Anthropic has released Claude Science, an AI workbench for scientists that consolidates fragmented research tools, including over 60 scientific databases and connectors pre-configured for genomics, proteomics, structural biology, and more, into a single reasoning layer. The platform joins an increasingly crowded ecosystem of tech platforms specialized for biology and aims to accelerate scientific discovery by making domain expertise more accessible.
Anthropic’s life science partners are delivering applications. Basecamp Research is targeting global public health, where drug-resistant infections play a role in nearly five million deaths per year. The London-based team has announced that its antibiotic design and vaccine target prediction EDEN models will now be available through Claude Science.
A metagenomic foundation model, EDEN demonstrated a 97% success rate when designing functional peptides with high potency against World Health Organization (WHO) critical-priority and multidrug-resistant pathogens. The work was done in collaboration with César de la Fuente, PhD, presidential associate professor at the University of Pennsylvania.
In a Claude Science demo, Oliver Vince, PhD, co-founder at Basecamp, uploaded a sample patient microbiology report. When given a simple natural language prompt, the platform designed peptides, predicted their efficacy, and provided a shortlist of candidates most likely to succeed in experiments in minutes.
While generating human-ready antibiotics at the click of a button is still a step away, Vince said democratizing these tools is a powerful first step, particularly for researchers in regions where accelerated computing infrastructure is not readily accessible.
“Most models require you to be a computational scientist,” Vince told GEN Edge. “Now, potentially any clinician in the world can chat with Claude and design an antibiotic that may work.”
“From a strategic perspective, you want the people with the most agency to solve the problem,” added Phil Lorenz, PhD, CTO at Basecamp. “Not the model builders who are two or three steps removed.”
Full stack
Founded in 2019, Basecamp has spent its initial years building a full computational stack spanning data, models, and therapeutic assets.
In addition to antibiotics and vaccines, the company’s U.S. office, based in Cambridge and led by Jonathan Finn, PhD, Basecamp CSO and former CSO of Tome Biosciences, has fine-tuned EDEN for programmable gene insertion. The approach places large therapeutic DNA sequences at precise locations in the human genome, expanding upon CRISPR-based approaches that use small edits to address a limited number of indications.
EDEN’s generalizability is enabled by training on BaseData, the company’s proprietary dataset composed of 9.8 billion protein sequences collected over 200 diverse and extreme locations, including thermal springs, polar ice, and high-altitude plateaus, across more than 30 countries. The database provides a 10-fold expansion of known protein diversity when compared to all public databases combined.
In March, the team published the compounding advantages of BaseData on model performance in a technical report on scaling laws for metagenomics. Basecamp is steadily pushing forward that data diversity through the Trillion Gene Atlas, a partnership with Anthropic, NVIDIA, PacBio, and Ultima Genomics that aims to scale BaseData 100-fold over the next two years.
Vince emphasizes that model deployment and integration into real-world workflows will be critical for these models to reach their full potential. Basecamp anticipates releasing more applications over the next year.
“I think it will surprise people what these models can do,” he said.
The post Claude Science is Here, Antibiotics Designed by Text Prompt Among Applications appeared first on GEN – Genetic Engineering and Biotechnology News.
