Why Biological Design Needs Open Standards: The Case for SBOL in Nanomedicine

The life sciences are building at a scale and complexity that informal documentation can no longer support. A shared design language is not optional infrastructure — it is the foundation on which reproducible, translatable, and safe nanomedicine depends.

Ashley Morgan · May 2026 · 18 min read

Keywords

1. Introduction: The Infrastructure Gap Nobody Is Talking About

Over the past several years, working at the intersection of biological data infrastructure, veterinary medicine, and nanomedicine design, I have become increasingly convinced that the biggest bottleneck in nanomedicine is not discovery — it is documentation.

That is not a comfortable observation. The field has invested enormously in the science of nanoparticle design. The returns have been real. But there is a structural gap that the science cannot close on its own: biology has built extraordinary discovery infrastructure. It has not built design infrastructure. The way biological systems are formally specified, transferred, and assessed has not kept pace with the complexity of the systems being designed. The result is a growing gap between what the science can do and what the system can reliably reproduce.

We are trying to manufacture twenty-first century biological systems using documentation formats designed for the twentieth. The consequences are visible, documented, and entirely preventable.

There is a peculiar paradox at the heart of modern nanomedicine. Lipid nanoparticles are delivering gene therapies, mRNA vaccines have been administered to billions of people, and nanoparticle-mediated drug delivery is redefining what is possible in oncology and infectious disease.

Yet the way these systems are designed, described, and shared between researchers, manufacturers, and regulators remains largely stuck in the documentation practices of a previous era. Nanoparticle formulations, lipid compositions, encapsulation parameters, surface functionalisations — these are still routinely communicated through prose descriptions in journal articles and PDF-based technical documents. The information is present. But it is not structured, not formally specified, and not designed to be faithfully reproduced by a different team, transferred to a manufacturing partner in another country, or assessed systematically by a regulatory body.

Consider what happened in 2021 when Emergent BioSolutions contaminated up to 15 million doses of COVID-19 vaccine at its Baltimore facility. The FDA inspection report cited inadequate documentation of process controls as a contributing factor. This was not an isolated incident. It was a visible example of a systemic failure in how biological manufacturing processes are specified and transferred.

This article argues that the solution already exists — and that the right frame for it is not simply ‘better documentation’ but something more precise: SBOL as the biological design layer in an emerging machine-readable CMC stack. The Synthetic Biology Open Language is a mature, community-developed, formally specified open standard for describing biological designs as structured data. Its adoption in nanomedicine has been limited by tooling gaps and the absence of a regulatory mandate. Both conditions are changing. The question is whether the field will move proactively or wait to be compelled.

2. What SBOL Is, and What It Looks Like in Practice

Before making the case for SBOL in nanomedicine, it is worth being precise about what SBOL actually is — and critically, what it looks like when applied to real nanoparticle systems.

SBOL is an open, community-maintained standard for representing biological designs as structured, machine-readable data. It was originally developed for synthetic biology — circuits, gene constructs, regulatory elements — but its architecture is extensible by design. SBOL3, the current version, provides a formal data model for describing biological parts, devices, systems, and their functional relationships, along with the experimental context in which they were characterised.

What SBOL is not is a proprietary software tool, a laboratory information management system, or an academic curiosity. It is a specification — like HTML is a specification, or like the DICOM standard is a specification for medical imaging. Any software, any database, any workflow system that implements the specification can read and write SBOL documents. This interoperability is the entire point.

In practice, an SBOL representation of a lipid nanoparticle system would encode the core formulation parameters as structured, queryable data: lipid molar composition ratios (ionisable lipid, phospholipid, cholesterol, PEG-lipid); the N/P ratio of lipid to nucleic acid cargo; mixing flow rates and temperature profiles during formulation.

It would also encode the measured properties that characterise the resulting system: particle size, polydispersity index, encapsulation efficiency, and surface charge (zeta potential). Critically, each parameter is formally associated with its measurement method, the instrument used, and the experimental conditions — not buried in a footnote, but linked to the design object it characterises as a machine-readable assertion.

This is the critical difference from current practice. In a conventional CMC document, a reviewer comparing two LNP formulations must locate relevant parameters across dozens of pages of prose, extract them manually, and compare. There is no guarantee the same parameter uses the same term in both documents, was measured using the same method, or is reported at the same level of precision. In an SBOL-formatted dataset, the comparison is computational. Discrepancies are flagged automatically. Missing parameters are identified systematically. The human reviewer focuses on interpretation, not extraction.

In an SBOL-formatted dataset, comparison is computational. The human reviewer focuses on interpretation, not extraction.

The SBOL community includes researchers from MIT, Edinburgh, Utah, Newcastle, and dozens of other institutions, as well as companies including Benchling and Synbiota. The standard is governed by an open community process with formal versioning. Benchling, the most widely used electronic lab notebook in the life sciences, has built SBOL import and export into its standard platform. This is not a niche academic tool — it is infrastructure already present in the workflows many researchers use daily.

2.1 Why SBOL Rather Than Existing Alternatives

It is worth being direct about the alternatives, because specialists will reasonably ask why SBOL rather than other data standards that already exist in adjacent domains.

ISA-Tab (Investigation/Study/Assay) is a widely used framework for describing experimental metadata in life sciences research. It is well suited to capturing the context of an experiment — who ran it, when, under what conditions — but it was not designed to represent biological design objects with formal, queryable structure. It describes the experiment around the design rather than the design itself.

AnIML (Analytical Information Markup Language) is a strong standard for encoding analytical instrument data — spectroscopy, chromatography, particle sizing. It is the right format for characterisation output. It is not the right format for encoding the design intent that generated that output — the formulation choices, the compositional rationale, the functional relationships between components.

Proprietary LIMS and CMC documentation platforms — Veeva Vault, MasterControl, and their equivalents — provide workflow management and document control for regulated industries. They manage documents well. They do not provide a formal semantic layer that allows the biological content of those documents to be computationally processed or compared across systems.

SBOL’s distinctive strength is that it was designed to represent biological design intent formally and extensibly. It encodes not just parameter values but the relationships between components, the functional roles of parts, and the provenance of design decisions. For nanomedicine — where a therapeutic system is a designed object whose properties emerge from the interaction of components — this design-first representation is exactly what is needed. The other standards complement SBOL; they do not substitute for it.

3. The Reproducibility Crisis in Nanomedicine

3.1 The Scale of the Problem

Reproducibility problems in nanomedicine are well documented and structurally serious. A substantial proportion of nanoparticle research cannot be reproduced from published methods alone — and where the cause has been examined, documentation failures (omitted parameters, ambiguous descriptions, insufficient methodological detail) feature more prominently than fraud or experimental error.

The causes are not primarily failures of scientific integrity. They are failures of documentation design. Methods sections in journal articles were developed to communicate to expert readers at a level of detail sufficient for approximate replication. They were not designed to specify a biological system precisely enough for automated processing, regulatory assessment, or manufacturing transfer to a team with no prior knowledge of the work.

This is not a problem confined to academic research. It propagates directly into clinical translation. When a nanoparticle system that has shown efficacy in a research laboratory is transferred to a manufacturing partner for scale-up, the same documentation failures that prevented academic reproducibility now prevent manufacturing consistency. The design that worked in the original lab cannot be reliably reconstructed because the description of that design was never precise enough to support reconstruction.

3.2 Why Current Documentation Fails

PDF-based technical documents — the standard format for CMC submissions — share the same fundamental limitation. A regulator comparing two lipid nanoparticle formulations must locate the relevant parameters across dozens of pages of prose, extract them manually, and make the comparison by hand. There is no formal schema requiring specific parameters to be present, and no standardised vocabulary ensuring the same term means the same thing in both documents.

The problem compounds as biological systems become more complex. A simple small-molecule drug can be fully specified by its molecular structure and a handful of parameters. A lipid nanoparticle carrying an mRNA payload is a multi-component system with process-dependent properties and formulation parameters that affect both efficacy and safety. Prose can describe it. Prose cannot describe it with the precision, completeness, and processability that manufacturing transfer and regulatory assessment require.

The problem is not that researchers are careless. It is that the documentation format being used was designed for a simpler era and a different purpose.

4. The mRNA Vaccine Case Study

The COVID-19 pandemic provided an involuntary stress test of the life sciences’ manufacturing infrastructure. The results were instructive, and the documentary record — FDA inspection reports, Congressional testimony, and published analyses — provides a clearer picture than the initial reporting suggested.

The speed at which mRNA vaccines were developed was a genuine scientific triumph. Moderna and BioNTech had vaccine candidates in clinical trials within weeks of the viral sequence being published. What followed was harder. Scaling mRNA vaccine manufacturing from clinical trial quantities to billions of doses required transferring complex manufacturing processes to facilities around the world — facilities working from documentation packages prepared under extraordinary time pressure, for processes that even the original developers did not fully understand at the parameter level.

The FDA’s inspection of Emergent BioSolutions in 2021 documented the consequences: contaminated batches, inadequate process controls, and documentation that was insufficient to support consistent reproduction of the manufacturing process. The agency’s inspection report — a public document — cited deficiencies in process validation and the specification of critical process parameters. Manufacturing failures of this kind are rarely caused by a single factor; training gaps, QA systems, and equipment all play a role. But failures in process specification and documentation are precisely the kind that structured design standards are designed to reduce.

More broadly, the mRNA manufacturing scale-up was widely reported to involve significant batch-to-batch variability at multiple facilities, with process deviations frequently attributed to ambiguities in transferred documentation rather than equipment failures or operator error. The precise scale of batch losses remains commercially sensitive and not fully disclosed.

SBOL-compliant documentation would not have eliminated all manufacturing variability. But it would have provided a formally specified, machine-readable representation of the design that could be validated computationally before manufacturing began, compared systematically across facilities, and updated with explicit version control as process improvements were made. The interpretation step — where most variability originated — would have been substantially reduced.

5. The CMC Regulatory Challenge

5.1 What CMC Requires

Chemistry, Manufacturing and Controls documentation is the formal technical submission that accompanies any application to a regulatory body — the FDA, EMA, MHRA — for authorisation to manufacture and sell a biological therapeutic. For nanomedicine products, CMC submissions are substantially more complex than for conventional pharmaceuticals. The multi-component nature of nanoparticle systems means that the characterisation burden is high, and the process-dependent properties of many nanoparticle formulations mean that manufacturing process specification is not supporting material — it is central evidence.

5.2 A Concrete Example of the Schema Problem

Consider two hypothetical CMC submissions for LNP-based mRNA therapeutics reviewed by the same agency in the same month. Submission A describes its ionisable lipid component as a molar percentage of total lipid content. Submission B describes the same parameter as a mass ratio relative to mRNA payload. Both are valid ways of reporting the same information. Neither is wrong. But they cannot be directly compared without manual conversion — and that conversion requires assumptions about molecular weights that may not be stated explicitly in either document.

This is not an edge case. It is representative of the parameter inconsistency that characterises the current CMC submission landscape for novel biological therapeutics. The FDA’s emerging Standardised Study Data initiative is an institutional acknowledgment of this problem. The agency has invested significantly in building data standards for small molecule submissions. The equivalent infrastructure for complex biological systems — including nanomedicines — is still developing, and SBOL is a credible foundation for it.

5.3 The Direction of Regulatory Travel

The FDA’s structured data work, the EMA’s work on data standardisation, and the MHRA’s post-Brexit regulatory modernisation programme all point in the same direction: machine-readable submissions, standardised data formats, and computational review as a complement to expert assessment. This transition is not speculative. It is in progress. Organisations that build structured data capabilities now are positioning themselves for a regulatory environment that will, in all likelihood, make them a requirement within the decade.

6. Open Standards as Strategic Infrastructure

6.1 Why SBOL Has Not Already Won

If the case for open standards in biological design is so compelling, it is reasonable to ask why SBOL has not achieved broader adoption in nanomedicine already. The answer involves three factors that are worth being honest about.

First, tooling. Until relatively recently, working with SBOL required specialist technical knowledge and tools that were not integrated into standard laboratory workflows. The situation has improved significantly — Benchling’s SBOL support, the SynBioHub repository, and the SBOL Designer visual editor have substantially lowered the barrier to entry. But the tooling ecosystem is still maturing, and the friction of adoption remains higher than it should be.

Second, regulatory mandate. The most powerful driver of documentation standard adoption in the life sciences is regulatory requirement. DICOM became universal in medical imaging because regulatory and accreditation bodies required it. HL7 FHIR is becoming standard in health informatics for the same reason. SBOL lacks an equivalent mandate in nanomedicine, and in the absence of one, the adoption incentive for any individual organisation is weakened by the collective action problem: the value of the standard increases with adoption, but early adopters bear disproportionate transition costs.

Third, community boundaries. SBOL developed in the synthetic biology community, which has different institutional culture, funding sources, and publication norms from the nanomedicine community. The conceptual overlap is high, but the community overlap has been limited. Building the bridges between these communities — which is partly what this article is attempting to do — is a prerequisite for broader adoption.

6.2 The Network Effect and Competitive Timing

Despite these constraints, the network effects argument for early adoption is strong. The value of a shared design language increases with the number of parties using it. A company that describes its nanoparticle designs in SBOL can communicate those designs to any other organisation using SBOL without translation, use any SBOL-compatible tool, and pull design data from any SBOL-compatible repository.

The comparison to HTML is instructive. Before HTML standardised web documents, every system that wanted to display formatted text invented its own format. The result was fragmentation and enormous duplication of effort. HTML did not eliminate diversity — it enabled it, by providing a common foundation. SBOL is attempting to do the same thing for biological design, and the network effects that made HTML’s adoption self-reinforcing will eventually apply to SBOL as well.

Organisations building SBOL competency now — developing internal workflows, building design libraries, training staff — will be ahead of a mandatory transition that competitors will make later under more pressure and at greater cost. The analogy is to electronic health record adoption: early adopters paid higher initial costs but built institutional capability that late adopters had to acquire in a hurry. The transition in biological design documentation will follow the same pattern.

The value of the standard increases with adoption, but early adopters bear disproportionate transition costs. This is the collective action problem that regulatory mandate will eventually resolve.

7. The Path Forward

7.1 A Phased Roadmap

Realising SBOL’s potential in nanomedicine does not require waiting for a complete solution. A phased approach — research adoption, then CDMO integration, then regulatory alignment — is both realistic and strategically sound.

In the research phase, the priority is extending SBOL’s data model to cover nanomedicine-specific parameters with the specificity the field requires. This means developing formal ontologies for nanoparticle characterisation — standardised vocabularies for lipid composition, formulation parameters, and physicochemical properties — and integrating these into the tools researchers already use. Benchling is the obvious starting point. Extending its SBOL support to include nanomedicine-specific parameters would immediately make structured design documentation available to a significant proportion of the sector.

In the CDMO phase, the priority is manufacturing process specification. Contract development and manufacturing organisations are the transfer points where documentation failures have the greatest consequences. Developing SBOL-compatible process specification templates for common nanoparticle manufacturing workflows — LNP formulation by microfluidics, conjugation chemistry, surface functionalisation — would address the most expensive failure mode in the current system.

In the regulatory phase, the priority is engagement with the FDA’s structured data programme, the EMA’s data standardisation work, and the MHRA’s regulatory modernisation initiative. SBOL should be positioned as the biological design layer of emerging structured submission frameworks — the format in which design data is provided when regulatory bodies are ready to receive it computationally.

7.2 The Role of Open Science

A biological design language that is openly specified, freely available, and community governed belongs to everyone. It cannot be captured by a single vendor, made proprietary, or withdrawn when a company changes strategy. The life sciences have benefited enormously from open infrastructure — the Human Genome Project’s open data release, the open publication of pathogen genomes during outbreaks, the creative commons licensing of fundamental research. Open standards for biological design are in the same tradition.

This is sound institutional design, not idealism. Open standards lower barriers to entry, enable competition, and ensure that the infrastructure of the field is not owned by anyone who might extract rent from it. The question is whether the nanomedicine community will build that infrastructure proactively, or wait until the costs of its absence become too large to ignore.

8. Conclusion

The biological design community is not short of intelligence or ambition. It is, in certain critical respects, short of infrastructure. The documentation practices currently used to describe nanoparticle systems were adequate for an era of small-scale academic research. They are not adequate for an era in which nanoparticle therapeutics are manufactured at global scale, transferred between dozens of facilities, and assessed by regulatory bodies under conditions of genuine clinical urgency.

SBOL offers a solution that is technically mature, community-validated, and aligned with the direction of regulatory travel — not as a replacement for existing characterisation and workflow standards, but as the biological design layer that those standards lack. Its adoption has been limited by tooling gaps and the absence of regulatory mandate. Both are changing. The case for early adoption does not require waiting for a perfect standard or a complete mandate. It requires only recognising that the current situation has documented, growing costs.

This paper has argued for a specific framing: SBOL not merely as a documentation improvement, but as the design layer in an emerging machine-readable CMC stack. That framing matters because it positions the standard where it can do the most work — not at the periphery of regulatory submission, but at its core.

The cost of inaction is already being paid — in contaminated batches, failed transfers, and therapies delayed while ambiguity is resolved manually. The technical solution exists. The policy direction is clear. What remains is the institutional decision to act before the next avoidable failure makes the argument for us.

References

1. SBOL Community (2023). Synthetic Biology Open Language (SBOL) Version 3.1.0 Specification. Journal of Integrative Bioinformatics, 20(1).
2. U.S. Food and Drug Administration (2021). Form 483 Inspectional Observations: Emergent BioSolutions, Bayview, Baltimore, April 2021.
3. Madsen, C., McLaughlin, J.A. et al. (2016). The SBOL Stack: A Platform for Storing, Publishing and Sharing Synthetic Biology Designs. ACS Synthetic Biology, 5(6), 487–495.
4. Beal, J. et al. (2019). Communicating Structure and Function in Synthetic Biology Diagrams. ACS Synthetic Biology, 8(8), 1818–1825.
5. European Medicines Agency (2022). Guideline on quality documentation for medicinal products when used with a medical device. EMA/CHMP/QWP/BWP/259165/2019.
6. Walsh, G. & Walsh, E. (2022). Biopharmaceutical benchmarks 2022. Nature Biotechnology, 40, 1722–1760.