Skip to content

Isomorphic Labs

Request correction

An Alphabet-backed artificial-intelligence company applying computational biology and machine learning to drug discovery.

Last updated October 14, 2026

AI crawler reads

8requests

4companies
  • OpenAI4
  • Amazon2
  • ByteDance1
  • Perplexity1
Profile
8
Controversies
0
JSON · MD
0
First read
Oct 29, 2026
Latest read
Nov 4, 2026

From this site's server log

Overview

Isomorphic Labs is a British artificial-intelligence company focused on applying machine learning to biomedical research and drug discovery. Based in London, the company was incorporated on February 24, 2021, and publicly announced on November 4, 2021. It was established under Alphabet as a spin-off from DeepMind, the AI research laboratory co-founded and led by Demis Hassabis. Hassabis is also the founder and chief executive of Isomorphic Labs. The company’s scientific premise is that advances in artificial intelligence can make the discovery of medicines more predictable, faster, and less expensive. Rather than treating drug research solely as a laboratory-screening exercise, Isomorphic Labs uses computational models to study biological structures, disease mechanisms, molecular interactions, and potential therapeutic compounds. Its work draws on technologies developed by DeepMind, especially the AlphaFold family of systems, which has demonstrated the ability to predict protein structures and other biomolecular configurations from sequence and related information. AlphaFold is relevant to drug discovery because the shape and interaction patterns of proteins influence how diseases develop and how medicines work. Better structural predictions can help researchers identify therapeutic targets, understand binding sites, design candidate molecules, and assess how proteins interact with ligands, antibodies, and other biologically important molecules. Isomorphic Labs presents this approach as an effort to build a more integrated computational platform for discovering medicines rather than as a single software product. The company expanded beyond its London base with a second office in Lausanne, Switzerland, announced in December 2022. It has also pursued collaborations with established pharmaceutical companies. In January 2024, Isomorphic Labs announced partnerships with Novartis and Eli Lilly focused on using AI in drug discovery and research. These relationships indicate a business model combining internal computational capabilities with pharmaceutical partners’ disease expertise, development infrastructure, and clinical experience. In May 2024, Google DeepMind and Isomorphic Labs announced AlphaFold 3, a newer AI and foundation-model system designed to predict the structures and interactions of proteins and other biomolecules, including drug-like ligands and antibodies. The model was trained using a large body of known protein information, and access to the AlphaFold server was made available for non-commercial research. AlphaFold 3 broadened the scope of earlier protein-folding systems, although subsequent benchmarking presented a more qualified picture: preliminary CASP16 results reported in November 2024 found that AlphaFold 3-based approaches did not significantly outperform older or hybrid methods for protein-ligand interaction prediction in the relevant pharmaceutical-target category. Isomorphic Labs raised $600 million in its first external funding round in April 2025, with Thrive Capital leading the round. The financing marked a significant step in the company’s development as a privately held business and supported its effort to build drug-discovery capabilities beyond research demonstrations. The company subsequently described a Drug Design Engine that it said improved protein-ligand structure-prediction performance, estimated small-molecule binding affinities, and identified possible binding pockets from amino-acid sequence information. Those claims represent the company’s stated capabilities and should be distinguished from independent validation. Isomorphic Labs remains an active, privately held AI-biotechnology company. Its identity is closely associated with the broader AlphaFold ecosystem, but its strategic objective is more expansive: to use AI models, biological data, and pharmaceutical collaboration to design and advance potential medicines. Public information continues to provide more detail about its r…

History

Isomorphic Labs was incorporated in the United Kingdom on February 24, 2021, and announced publicly on November 4 of the same year. It was created under Alphabet as a spin-off from DeepMind, Alphabet’s AI research laboratory. Demis Hassabis, a DeepMind co-founder and chief executive, founded Isomorphic Labs and became its chief executive as well. The new company was intended to apply advances in artificial intelligence to one of the most difficult and costly areas of biomedical research: discovering medicines that can interact with disease-relevant biological targets. The company’s scientific foundation is closely connected to DeepMind’s AlphaFold work. AlphaFold demonstrated that machine-learning systems could predict protein structures with high accuracy, providing researchers with useful information about the three-dimensional forms of proteins. These structures are important in drug discovery because a candidate molecule often works by binding to a particular region of a protein and altering its activity. Structural prediction can therefore support target identification, binding-site analysis, molecular design, and prioritization of compounds for laboratory testing. Isomorphic Labs was established as a separate company rather than simply as a research program within DeepMind. Its remit is broader than protein folding alone: it aims to combine AI models, biological data, computational chemistry, and pharmaceutical expertise in an integrated drug-discovery process. Public descriptions have emphasized the potential to improve the efficiency and success rate of research, although the company has not publicly disclosed a conventional portfolio of approved medicines or detailed clinical results. In December 2022, Isomorphic Labs announced that it would establish a second office in Lausanne, Switzerland. The expansion gave the company an additional European location and reflected its effort to develop a broader scientific and operational base. The company’s pharmaceutical relationships became more prominent in January 2024, when it announced partnerships with Novartis and Eli Lilly. The collaborations were intended to apply AI to drug discovery and related research. Such partnerships connect Isomorphic Labs’ computational capabilities with the pharmaceutical companies’ experience in therapeutic areas, experimental validation, development, and commercialization. In May 2024, Google DeepMind and Isomorphic Labs announced AlphaFold 3. Compared with earlier AlphaFold systems, AlphaFold 3 was presented as a broader model for predicting interactions among proteins and other biological molecules, including drug-like ligands and antibodies. The system was trained using a large collection of known protein information, and an AlphaFold server made the technology available for non-commercial research. The announcement reinforced Isomorphic Labs’ connection to the AlphaFold family while also highlighting the company’s interest in molecular design and drug development. The reception of AlphaFold 3 was not uniformly positive. Preliminary results from CASP16, reported in November 2024, indicated that AlphaFold 3-based models did not significantly outperform older methods for protein-ligand interaction prediction in the relevant pharmaceutical-target category. ClusPro and CoDock were reported as the strongest performers in that section, using combinations of AlphaFold 2-based predictions, visual inspection, and manual adjustments. These results illustrated the continuing difficulty of translating protein-structure prediction into consistently superior drug-binding predictions. In April 2025, Isomorphic Labs announced a $600 million external funding round led by Thrive Capital. It was the company’s first reported external financing round and represented a major capital commitment to its private-company expansion. The financing helped distinguish Isomorphic Labs from a purely internal Alphabet research initiative and supported its stated ambition to develop practical AI-driven drug-discovery systems. The company later announced a Drug Design Engine. Isomorphic Labs said the system doubled AlphaFold 3’s performance on a protein-ligand structure-prediction generalization benchmark, estimated small-molecule binding affinities more accurately than physics-based methods at a fraction of the time and cost, and found new binding pockets using only amino-acid sequence information. These are company-reported capabilities; the available reference material does not establish independent clinical validation or regulatory approval. Isomorphic Labs remains a privately held, active company headquartered in London, with a second European office in Lausanne and an international research and partnership orientation. Its historical development has followed three connected tracks: separation from DeepMind under Alphabet, expansion of AlphaFold-related biological modeling into drug discovery, and the creation of pharmaceutical partnerships and externally funded product-development capabilities.

  • 2026

    Drug Design Engine announced

    [1]

    The company announced a Drug Design Engine and reported improvements in protein-ligand prediction, binding-affinity estimation, and binding-pocket discovery.

  • 2025

    First external funding round

    [1]

    Isomorphic Labs raised $600 million in its first external funding round, led by Thrive Capital.

  • 2024

    Pharmaceutical partnerships announced

    [1]

    Isomorphic Labs announced AI drug-discovery collaborations with Novartis and Eli Lilly.

  • 2024

    AlphaFold 3 announced

    [1]

    Google DeepMind and Isomorphic Labs announced AlphaFold 3, extending structure and interaction prediction to additional biomolecular systems.

  • 2022

    Second office announced in Lausanne

    [1]

    The company announced Lausanne, Switzerland, as the location of its second office.

  • 2021

    Company incorporated and publicly announced

    [1]

    Isomorphic Labs was incorporated on February 24, 2021, and publicly announced on November 4, 2021, as an Alphabet-backed spin-off from DeepMind.

Products and positioning

A frontier AI-biotechnology company seeking to transform pharmaceutical research through computational biology, structure prediction, molecular design, and strategic partnerships.

  • Computational drug discovery · 2021

    AI Drug Discovery Platform

    Isomorphic Labs develops computational systems intended to support multiple stages of pharmaceutical research, including biological-target analysis, molecular interaction prediction, compound design, and prioritization of candidates for experimental work. Public materials describe the platform as an integrated AI-driven approach rather than a single consumer software product. The company’s systems are designed to complement laboratory and pharmaceutical expertise, with partnerships providing routes for testing and advancing potential discoveries.

  • Protein structure and biomolecular modeling · 2021

    AlphaFold-related applications

    Isomorphic Labs draws on the AlphaFold technology developed by DeepMind to analyze protein structures and biological interactions relevant to drug discovery. These applications can help researchers study disease targets, infer possible binding regions, and model interactions between proteins and drug-like molecules. AlphaFold is not itself an approved medicine or a conventional commercial drug product; in Isomorphic Labs’ context, it is part of a broader computational research strategy.

  • Biomolecular interaction prediction · 2024

    AlphaFold 3

    Announced in 2024 by Google DeepMind and Isomorphic Labs, AlphaFold 3 is an AI and foundation-model system intended to predict structures and interactions involving proteins and other biomolecules, including ligands and antibodies. An AlphaFold server made the system available for non-commercial research. Subsequent benchmarking indicated that its performance advantages were not uniform across protein-ligand prediction tasks, making independent evaluation important when assessing its practical drug-discovery value.

  • AI molecular design · 2026

    Drug Design Engine

    The Drug Design Engine is a company-announced system for computational drug design. Isomorphic Labs said it improved protein-ligand structure-prediction performance relative to AlphaFold 3 on a generalization benchmark, estimated small-molecule binding affinities, and identified previously unrecognized binding pockets using amino-acid sequence information. The available reference does not establish whether the engine is commercially available or whether its reported results have been independently validated.

Brand decisions

  • 2025

    Raise external capital for AI drug discoveryStrategy

    Background

    Isomorphic Labs sought capital to develop its private-company operations and expand computational drug-discovery capabilities beyond its origins as an Alphabet and DeepMind spin-off.

    Action taken

    The company raised $600 million in its first external funding round, led by Thrive Capital.

    Aftermath

    The financing provided a substantial external funding base for continued platform development and pharmaceutical activities.

    External funding raised. $600 million (April 2025)

  • 2024

    Launch AlphaFold 3Product launch

    Background

    Earlier AlphaFold systems focused primarily on protein-structure prediction, while drug discovery also requires understanding interactions with ligands, antibodies, and other biomolecules.

    Action taken

    Google DeepMind and Isomorphic Labs announced AlphaFold 3 and provided access through the AlphaFold server for non-commercial research.

    Aftermath

    The release broadened the public discussion of AI-assisted molecular modeling, although CASP16 preliminary results later showed that the system did not significantly outperform older approaches in one protein-ligand prediction category.

  • 2021

    Establish Isomorphic Labs as a DeepMind spin-offStrategy

    Background

    Alphabet and DeepMind had developed AI capabilities with potential applications in biology, while drug discovery remained a lengthy and high-risk research process.

    Action taken

    Isomorphic Labs was formed as a separate Alphabet-backed company led by Demis Hassabis, with a mandate to apply AI and computational biology to medicine discovery.

    Aftermath

    The company became the principal corporate vehicle associated with Hassabis’s effort to extend AlphaFold-related advances into pharmaceutical research.

Sources2 sources

Cite this profile: Cite the canonical profile. /brand-wiki/isomorphic-labs · Editorial policy · How profiles are compiled