Application of machine learning in molecular materials discovery

Lead Research Organisation: Imperial College London
Department Name: Dept of Chemistry

Abstract

This project will investigate the use of machine learning, in particular neural networks, for the discovery of new molecular materials. This will focus on porous molecular materials, investigating their formation reactions and properties for separation applications. It will also investigate other organic molecular materials, such as organic molecules for water splitting and hole transport materials. A significant portion of the early work will require building up training sets of data and the use of an evolutionary algorithm in this area.

Publications

10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/N509486/1 01/10/2016 30/09/2021
1805162 Studentship EP/N509486/1 01/10/2016 30/09/2019 Lukas Turcani
 
Description Developed new software that autmatically designs molecules.
Exploitation Route Other people can design new molecules and extend the software to other types of molecules.
Sectors Chemicals,Healthcare,Pharmaceuticals and Medical Biotechnology

 
Title stk 
Description A Python library for the automated design of molcules. 
Type Of Technology Software 
Year Produced 2017 
Open Source License? Yes  
Impact Used in research for materials discovery. 
URL https://github.com/lukasturcani/stk