the greifswald protein screening platform LARA

LARA - real robotics at the institute for biochemistry

• protein screening engineering
• screening for new substrate scopes / altered substrate specificity
• enhanced enzyme stability
• directed evolution approaches
• findind the right enzyme in 1E5 to 1E9 variants

KIWI project - overview

• highly interdiciplinariy project between Biotechnology, Mathematics, Machine Learning, Robotics and Software Engineering
• developing autonomous biolabs by creating and tailoring existing ML tools for biotech
• closing the feedback loop between Mathematical Model/Machine Learning and real Robotics - toward autonomous experimentation
• connecting different automated labs, located at different sites.
• biotechnological goals: optimisation of protein activity and protein yield for production
• the project will have:
• 5 PhD students - open position at University-Greifswald
• ~10 scientists
• ~20 partners from academia - as consultands
• > 4 companies involved

KIWI - partners

scientific data - structure

LARA suite ( gitlab.com/larasuite )

• planning of experiments
• storing all required data for the planning, like literature, substances, material, devices, experimentalists ...
• generating the processes
• execution of the processes, communication with the lab devices
• collection of the data (very structured, well prepared to learn from it)
• evaluation and visualisation of the data (also DoE and machine learning)
• reporting / publishing / exchange between labs

LARA suite - technologies

• all open source based

• languages: python3, (R) and JavaScript
• very compact code, usage of standard library, minimum dependencies
• gRPC as universal wire format (SiLA2 communication) and web front-end / backend (no REST !!)
• AnIML as long term storage and data exchange format (planned)
• one language paradigm for all !
• python-django web framework and database ORM- very simple access
• flexible relational database schema, fitting many lab workflows

summary: KIWI-biolabs, LARA and AnIML

building the lab of the future - how do we all benefit ?

1. machine learning in BioTech and LifeScience applications
2. closing the loop of autonomous, machine guided experimentation
3. standadised data exchange between remote robotics labs
4. high meta-data content for improved AI/Machine Learning
5. possiblity to re-interpret / re-evaluate the data in the future with evolved knowledge
6. SiLA as commincation standard between the components
7. long term data storage
8. free and open source python AnIML implementation: https://gitlab.com/LARAsuite/animl_python

acknowledgements