Can we teach generative artificial intelligence the design language of engineered living materials?

TL;DR

This paper develops a comprehensive ontology and codification scheme for engineered living materials (ELMs), demonstrating its utility and exploring its integration with generative AI to facilitate ELM design and discovery.

Contribution

It introduces a versatile ontology and codification scheme for ELMs and validates its usability with generative AI to support innovative ELM design.

Findings

The ontology effectively classifies diverse ELMs across applications and synthesis methods.

Generative AI can understand and apply the ontology for describing and designing ELMs.

The combined approach accelerates communication and innovation in the ELM field.

Abstract

This study presents a versatile ontology and a useful codification scheme for describing all kinds of engineered living materials (ELMs). The different components of the ontology, namely: families according to the taxonomy for ELMs, industrial applications and synthesis or processing methods, are systematically organized, enumerated, classified, codified and explained. The methodic application of the ontology to a set of 100 relevant examples of ELMs helps to demonstrate its utility and adaptability to many different types of ELMs with a wide range of industrial applications and obtained through numerous synthesis and processing methods. This proves that the developed ontology and codification schemes, with the glossary provided to support its implementation and application, can serve as a comprehensive classification tool for the emergent field of ELMs. Furthermore, the usability of the ELMs ontology and codification by a generative artificial intelligence (AI) is explored and validated by different means, checking that both natural language and the codification are understandable for describing ELMs, verifying that the generative AI adequately codifies examples of ELMs according to the ontology, and validating the synergic applicability of the ontology and codification with generative AI tools for illustrating novel ELMs and supporting their conceptual design. This study is expected to provide a universal language to facilitate communication in the ELMs field and to foster the discovery of new ELMs and related innovations, hoping it may accelerate scientific and technological discoveries.