Scientific Roadmap

Toward a Science of Biological Observability

Lunan Foldomics is developing a long-term scientific program centered on a fundamental question:

How does biological information become biological form?

Modern biology has made extraordinary progress in describing genes, proteins, signaling pathways, regulatory networks, and cellular states. Yet the relationship between these microscopic processes and the emergence of tissues, organs, spatial organization, and morphology remains only partially understood.

Between molecular regulation and macroscopic anatomy lies an intermediate level of biological organization where cells interact, compete, cooperate, differentiate, move, persist, and collectively generate observable structures.

We refer to this intermediate domain as the mesoscopic level.

At this scale, morphology is not treated simply as the final visible outcome of molecular processes. Instead, biological form may be studied as a structured observable: partial, compressed, history-dependent, and influenced by environmental context, but still capable of retaining recoverable information about the regulatory processes that generated it.

The long-term objective of Lunan Foldomics is to contribute to the development of a science of biological observability: a framework for understanding how morphology, spatial organization, and collective form can be used to study the hidden regulatory and generative processes of living systems.

This program sits at the intersection of systems biology, spatial biology, artificial intelligence, complex systems, and the epistemology of biological observation.

It is currently organized as a four-stage scientific roadmap that we refer to as:

The Mesoscope Tetralogy

The Mesoscope Tetralogy is a long-term research program designed to move progressively from controlled synthetic systems toward richer mesoscopic models, formal morphological languages, and ultimately real biological data. 

Its aim is to investigate whether biological form can be studied as a structured observable of hidden regulatory organization.

Stage I — Evoscope

Minimal Biological Observability

The first stage asks whether a minimal regulatory system can generate morphologies that retain recoverable information about the regulatory processes that produced them. 

To investigate this question, we developed Evoscope, a controlled generative system in which the internal regulatory state is known and the resulting collective morphology can be analyzed.

Evoscope is not intended to reproduce the full complexity of real tissues. Instead, it provides a simplified experimental environment where the relationship between regulation, spatial organization, and morphology can be studied under controlled conditions.

This stage provides the initial conceptual foundation for mesoscopic biological observability.

Stage II — Toward a Mesoscopic Configuration Atlas

The second stage investigates how richer physical interactions and environmental constraints expand the space of possible biological configurations.

The objective is to progressively construct an Atlas of Mesoscopic Configurations: a systematic characterization of recurrent spatial organizations emerging from controlled regulatory systems.

Rather than focusing on isolated simulations, this stage aims to identify families of mesoscopic organization, their latent representations, their transitions, and their relationships to observable morphology.

This atlas is intended to serve as a bridge between synthetic morphogenesis and a more formal description of biological form.

Stage III — Meso-L

A Language for Mesoscopic Biology

Scientific disciplines mature when they develop a language capable of describing the phenomena they study.

The third stage introduces Meso-L, a proposed language for mesoscopic biological organization.

The objective is to define a vocabulary of recurrent morphological configurations, latent representations, and compositional rules that allows biological organization to be described, queried, compared, and eventually generated through structured descriptions.

Meso-L aims to transform mesoscopic configurations from isolated visual patterns into interpretable biological tokens.

In this framework, morphology becomes not only something to observe, but something that can be represented, combined, and reasoned about.

Stage IV — Toward Real Biological Observability

The final stage investigates whether the concepts developed in synthetic systems can be applied to real biological data.

By combining spatial transcriptomics, biological imaging, and generative computational models, this work aims to explore whether mesoscopic representations can support the generation of new hypotheses about developmental organization and regulatory programs.

Embryonic development represents a particularly powerful system for this effort, because it offers a natural setting in which regulatory information, spatial patterning, cellular differentiation, and morphology are deeply intertwined.

The goal is not to replace experimental biology, but to provide computational frameworks that help interpret the relationship between gene regulation, spatial organization, and biological form.

Looking Forward

The Mesoscope Tetralogy represents the current scientific direction of Lunan Foldomics.

Its broader objective is to investigate whether biological form can be studied as an observable of hidden regulatory organization.

This vision remains exploratory. It does not assume that morphology contains all the information needed to reconstruct biological state. Rather, it asks a more precise question:

What part of biological regulation remains visible in form?

Lunan Foldomics was created to investigate this question through theory, simulation, representation learning, and the progressive integration of synthetic and experimental biological systems.