVere Totworth Court Hotel, Gloucestershire

Overview:

This joint meeting between the Dial-a-Molecule, Directed Assembly and AI3SD (Artificial Intelligence and Augmented Intelligence for Automated Investigations for Scientific Discovery) Networks was held just as the COVID-19 outbreak began to be significant in the U.K.. A number of participants and speakers had to withdraw on company / university instructions, but we still had an enthusiastic audience of over 100. Most of the speakers who were not able to attend were still able to present their talks remotely. A full report on the meeting is being prepared by Wendy Warr. Many of the talks are already available here.

Scope of meeting.

The meeting examined the state of the art and future opportunities in the use of Artificial Intelligence to predict the outcome of unknown chemical reactions, and consequently design optimum synthetic routes to desired molecules.
A wide variety of AI approaches were illustrated, including expert systems, statistical methods, mechanism based and Machine Learning.
The meeting also considered:
Data sourcing, sharing, and quality including data standards and ontologies.
Automated experimentation to generate reaction knowledge and rapidly synthesis molecules.
Theoretical calculations to enrich or replace experimental data.

There was the opportunity to try a variety of software / database systems from our exhibitors during the meeting.

Paperless Content including the final programme may be found here.

Plenary Lectures:

Professor Pierre Baldi – University of California, Irvine

“AI for Chemistry”

Professor Lee Cronin – University of Glasgow

“A Non-Deterministic Chemputer for Running Chemical Programs”

Dr Connor Coley – Massachusetts Institute of Technology, USA.

“ASKCOS: data-driven chemical synthesis”

Invited Lectures:

Dr Colin Batchelor – Royal Society of Chemistry

“Chemistry ontologies and artificial intelligence”

Dr Natalie Fey – University of Bristol

“Intelligence from Data: Towards Prediction in Organometallic Catalysis”

Professor Val Gillet – University of Sheffield

“Using Reaction Data to Enhance De Novo Design”

Professor Fernanda Duarte – University of Oxford

“Computational design via metal-driven self-assembly: From molecular building blocks to emerging functional materials”

Professor Jonathon Goodman – University of Cambridge

“Gathering molecules: representations and machine learning with minimal data”

Professor Peter Johnson – University of Leeds

“Computer-Assisted Design of Complex Organic Syntheses – 50 years on”

Dr Sami Kanza – University of Southampton
“The Semantic Laboratory”

Dr Mario Latendresse – SRI Biosciences

“Integrating AI with Robust Automated Chemistry: AI Driven Route Design and Automated Reaction & Route Validation”

Professor Harris Makatsoris – Kings College, London.

“Evolutionary computing strategies and feedback control for directed execution and optimisation of chemical reactions”

Professor Mahesan Niranjan – University of Southampton

“Introduction to ML and Structured matrix methods for learning outliers”

Dr Koen Paeshuyse and Dr William Maton – Janssen, and Dr Allyson McIntyre – AstraZeneca
“A Structured Recipe Based Approach in Process Research and Development”

Professor Per-Ola Norrby – AstraZeneca

“From mechanisms to reaction selectivity”

Dr. Quentin Perron – IKTOS
“The importance of false reactions and rule definition for efficient data-driven retrosynthetic analysis”

Dr. Orr Ravitz, CAS

“Making sense of predicted routes: the use of data as evidence for predictions in SciFinder N”

Dr. Lindsey Rickershauser, Merck KGaA

“Synthia (formerly Chematica) Retrosynthesis Software: Validation at the Bench.”

Dr Dobrila D. Rudnicki – NCATS, ASPIRE

“A Specialized Platform for Innovative Research Exploration (ASPIRE): mapping unexplored biologically active chemical space”

Dr Roger Sayle – NextMove Software

“Automated mining of a database of 9.2M reactions from the patent literature, and its application to synthesis planning”

Dr Marwin Segler – Benevolant.ai

“Retrosynthesis via Machine Learning”

Dr Jarek Tomczak – Pistoia Alliance

“UDM – a community-driven data format for the exchange of comprehensive reaction information”

Dr Stanisław Jastrzębski – Molecule.One

“Applying AI to retrosynthesis in the wilderness”

Dr. Jules Tilley – Rahko

“Accurate excited states calculations on near term quantum computers.”

Oral contributions:

Dr Benjamin Deadman – Imperial College – ROAR

“Data-driven exploration of the catalytic reductive amination reaction”

Dr Ella M Gale – University of Bristol

“Encoding solvents and product outcomes to improve reaction prediction systems”

Dr Christopher A Hone, Research Center Pharmaceutical Engineering (RCPE)

“Machine-Assisted Flow Chemistry for Organic Synthesis”

Dr Kjell Jorner, AstraZeneca, United Kingdom

“Reaction prediction in process chemistry with hybrid mechanistic and machine learning models”

Dr Timur Madzhidov, Kazan Federal University

“Predictive models for assessing reaction conditions”

The event will also include ample discussion sessions and opportunities to develop collaborations will be a key aspect .

Running from 10:30 on Monday 9th March through to 4pm on Wednesday 11th March, this 3-day event is residential and registration includes 2 nights accommodation and all meals at the lovely De Vere Tortworth Court Hotel.

Registration fees:

Academia and SMEs £200; Industry Delegates £300
Registration is now closed, but we are operating a reserve list in case of cancellations. Please get in touch with dialamol@soton.ac.uk.

Note that a number of ECR bursaries are available from Dial-a-Molecule for this event. Please contact us

Richard Whitby, Gill Smith and Samantha Kanza on behalf of the organising committee.

Exhibitors and Sponsors: