A robot scientist for drug design and chemical genetics
Lead Research Organisation:
Aberystwyth University
Department Name: Computer Science
Abstract
In Aberystwyth we are developing the world's first 'Robot Scientists'. These are robotic system that can fully automate simple forms of scientific reasoning and experiment. Given a problem they can automatically execute cycles of: hypothesis formation, generation of experiments to discriminate between hypotheses, execution of these experiments using laboratory robotics, and analysis of results. We have previously demonstrated the proof-of-concept of this idea, and built the Robot Scientist 'Adam' which can automate growth experiments in bakers/brewers yeast (S. cerevisiae). This organism is used a 'model' for more complicated biological cells, such as our own. We have been given £540,000 from the University of Wales, Aberystwyth, to develop a Robot Scientists for chemical genomics and drug design - Eve. This money will only pay for the hardware and software for Eve. In this application we are requesting the salary and running costs necessary to demonstrate that chemical genomics and drug design can be successfully automated. Drug design is an ideal candidate for increased automation: automation is already far advanced, with large (relatively stupid) robotic systems widely deployed, and drug design experiments are arguably the most important type of experiment done in the UK / both medical and economically. The core of Eve will be a robotic assay platform that will allow the automated testing of thousands of compounds from a library using automated plate readers and microscopy. Eve will be linked to Adam to enable different types of mutant yeast cells to be tested. The AI brain of Eve will analyse the results of experiments, then design new experiments by choosing new chemical compounds from the library and requesting new yeast strains from Adam. This will result in the automatic accumulation of knowledge. We will look at two specific systems in yeast: pheromone/glucose detection, and the MAPK pathway. These are both intrinsic biological interest, and have significant links to human medicine. We will also make human/yeast chimeric systems that will allow us to test drugs that will work on humans.
Technical Summary
We propose to build and deploy a Robot Scientist for chemical genetics and drug design - Eve. A Robot Scientist is a combination of AI software and laboratory automation hardware that can fully automatically execute cycles of: hypothesis formation, selection of efficient experiments to discriminate between hypotheses, execution of experiments using laboratory automation equipment, and analysis of results. Eve will be the first fully automated system for chemical genetics and drug design experiments. Robot Scientists have the potential to transform the practice of chemical genetics and drug design. Their ability to automate the full cycle of the scientific process enables them to significantly increase the speed and effectiveness of the scientific discovery process and thereby reduce its cost. The hardware and basic software for Eve will be mainly funded through a SRIF3 grant for £540,000. The core of Eve will be a processing station where a range of high-throughput assays will be able to be prepared. The station will have: storage incubator(s)/freezer(s) to hold strains and compound libraries, a compound library, plate transport robotics, a liquid handling station to enable high-throughput assay preparation, automated plate readers, an automated microscopic system, integration hardware, computers, and associated software. We have previously built the Robot Scientist 'Adam' which can automate growth experiments in yeast. Adam and Eve will be coupled so that they can collaborate scientifically. As a proof-of-principle of the automation of chemical genetics we will apply Eve to analysing in yeast (S. cerevisiae) the two GPCR systems (pheromone and sugar sensing), and the MAPK pathway. These systems are both of high intrinsic biological interest, and becoming models for Systems Biology.
Publications
Aubrey W
(2015)
A Tool for Multiple Targeted Genome Deletions that Is Precise, Scar-Free, and Suitable for Automation.
in PloS one
Bilsland E
(2013)
Yeast-based automated high-throughput screens to identify anti-parasitic lead compounds.
in Open biology
Bilsland E
(2018)
Antiplasmodial and trypanocidal activity of violacein and deoxyviolacein produced from synthetic operons.
in BMC biotechnology
Bilsland E
(2011)
Functional Expression of Parasite Drug Targets and Their Human Orthologs in Yeast
in PLoS Neglected Tropical Diseases
Bilsland E
(2018)
Plasmodium dihydrofolate reductase is a second enzyme target for the antimalarial action of triclosan.
in Scientific reports
Coutant A
(2019)
Closed-loop cycles of experiment design, execution, and learning accelerate systems biology model development in yeast
in Proceedings of the National Academy of Sciences
King RD
(2009)
An interview with Ross D. King, Ph.D. (by Vicki Glaser).
in Assay and drug development technologies
King RD
(2009)
Make way for robot scientists.
in Science (New York, N.Y.)
King RD
(2014)
An investigation into eukaryotic pseudouridine synthases.
in Journal of bioinformatics and computational biology
King RD
(2011)
On the formalization and reuse of scientific research.
in Journal of the Royal Society, Interface
King RD
(2009)
The automation of science.
in Science (New York, N.Y.)
Lanthaler K
(2011)
Genome-wide assessment of the carriers involved in the cellular uptake of drugs: a model system in yeast.
in BMC biology
Lu C
(2009)
An investigation into the population abundance distribution of mRNAs, proteins, and metabolites in biological systems.
in Bioinformatics (Oxford, England)
Olier I
(2021)
Transformational machine learning: Learning how to learn from many related scientific problems.
in Proceedings of the National Academy of Sciences of the United States of America
Oliver S
(2022)
From Petri Plates to Petri Nets, a revolution in yeast biology.
Sparkes A
(2012)
AutoLabDB: a substantial open source database schema to support a high-throughput automated laboratory.
in Bioinformatics (Oxford, England)
Wang K
(2021)
NERO: a biomedical named-entity (recognition) ontology with a large, annotated corpus reveals meaningful associations through text embedding.
in NPJ systems biology and applications
Williams K
(2015)
Cheaper faster drug development validated by the repositioning of drugs against neglected tropical diseases.
in Journal of the Royal Society, Interface
Zhou F
(2014)
The use of weighted graphs for large-scale genome analysis.
in PloS one
Description | The project resulted in the first scientific knowledge autonomously discovered by a machine: both hypothesised and experimentally confirmed. |
Exploitation Route | We have made drug design more efficient. We have discovered how the compound triclosan works as an anti-malarial compound. |
Sectors | Aerospace Defence and Marine Agriculture Food and Drink Chemicals Communities and Social Services/Policy Construction Creative Economy Digital/Communication/Information Technologies (including Software) Education Electronics Energy Environment Financial Services and Management Consultancy Healthcare Leisure Activities including Sports Recreation and Tourism Government Democracy and Justice Manufacturing including Industrial Biotechology Culture Heritage Museums and Collections Pharmace |
Description | Research on the Robot Scientists Adam and Eve have inspired industrial research around the world, e.g. in Google. The research also resulted in much media interest - see BBSRC press office. We have discovered how the compound triclosan works as an anti-malarial compound. Malaria kills ~1M people a year. Our paper on triclosan also received a large amount of press interest - ranked top 0.1%. |
First Year Of Impact | 2018 |
Sector | Aerospace, Defence and Marine,Agriculture, Food and Drink,Chemicals,Creative Economy,Financial Services, and Management Consultancy,Healthcare,Leisure Activities, including Sports, Recreation and Tourism,Manufacturing, including Industrial Biotechology,Culture, Heritage, Museums and Collections,Pharmaceuticals and Medical Biotechnology |
Impact Types | Societal Economic |
Description | Broadcast BBC Cambs Breakfast Show |
Form Of Engagement Activity | A broadcast e.g. TV/radio/film/podcast (other than news/press) |
Part Of Official Scheme? | No |
Geographic Reach | Regional |
Primary Audience | Public/other audiences |
Results and Impact | Interview on Breakfast show of BBC Radio Cambridgeshire. This related to our discovery, in work funded by the BBSRC and the Bill & Melinda Gates Foundation, that triclosan (an antimicrobial commonly used in toothpaste) had a second enzyme target in the malaria parasite. This opens up the prospect of the development of a dual-action antimalarial drug.The pair describing this work: Bilsland E, van Vliet L, Williams K, Feltham J, Carrasco MP, Fotoran WL, Cubillos EFG, Wunderlich G, Grøtli M, Hollfelder F, Jackson V, King RD, Oliver SG (2018) Plasmodium dihydrofolate reductase is a second enzyme target for the antimalarial action of triclosan. Scientific Reports 8: 2018 (8 pages) has (thus far) an Altmetrics score of 1019 and has engendered 131 News articles. |
Year(s) Of Engagement Activity | 2018 |
Description | Broadcast Naked Scientists |
Form Of Engagement Activity | A broadcast e.g. TV/radio/film/podcast (other than news/press) |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Public/other audiences |
Results and Impact | Interview on Breakfast show of BBC Radio Cambridgeshire. This related to our discovery, in work funded by the BBSRC and the Bill & Melinda Gates Foundation, that triclosan (an antimicrobial commonly used in toothpaste) had a second enzyme target in the malaria parasite. This opens up the prospect of the development of a dual-action antimalarial drug.The pair describing this work: Bilsland E, van Vliet L, Williams K, Feltham J, Carrasco MP, Fotoran WL, Cubillos EFG, Wunderlich G, Grøtli M, Hollfelder F, Jackson V, King RD, Oliver SG (2018) Plasmodium dihydrofolate reductase is a second enzyme target for the antimalarial action of triclosan. Scientific Reports 8: 2018 (8 pages) has (thus far) an Altmetrics score of 1019 and has engendered 131 News articles. |
Year(s) Of Engagement Activity | 2018 |