Repository logo
 

Research Data supporting "A Dynamic Knowledge Graph Approach to Distributed Self-Driving Laboratories"


No Thumbnail Available

Type

Dataset

Change log

Authors

Mosbach, Sebastian 
Taylor, Connor J 
Karan, Dogancan 
Lee, Kok Foong 

Description

Experimental data and provenance records captured in ontological format during the closed-loop optimisation of the manuscript "A Dynamic Knowledge Graph Approach to Distributed Self-Driving Laboratories".

Source Data Figure 6: Extractable form of the experiments recorded in the knowledge graph during the collaborative optimisation campaign in chronological order. For autosampler sites from which the chemicals were sourced for runs conducted in the Cambridge lab, as well as additional notes on the data points, please see Supplementary Table S7. The data points contained in this file were used to produce Fig. 6.

Source Data Supplementary Table 7: Extractable form of the Supplementary Table S7 provided in the Supplementary Information. For more details, please see the table caption and related text in the Supplementary Information.

Supplementary Data: Complete records of the collected knowledge graph triples during the collaborative optimisation campaign, including both experimental data and their provenance. The digital twins of the laboratory hardware are anonymised. This file is accessible through common plain text editors, such as Notepad, Notepad++, and Sublime Text.

Supplementary Information: Technical details on the knowledge graph implementation and experimental procedures to reproduce this work.

Supplementary Movie 1: An interactive animation of the progress of the Pareto front advancement (Fig. 6(a)) during the collaborative optimisation campaign.

Supplementary Movie 2: An interactive version of the 3D plot for the cost objective (Fig. 6(b)) during the collaborative optimisation campaign.

Supplementary Movie 3: An interactive version of the 3D plot for the yield objective (Fig. 6(c)) during the collaborative optimisation campaign.

Version

Software / Usage instructions

All the codes developed are publicly available on The World Avatar GitHub repository https://github.com/cambridge-cares/TheWorldAvatar or the Zenodo repository at https://doi.org/10.5281/zenodo.10151236. The docker images of agents are available at GitHub’s public registry located at ghcr.io/cambridge-cares/: doe_agent:1.2.0, vapourtec_schedule_agent:1.2.0, vapourtec_agent:1.2.0, hplc_agent:1.2.0, hplc_postpro_agent:1.2.0, rxn_opt_goal_iter_agent:1.2.0, and rxn_opt_goal_agent:1.0.0. The deployment instructions can be found in folder TheWorldAvatar/Deploy/pips.

Keywords

Digital Twin, Distributed Laboratory, Dynamic Knowledge Graph, Goal-Driven Self-Optimisation, Laboratory Automation, Multi-Agent System

Publisher

Sponsorship
Agency for Science, Technology and Research (A*STAR) (via Cambridge Centre for Advanced Research and Education in Singapore (CARES)) (Unknown)
European Regional Development Fund (ERDF) (via Department For Communities & Local Government) (13R17P02073)
EPSRC (via Alan Turing Institute) (T2-16)
Engineering and Physical Sciences Research Council (EP/S024220/1)
This research was supported by the National Research Foundation, Prime Minister’s Office, Singapore, under its Campus for Research Excellence and Technological Enterprise (CREATE) programme, and Pharma Innovation Platform Singapore (PIPS) via grant to CARES Ltd "Data2Knowledge, C12". This project was cofunded by European Regional Development Fund via the project "Innovation Centre in Digital Molecular Technologies", UKRI via project EP/S024220/1 "EPSRC Centre for Doctoral Training in Automated Chemical Synthesis Enabled by Digital Molecular Technologies". Part of this work was also supported by Towards Turing 2.0 under the EPSRC Grant EP/W037211/1. The authors thank Dr. Andrew C. Breeson for his helpful suggestions on graphical design. J. Bai acknowledges financial support provided by CSC Cambridge International Scholarship from Cambridge Trust and China Scholarship Council. C.J. Taylor is a Sustaining Innovation Postdoctoral Research Associate at Astex Pharmaceuticals and thanks Astex Pharmaceuticals for funding, as well as his Astex colleagues Chris Johnson, Rachel Grainger, Mark Wade, Gianni Chessari, and David Rees for their support. S.D. Rihm acknowledges financial support from Fitzwilliam College, Cambridge, and the Cambridge Trust. M. Kraft gratefully acknowledges the support of the Alexander von Humboldt Foundation. For the purpose of open access, the author has applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising.
Relationships
Supplements: