Brands
Latest top stories
Start-ups
Technology

WhiteLab Genomics: using AI to tackle gene therapy’s delivery problem

9 October 2026

The WhiteLab Genomics team (photo: WhiteLab Genomics)

 

A gene therapy can contain precisely the instructions a patient needs, yet still struggle to deliver them to the cells where they would make a difference. Overcoming that obstacle is the business WhiteLab Genomics is building, with its latest funding intended to move more of its computational designs through biological testing.

On 6 October 2026, the Paris-based company announced a $26 million Series B led by AVP, with new investors Yaday Health and Blast Club joining existing backers Omnes Capital and Debiopharm Innovation Fund. The financing will support further validation of its technology and commercial expansion across North America, Europe and Asia.

For pharmaceutical customers, the potential attraction is a better starting point for developing genetic medicines: delivery systems and payloads selected through computational design and supported by experiments. For WhiteLab, that creates an opportunity to develop valuable components that could be used in more than one therapeutic programme.

 

WhiteLab Genomics’ founders and key milestones

 

WhiteLab was founded in 2019, bringing together healthcare business experience and genetics research. Early investor announcements identify David Del Bourgo and Julien Cottineau as its founders, while current accounts also recognise Lucia Cinque as a co-founder.

Del Bourgo, the chief executive, trained as a biomedical engineer and previously worked at GE Healthcare. Cottineau’s background includes genetics, molecular biology and immunology, with research at Rockefeller University and Necker Hospital. Cinque, now chief operating officer, has helped develop the organisation’s operating structure as its scientific and commercial activities have expanded.

That combination became important as the company moved beyond its initial computational offering. After joining Y Combinator’s Winter 2022 cohort, WhiteLab announced a $10 million Series A that September, backed by Omnes and Debiopharm. Its early work used simulations to support target discovery and the design of delivery vehicles and genetic payloads for gene and cell therapies.

Partnerships subsequently connected those capabilities to specific development challenges. In October 2023, WhiteLab joined Sanofi, Nantes University’s TaRGeT laboratory and Institut Imagine in WIDGeT, a France 2030-supported consortium developing AI-designed viral vectors for eye and kidney diseases.

Its work also extended into manufacturing through a September 2025 collaboration with Cytiva. WhiteLab’s computational tools would help improve the selection of stable cell lines used to produce viral vectors, with the aim of shortening development.

The latest financing builds on this progression. Alongside collaborative research, WhiteLab now wants a portfolio of experimentally validated biological assets that it can advance independently or with pharmaceutical partners. The commercial logic is that a tested component could offer a customer more than a prediction and provide a stronger basis for a licensing agreement.

 

What a drug developer might buy

 

A hypothetical biotechnology company developing a treatment for a neurological disease illustrates how that relationship could work. It might have a promising therapeutic gene but lack a delivery vehicle that carries enough of it into the intended brain cells without excessive exposure elsewhere.

WhiteLab could help design the vehicle or the sequences controlling the gene’s activity, using computational analysis to select candidates for experimental testing. If the results were promising, the customer might license a component or continue its development through a collaboration.

The resulting treatment would still require further safety and efficacy studies, manufacturing development, clinical trials and regulatory review. WhiteLab’s contribution would therefore be an enabling part of the medicine, whose value would depend on how reliably it helped the complete treatment perform.

 

How ALFRED designs gene therapy delivery systems

 

The platform behind that work is ALFRED, short for AI-Led Framework for Rational Exploration in Drug Design. One application is engineering vectors derived from adeno-associated virus, or AAV, which are widely used to carry genetic material into cells.

A vector’s outer protein shell, known as its capsid, influences its interactions with cells and tissues. Altering the protein sequence can change those properties, but the number of possible designs makes exhaustive testing impractical.

WhiteLab’s published technical approach combines protein language models, which learn patterns in protein sequences, with reinforcement learning to generate and refine capsid candidates. Its broader strategy also uses predicted interactions with cell-surface receptors to guide designs towards particular targets.

The intended benefit is a more focused experimental search, although biological testing remains necessary to establish whether the predictions are useful. WhiteLab is applying related design principles to non-viral carriers, including lipid nanoparticles, and synthetic promoters, which are DNA sequences that regulate gene expression.

Together, these capabilities could help a developer coordinate how genetic material reaches a cell with how it behaves once it arrives. 

 

What WhiteLab’s mouse brain study shows

 

In May 2026, WhiteLab reported work with the Paris Brain Institute in which its AI-designed vectors achieved approximately 50-fold greater DNA enrichment in mouse brains than the reference vector AAV9 following intravenous injection, with no detectable liver signal.

The company-reported result suggests potential for more selective delivery, but the measurement should not be confused with a 50-fold improvement in treatment effectiveness. Detecting genetic material in brain tissue does not, by itself, establish that a therapeutic gene functions in the intended cells or improves disease, while an undetectable liver signal is not proof of zero exposure or toxicity.

The next challenge is establishing how the designs perform beyond mice. Published research on other AAV vectors has shown that targeting can depend on receptors that differ between species, making translation a substantive engineering problem. WhiteLab said larger-animal studies were under way; human safety and therapeutic benefit remain to be demonstrated.

 

What the $26 million Series B will fund

 

The Series B will accelerate validation across viral and non-viral delivery technologies and programmable genetic payloads. This should help WhiteLab assess which designs have the biological properties required for further therapeutic development.

Commercial expansion will accompany that work, with plans to strengthen Boston as a hub for North American partnerships, develop a US West Coast presence and pursue opportunities in Japan and South Korea alongside European activities.

Developing more evidence before offering a component to partners could increase its commercial value.

 

Competitors in AI-designed gene delivery

 

WhiteLab enters a field where competitors have already secured significant pharmaceutical relationships. US-based Dyno Therapeutics is a close comparison in AI-enabled AAV capsid engineering, with partnerships involving Roche and Astellas.

Dyno received $50 million upfront for an expanded Roche collaboration announced in 2024. Its Astellas partnership included $18 million upfront and potential milestone and royalty payments exceeding $1.6 billion. According to Dyno, Roche exercised a capsid licence in early 2025 and Astellas did so in 2026, demonstrating progress beyond the initial research agreements.

London-based Sixfold Bioscience addresses an adjacent delivery problem through programmable RNA tags developed using machine learning and experimental testing. Although its technology differs from AAV engineering, it offers another route for developers seeking to direct genetic medicines towards particular cells.

WhiteLab’s proposed distinction is the breadth of its work across receptor-guided design, delivery, payload control and manufacturing support, which is useful to customers facing several connected development problems.

The next opportunity to examine WhiteLab's progress will come at the European Society of Gene and Cell Therapy congress in Hamburg on 27–30 October, where WhiteLab says it will present additional partner data. Beyond those presentations, larger-animal results, functional gene expression, manufacturing performance and asset agreements will help show whether its designs can become dependable tools for drug developers.