28 September 2026

^{Basecamp Research team and biodiversity data partners in Malawi (photo: Basecamp Research)}

 

Off Malta, divers collected tubes of sediment from the seabed beside historic shipwrecks. The wrecks were easy to see. The communities of microbes living in the mud were not.

Basecamp Research and its Maltese collaborators studied the DNA in those samples to find out how the wrecks had changed life on the seabed. They found greater microbial diversity close to the wrecks than farther away, although the type of sediment had an even stronger influence. It was one small investigation in a much larger effort: Basecamp is collecting genetic data from ecosystems around the world to help its AI models design new medicines. 

The London biotech has now raised $140 million in Series C funding to expand its biological dataset and advance its proposed treatments. 

 

A genetic map of biodiversity

 

In January, MTN covered Basecamp’s claim that its AI could design proteins for targeted gene insertion. Behind that result sits a less familiar operation: a network of scientists collecting samples across different habitats, then sequencing and organising the genetic material they contain.

A scoop of sediment holds DNA from many organisms, including microbes that have never been grown in a laboratory. Sequencing turns the mixed material into fragments of genetic code. Researchers then try to assemble those fragments, identify genes and connect them to the organisms and environments from which they came. A gene linked to a habitat, its neighbours and an evolutionary lineage offers more clues about what it might do than a sequence on its own.

This is what Basecamp means by mapping biodiversity. It is building a genetic map of sampled life, heavily informed by microbes, rather than locating and naming every plant or animal on Earth. Its data page now claims more than 100 billion novel genes and more than 10 million species “unknown to science”. Those company figures have not been independently audited in public, and genetic estimates of distinct species should not be read as 10 million formally described organisms. Another section of the same page still gives earlier totals of 10 billion genes and one million species.

The planned expansion is called the Trillion Gene Atlas. Basecamp aims to extend the dataset across more than 100 million species. Sequencing partner PacBio expects roughly 100,000 samples from at least 31 countries to be deeply sequenced for the project. These are targets, rather than a completed inventory of life.

 

From environmental DNA to AI-designed medicines

 

Evolution has spent billions of years testing biological solutions. Microbes compete, survive chemical assaults, enter cells and repair their DNA using molecular machinery that can inspire drug designers. Looking only at familiar laboratory organisms leaves much of that experimental history unseen.

Basecamp trains its EDEN biological AI models on its collection of genetic sequences and related context. The company’s proposition is that a model exposed to a wider variety of evolutionary solutions can propose proteins with useful properties. These include enzymes intended to insert genetic material at chosen sites and peptides that might act against bacteria.

 

Who benefits from the biodiversity map?

 

The samples are collected in places with Basecamp's own scientists, laws and claims on biological resources. The company says it has more than 200 access and benefit-sharing agreements and reports over 130 royalty payments across 21 countries. It says its sequences can be traced to the permits covering their collection. 

In the Malta study, local partners appear as co-authors, government authorisation is stated and a permit number is disclosed. The paper does not tell us how much any partner has earned from Basecamp’s wider commercial work. The company’s aggregate payment count likewise says little about the amounts received, how agreements differ or how a future successful medicine would be valued. As genetic information becomes an industrial input, those details will shape whether collaborators share in the rewards.

 

From a promising design to a patient

 

Basecamp’s six therapeutic programmes are still at the discovery or lead optimisation stage. Its most ambitious idea is to make CAR-T cells inside a patient’s body, rather than removing immune cells, modifying them in a laboratory and returning them. The company says its early results are promising, but none of these programmes has reported results from a human trial.

The $140 million gives Basecamp more time and resources to test whether its AI-designed molecules can become treatments. It also lets the company gather genetic data from more places. Somewhere in those samples may be a useful biological tool that researchers would never have found by studying familiar laboratory organisms alone.