The Bayes Centre welcomes new resident: Amytis

Amytis began as an attempt to solve a practical problem their founding team faced during their PhDs: how do you keep context, relationships and data intact as a research project evolves?

As AI has rapidly entered life science research, that question has become even more important: how do we keep processing steps not only intact, but visible, when AI is used in a workflow? 

Most AI tools for science work through one-off chat windows, requiring users to re-explain projects, move between workspaces and files or commit to a single model for all of their work. Critically, the processing steps and changes made by an AI model may not be immediately visible or understood by the user, which compromises the overall reliability of a project. 

They believe AI should augment rigorous research, not replace it. 

Introducing Amytis

Amytis is a visual AI workflow platform for life science research. It lets researchers create and run workflows based on real scientific questions, while keeping all inputs, prompts, model choices, parameters and steps visible and editable. 

Rather than a linear chat window, Amytis uses a knowledge graph or mind-map format to bring together all of the ‘building blocks’ of science. Whether that be data, protocols, materials or academic papers. You can connect all of these within Amytis and use your choice of AI models to summarise, analyse or generate. 

Their principles

At Amytis they believe that the application of AI in life science research is one of the primary ways it can be used for good. It's already being used to expedite disease diagnosis and subsequently discover new treatment pathways. What if we put this tool into the hands of every life scientist? What could be achieved?

Amytis is designed to reduce the technical barriers to using AI in scientific research, but by fundamentally changing the way scientists interact with it, they also ensure that control, trust and transparency aren’t sacrificed in favour of ease of use. 

Collaborate with Amytis

They are a team of researchers, building for researchers. Amytis is always open to feedback and collaboration from students, lab groups and biotech companies who are tackling complex problems. They want Amytis to be the most useful and trusted tool, so if you have an idea or want some help using AI in your project - get in touch!

They are also always looking for new and innovative ways to instill control, trust and transparency in their product. If you have expertise in the field of local AI, human computer interaction or similar then get in touch to collaborate.