Skip to main content

DeepCyte

Category: AI in Biotech / Drug Discovery

DeepCyte is a techbio company building AI-powered single-cell toxicology tools for drug development, combining single-cell metabolomics with machine learning to predict and explain drug toxicity before clinical trials. DeepCyte was founded in 2026. The company is led by Theodore Alexandrov. Based in Wilmington, Delaware, United States; Copenhagen, Denmark. Team size: 2-10. Total funding raised: $1.5M. Latest round: Seed. Key investors include Carl J. G. Evertsz.

Founded
2026
Headquarters
Wilmington, Delaware, United States; Copenhagen, Denmark
Team size
2-10
Total funding
$1.5M

Value proposition

DeepCyte combines single-cell metabolomics with AI to detect metabolic changes within individual cells, predicting toxicity and revealing mechanisms invisible to conventional methods. DeeImmuno predicted 17 detailed toxicity mechanisms with 94% accuracy on 100 held-out drugs.

Products and solutions

MetaCore (high-throughput single-cell metabolomics platform built on laser-based sampling and mass spectrometry technology), DeeImmuno (AI toxicology solution that predicts toxicity class, identifies biomarkers, and infers molecular mechanisms of toxicity, trained on proprietary single-cell metabolomics atlases)

Unique value

Only company combining high-throughput single-cell metabolomics (MetaCore) with purpose-built AI (DeeImmuno) to provide mechanistic toxicity insights at single-cell resolution, moving beyond animal models and bulk assays toward human-centric drug safety testing.

Target customer

Biopharma companies (preclinical safety assessment teams), drug discovery and development organizations

Industries served

Pharmaceutical drug development, Biotechnology, Drug safety/toxicology

Technology advantage

World-leading high-throughput single-cell metabolomics technology built on 10+ years of academic research; award-winning laser-based sampling and mass spectrometry; AI models trained on proprietary single-cell metabolomics atlases; MetaCore generates atlas-scale AI-ready datasets with minimal sample preparation at enabling cost

How they differentiate

Single-cell resolution metabolomics platform (MetaCore) combined with AI models (DeeImmuno) trained on proprietary atlases; achieves 94% accuracy across 17 detailed toxicity mechanisms; provides mechanistic insight (not just risk scores) showing how cellular programs get rewired by drugs; human-centric approach aligned with FDA/EMA shift away from animal models

Main competitors

Quris AI (AI drug safety prediction using organ-on-chip + ML), Insilico Medicine (AI drug discovery with toxicity prediction), Recursion Pharmaceuticals (AI-enabled drug discovery with safety assessment)

Key partnerships

Bruker (partnership leveraging Bruker mass spectrometry technology for HT SpaceM and 13C-SpaceM single-cell metabolomics methods)

Major milestones

Launched April 2026 with $1.5M seed funding, Introduced MetaCore platform and DeeImmuno AI solution, DeeImmuno achieved 94% accuracy predicting 17 toxicity mechanisms on 100 held-out drugs, CEO's prior company SCiLS GmbH acquired by Bruker

Market positioning

Early-stage techbio startup positioned at the intersection of single-cell biology and AI for drug toxicology, targeting the critical gap in preclinical safety assessment where 20-30% of drug failures are linked to toxicity

Geographic focus

Global (US and EU biopharma market)

About Theodore Alexandrov

Ex-EMBL Team Leader / Head of Metabolomics Core Facility; Ex-UC San Diego; Developed METASPACE cloud platform used by thousands of researchers globally; Co-founded SCiLS GmbH (later acquired by Bruker, NASDAQ: BRKR); PhD in Computer Science

Official website: