Back to search

University of Strathclyde

The application of active learning to the lead optimisation of a novel series of antimalarials

Abstract

dc:description.abstract

Malaria remains a significant global health problem, affecting the lives of hundreds of millions of people each year. Plasmodium falciparum (the parasitic species responsible for almost all malaria-related deaths) has developed resistance to front-line treatment, threatening to reverse decades of progress made towards eradication of the disease. To overcome this, novel medicines that are effective against resistant strains are required urgently.During previous work at GSK Tres Cantos, a series of aminoquinazolines were discovered that showed potential of meeting this medical need. The research reported in this thesis focussed on lead optimisation of the series by applying active learning – an iterative form of machine learning whereby the model identifies data points from which it would learn most, in order to maximise improvement in prediction accuracy. To evaluate the current effectiveness of this technology when applied prospectively to a live drug discovery programme, this project was conducted independently and in parallel to conventional lead optimisation efforts by the Team in Tres Cantos.Both the conventional medicinal chemistry strategy and the active learning strategy identified significantly improved compounds with respect to the initial lead. The best compounds from both efforts were similar both in terms of their structure and properties. However, the active learning method reached this comparable end-point with the synthesis and testing of one-third the number of compounds and with one-third the number of chemists over the same time period. Hence, this thesis demonstrates that active learning can increase the efficiency of lead optimisation and, as such, could help to tackle declining R&D productivity, a problem the pharmaceutical industry has been faced with over the last two decades.

Degree

thesis:*
Name dc:type.qualificationname
phd
Level dc:type.qualificationlevel
doctoral-pg
Grantor dc:publisher.institution
University of Strathclyde
Year dc:date.issued
2022

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Roberts, Jack P.M.
Advisors dc:contributor.advisor
  • Tape, Daniel
  • Hirst, David
  • Murphy, John
Author Identifier dc:creator.authoridentifier
  • 201773522

Rights

Identifiers

dc:identifier.*
Identifier
T17029

Chain of custody

source
Harvested from
University of Strathclyde
Base URL
stax.strath.ac.uk/catalog/oai
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Roberts, Jack P.M.. The application of active learning to the lead optimisation of a novel series of antimalarials. doctoral-pg thesis, University of Strathclyde, 2022. https://stax.strath.ac.uk/concern/theses/rx913q511