DeepCyte, a techbio company using AI and single-cell biology to identify and explain drug toxicity, has launched with $1.5 million in seed funding. Co-founded by Theodore Alexandrov, Ph.D., and Shawn Owens, the company is building tools that tell biopharma teams not just whether a drug causes harm, but which cells it affects and why. It operates out of Delaware and Copenhagen.

Around 30% of clinical trial failures are linked to toxicity or unmanageable side effects, and that figure has not improved meaningfully despite decades of new screening methods. The standard tools, animal models and bulk assays, measure an average response across a tissue sample rather than tracking how individual cells react. That distinction matters. A drug can affect certain cells very differently from others, and those differences are often where the real warning signs live. The FDA has formally called for drug development to move away from animal testing as the default, toward human-based models that predict drug reactions more accurately before trials begin.

The round was backed by Carl J. G. Evertsz, a medtech executive, former CEO, and investor, who will serve as Board Chair.

Why Drug Toxicity Testing Has Long Been a Problem for Biopharma

Both the FDA and the European Medicines Agency are updating their regulations to support non-animal testing methods, including organ-on-a-chip systems and AI-based models. Organ-on-a-chip refers to small laboratory devices engineered to mimic how human organs behave, allowing researchers to observe drug effects on human-like tissue. Both agencies are now working on formal guidance for validating and accepting these approaches. For biopharma teams under pressure to avoid expensive late-stage failures, the current options are narrow. Most available methods either cannot explain the mechanism behind a toxic effect or cannot separate how a drug behaves across different cell types within the same sample.

Single-cell biology does what the name suggests: it studies one cell at a time, rather than producing an average across millions. The field has advanced considerably over the past decade. When that level of precision is combined with AI, patterns of toxicity that standard tests would miss become visible. DeepCyte is working from the view that this kind of cell-level, mechanism-aware evidence will become a routine expectation in drug safety testing, particularly as regulators push for data that is more specific and directly relevant to human biology.

How DeepCyte Plans to Use the Seed Funding

The company has not published a detailed breakdown of how the capital will be spent. Based on its launch materials, the focus is on developing its two products, expanding the proprietary datasets that train the AI, and building commercial relationships in the biopharma market. Both products are at an early stage, and the data atlases that the AI depends on will require continued investment to grow.

The stated near-term goal is to give drug developers clearer, earlier signals about toxicity risk, before problems force a trial to stop. Further out, DeepCyte aims to help move the industry away from animal models and bulk assays toward testing grounded in human cellular data. As the FDA continues to update its regulatory expectations, that direction becomes more commercially relevant.

DeepCyte’s Two Products: MetaCore and DeeImmuno Explained

DeepCyte launched in April 2026 with two products designed to work together. The first, MetaCore, is a high-throughput platform that analyzes the molecular contents of individual cells. It uses laser-based sampling and mass spectrometry, a method where a laser extracts material from a cell so its chemical makeup can be measured, to build detailed profiles of what is happening inside cells. MetaCore is built to process large numbers of samples with minimal preparation and at what the company calls an enabling cost. The data it generates forms large, structured atlases used to train the AI.

The second product, DeeImmuno, takes MetaCore data and applies machine learning to predict how toxic a drug is, identify biological markers linked to that toxicity, and trace the molecular mechanisms behind it. In testing on 100 drugs kept separate from the training data, DeeImmuno identified 17 specific toxicity mechanisms with 94% accuracy. According to the company, that level of mechanistic detail is not achievable with conventional testing methods.

“DeepCyte’s mission is to reveal and prevent toxicity in every cell, at scale, before drugs reach patients. By combining advances in AI and single-cell biology, we predict not only whether a drug is toxic, but also why.”

Theodore Alexandrov, Ph.D., CEO and co-founder of DeepCyte

Alexandrov previously developed METASPACE, a cloud platform used by thousands of researchers globally, and co-founded SCiLS GmbH, which was later acquired by Bruker. His background spans computational research and life sciences commercialisation, which reflects the mix of skills DeepCyte is built around.

Origianl Creator: Ekaterina Pisareva
Original Link: https://justainews.com/companies/funding-news/deepcyte-raises-1-5m-to-build-ai-tools-that-detect-drug-toxicity-before-clinical-trials/
Originally Posted: Sun, 12 Apr 2026 14:18:02 +0000