Research
Infectious disease and immunology as computational biological systems.
We study infectious disease and immunology through the combined lenses of biology, computation and engineering — concerned not simply with applying computational methods to biological datasets, but with how the biological system itself is represented.
The programme
From representation to reasoning to design.
Our interest is not simply in applying computational methods to biological datasets. We are concerned with how the biological system itself is represented: which entities and mechanisms are present, how they interact, what evidence supports those relationships, and which aspects of context must survive translation into a computational form.
This creates a research programme extending from biological representation through modelling and reasoning to the longer-term possibility of biological design.
We do not treat prediction as an endpoint in itself. Mechanistic explanation, inspectability and traceable evidence remain central to how we approach every stage of the work — from the first representation of a system to the eventual possibility of designing an intervention.
Research areas
Five connected lines of enquiry.
-
1
Infection and host–pathogen interaction
Infection is an interaction rather than an isolated property of a pathogen. We are interested in the molecular and cellular processes through which pathogens encounter hosts, establish infection, exploit or alter host systems, evade defence and generate disease. Representing these processes computationally requires more than lists of genes or proteins; it requires relationships, mechanisms, temporal context and evidence.
-
2
Immunology, recognition and response
Immune systems provide a particularly rich problem for computational biology because recognition, signalling, regulation, memory and adaptation occur across multiple biological scales. We are interested in representations of innate and adaptive immune processes and in the ways pathogens alter, evade or exploit them.
-
3
Biological representation
Before a system can be modelled or reasoned over, it must be represented. Noviota investigates computational forms capable of expressing biological entities, interactions, mechanisms, context, evidence and provenance without reducing complex biology to a vocabulary too impoverished to support useful reasoning.
-
4
Modelling and computational reasoning
Structured representations provide a substrate on which models can be built and questions can be asked. Our longer-term aim is to explore computational methods capable of tracing relationships, evaluating mechanistic hypotheses, identifying gaps or inconsistencies, and comparing possible interventions while retaining links to the biological evidence on which those operations depend.
-
5
From reasoning to design
The eventual objective is not merely to describe biological systems more elegantly. If biological representations and models become sufficiently expressive and inspectable, they may support the design of interventions with intended behaviours. This connects Noviota’s research to biological engineering while keeping mechanistic understanding and traceability at the centre of the process.
Research principles
How we hold the work to account.
- →
Biological meaning should survive computational representation.
- →
Evidence and provenance are part of the scientific object, not administrative metadata added afterwards.
- →
Mechanistic explanation and inspectability matter alongside predictive performance.
- →
Standards and interoperability are preferable to isolated representations where they can preserve the required meaning.
- →
Human and animal infectious disease should be considered across shared biological mechanisms without erasing important species and contextual differences.
- →
Research software should make scientific assumptions visible wherever practicable.
Open projects
Open and experimental tools, where they help test the programme.
Noviota develops open and experimental tools where doing so helps test the research programme. This is a growing project index rather than a fixed product catalogue.
Sybil
An experimental domain-specific language for computational biology. It explores how biological systems and designs can be expressed in a structured, inspectable form suitable for validation, transformation and computational reasoning. Interoperability with SBOL3 and related standards is an important design consideration — not an attempt to replace established biological standards.
Libraries, validation and representations
Further open work in this area may include SBOL libraries, validation tools, ontologies, templates, parsers and research utilities developed as part of the representation and design programme. Projects appear here as they are released.