Adaptive Automated Scientific Laboratory

Lead Research Organisation: University of Manchester
Department Name: Computer Science

Abstract

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
 
Description We have integrated systems modelling, machine learning, ontology reasoning, and the Robot Scientist Eve.

This component of the AdaLab project is part of a wider collaborative project. The UM contribution to AdaLab aims to integrate all components of the project within the AdaLab framework, and to work on a real-world problem: the yeast diauxic shift.

The major task focussed on for the first year of the project was the "integration of Eve with the developed knowledge representation". This task is expected to run until Month 34 of the project and involves integrating the OCUK ontology, the knowledge base developed as part of task 2.2 and the human - computer interaction mechanism developed as part of task 2.3 into the software of the robot scientist, Eve. The work in the first year of the project has concentrated on establishing the future feasibility of this integration work through contributions to the relevant knowledge frameworks being developed in the laboratories of other partners (specifically, Workpackage 2). Within the reported period an Equipment ontology has been developed, which describes Eve's equipment and its functionality. This is being integrated with the proposed communication protocol (the deliverable report related to the communication protocol is available on request).

Eve's robotics were originally designed to execute high-throughput drug screening assays. We have modified Eve's robotic work-flow to enable it to execute quantitative diauxic shift experiments. With this new experimental capability we have investigated the conditions (genetic and environmental) under which a diauxic shift either occurs or not. We have replicated results obtained using the Robot Scientist Adam, and are in the process of replicating results from Chalmers University of Technology, Sweden. These experiments have provided data for the knowledge base being developed as part of Workpackage 2.

In addition to this planned work, some of the work envisaged as happening in months 22-36 of the project has been brought forward. This was done because the future evaluation of automatically generated hypotheses regarding diauxic shift required that the conditions under which diauxic shift occurred within experiments on Eve were well understood. The initial optimisation experiments for the protocols have also added data to the knowledge base being developed as part of Workpackage 2.
Exploitation Route We are developing a model for future laboratory based science.

A "Robot Scientist" is a laboratory automation system that exploits techniques from the field of artificial intelligence to execute cycles of scientific experimentation. This project aims to develop a framework for semi-automated and automated knowledge discovery by teams of human and robot scientists.

Our vision is that within 10 years many scientific discoveries will be made by teams of human and robot scientists and that such collaborations will generate scientific knowledge more efficiently than either could alone. In this way the productivity of science will be increased, which in turn will lead to societal benefits.

A sustainability plan has been generated, describing the potential for the use of the AdaLab system (as well as its individual components) beyond the formal end of the project. This can be found in Deliverable Report no. 5.2 (available on request). The main outcome of the project will be the integrated AdaLab system, which will comprise all of the project's results working as a cohesive workflow to perform experiments, evaluate and analyse results and plan future experiments based on these results. In addition to this, there are several individual components of the system which have the potential to be used by the wider scientific community after the project has ended. These include:
• The machine learnt models of yeast behaviour and diauxic shift,
• The machine learnt models for selecting new experiments to conduct,
• The AdaLab ontologies,
• The database of yeast growth data produced within the project,
• Communication mechanisms between robot and human scientists,
• Various smaller bits of software developed during the AdaLab project,
• A compendium of relevant literature and datasets to be fully accessible by the scientific community.

Specifically, several software components are planned to be released by the end of the project, for example, in the form of R packages for Workpackage 4; these are currently available as intermediate versions, under GIT repositories (https://bitbucket.org/alexandros_sarafianos/adalab). These include CoRegFlux, an R Bioconductor package for metabolic simulations using gene expression and regulator influence that will be released with an accompanying publication.
Sectors Agriculture

Food and Drink

Chemicals

Creative Economy

Digital/Communication/Information Technologies (including Software)

Electronics

Energy

Environment

Healthcare

Manufacturing

including Industrial Biotechology

Pharmaceuticals and Medical Biotechnology

 
Description The Robot Scientist was exhibited at the Science Museum in August 2015. This led to engagement of, and interaction with, high numbers of the general public around the topic of the Robot Scientist. Prof. King, the Scientific Leader of the AdaLab project, was interviewed by Ann Dooms from the Belgian public television channel 'Canvas' (www.canvas.be; https://www.canvas.be/wetenschap/de-ontmoetingen-van-ann?gid=20801pd=23403). Subsequently in December 2016, a documentary about the life of Alan Turing was broadcast on Canvas TV, featuring the interview with Prof. King and the Robot Scientist 'Eve'. EPSRC workshop: Dr. Soldatova was invited to a dedicated EPSRC workshop entitled "Automating science discovery with artificial intelligence technologies" on March 28, 2017. This workshop was held, together with key stakeholders, "to articulate and explore the future of artificial intelligence (AI) techniques", and thus was an extremely useful forum to be able to discuss the project as well as the future direction of the field. Larisa Soldatova took part in a 'Society 5.0' event at The Royal Society in London (Friday 12th May 2017): to discuss future visions for AI in Scientific Research. Ross King gave a presentation of his work ("The Automation of Science") at an OECD event in Paris (October 2017). The topic of the event was "AI: Intelligent Machines, Smart Policies", thus providing the opportunity for potential influence on future policies concerning Artificial Intelligence.
First Year Of Impact 2015
Sector Creative Economy
Impact Types Cultural

Policy & public services

 
Description EU
Geographic Reach Europe 
Policy Influence Type Contribution to a national consultation/review
Impact Helped shape future EU science policy.
URL https://research-and-innovation.ec.europa.eu/research-area/industrial-research-and-innovation/key-en...
 
Title Closed-loop AI experimentation 
Description Our work is now causing a revolution in science. 
Type Of Material Improvements to research infrastructure 
Year Produced 2009 
Provided To Others? Yes  
Impact Closed-loop AI experimentation 
 
Description Arctoris 
Organisation Arctoris
Country United Kingdom 
Sector Private 
PI Contribution We have started a collaboration with the SME Arctoris on research into cancer, lab automation, and AI.
Collaborator Contribution We have started a collaboration with the SME Arctoris on research into cancer, lab automation, and AI.
Impact EPSRC Ambition project
Start Year 2020
 
Description Imagen therapeutics 
Organisation Imagen Therapeutics
Country United Kingdom 
Sector Private 
PI Contribution Collaboration in AI and cancer.
Collaborator Contribution Collaboration in AI and cancer.
Impact none so far
Start Year 2021
 
Description RIKEN 
Organisation RIKEN
Country Japan 
Sector Public 
PI Contribution Cooperation Laboratory automation and AI: RIKEN Professor Jun Seita, Tokyo Professor Koichi Takahashi, Kobe
Collaborator Contribution Cooperation Laboratory automation and AI: RIKEN Professor Jun Seita, Tokyo Professor Koichi Takahashi, Kobe
Impact https://www.nobelturingchallenge.org/
Start Year 2022
 
Company Name REGEMUS TECHNOLOGIES, LLC 
Description Lab automation 
Year Established 2022 
Impact Early stage
 
Description Horizons article 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Ross King was interviewed for an article in Cartlidge, E. "Let the Robots do the tedious work", Horizons (Swiss magazine for Scientific Research); Vol 113; pg 10-11
Year(s) Of Engagement Activity 2017
URL http://www.snf.ch/SiteCollectionDocuments/horizonte/Horizonte_gesamt/SNSF_horizons_113_en.pdf
 
Description Interview with Prof. King featured on Belgian television 
Form Of Engagement Activity A broadcast e.g. TV/radio/film/podcast (other than news/press)
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Public/other audiences
Results and Impact Prof. King, the Scientific Leader of the AdaLab project, was interviewed by Ann Dooms from the Belgian public television channel 'Canvas' (www.canvas.be) for a documentary about the life of Alan Turing. The documentary, featuring the interview with Prof. King and the Robot Scientist Eve, was broadcast on Canvas TV in December 2016.
Year(s) Of Engagement Activity 2016
URL https://www.canvas.be/wetenschap/de-ontmoetingen-van-ann?gid=20801&pid=23403
 
Description Science Museum Antenna Live Event 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Public/other audiences
Results and Impact We presented the Robot Scientist and discussed the types of experiments it was capable of undertaking, within the framework of 'how to think like a scientist'. We also had an interactive (computer simulation) demonstration of drug design in which visitors could ascertain what features of a compound rendered it as a 'good' or a 'bad' drug. Both activities provoked significant interest and enthusiasm from members of the public of all ages, from 8 to 80! The robot itself sparked more general discussion about the potential uses of a robot scientist, as well as the technicalities of how it operates, whereas the computer simulated demonstration enabled those who took part to think about the characteristics looked for in drug design, which also generated much discussion. Visitor records to this specific exhibit recorded more than 3500 visitors either 'spectating' or actively 'engaging' with the scientists presenting the robot.
Year(s) Of Engagement Activity 2015
URL http://www.youtube.com/watch?v=wMIcMrzDgNc