CellMetrix AI

What CellMetrix AI does

CellMetrix AI is a Boston-based, privately held AI-native drug discovery company that builds foundation models directly on multi-omic single-cell data — transcriptomic, epigenomic, and proteomic readouts across tens of thousands of individual cells — to predict how compounds will behave in real human cells. Its core technical premise is that most computational drug discovery tools learn from molecular structure alone and lack a model of the cellular state the compound must act on; CellMetrix's models are designed to provide predictions grounded in actual cellular context rather than averaged bulk-tissue proxies.

The platform is delivered as a connected stack of three products currently in private beta: CellMetrix Atlas, a queryable single-cell reference atlas spanning 40+ tissue types that unifies public and partner datasets with AI-predicted perturbation responses; CellMetrix Predict, a REST API and web dashboard that returns predicted toxicity, mechanism of action, and off-target scores with confidence intervals in under 10 minutes and supports batch compound-library screening; and CellMetrix Lab Copilot, an early-access wet-lab workflow assistant that helps scientists design perturbation experiments, choose sequencing panels, and interpret results in plain language. The underlying models are described as GPU-native foundation models trained at scale, with a live model visualization referencing 160,482 cells.

The company is targeting three customer segments — pharma companies, biotech companies, and academic research institutions — with use cases spanning compound toxicity screening, mechanism-of-action prediction, off-target analysis, perturbation experiment design, and reference data access. Distribution is digital-first via REST API and web dashboard with direct partner onboarding through the private beta program. Revenue mechanics are described as likely subscription or usage-based across three streams (Platform Access, Atlas Subscription, Lab Copilot Early Access), but pricing is not publicly disclosed and the company is currently accepting design partners rather than charging for production access. The company is founder-controlled by Dr. Elena Marsh, has no disclosed institutional funding, and operates from 201 Brookline Ave, Suite 900, Boston, MA.

Name

CellMetrix AI

Legal name

CellMetrix AI, Inc.

Website

https://cellmetrix.site

Company type

Private

Operating status

Operating

Short description

CellMetrix AI builds GPU-native foundation models on multi-omic single-cell data to predict compound toxicity, mechanism of action, and off-target effects in real human cells. The Boston-based platform serves pharma, biotech, and academic researchers through Atlas, Predict, and Lab Copilot.

Ownership category

akta.pro rank

Product category

AI Drug Discovery Platform

NAICS

Research and Development in Biotechnology (except Nanobiotechnology) (541714), Scientific Research and Development Services (5417)

SIC

Services-Prepackaged Software (7372), Services-Commercial Physical & Biological Research (8731)

akta.pro primary industry

AI/ML-Enabled Biologic Discovery Platforms (computational design, in silico screening) (HLAAAIAE)

akta.pro secondary industries

Single-Cell & Spatial Omics Discovery Platforms (target ID/validation, functional profiling) (HLAAAIAF), Bioinformatics & Multi-omics Analysis Software (HLAGAJAD), AI/ML Platforms for Drug Discovery & Experiment Optimization (HLAGAJAO)

Keywords

• Single-cell genomics
• Drug discovery platform
• Multi-omic data analysis
• Foundation models biology
• Compound toxicity prediction

Headquarters

HQ city

Boston

HQ country

United States

HQ region

North America

Offices1 record

Markets served

GTM type

B2B

Offering type

Software

Cost components

Technology or R&D, Personnel, Infrastructure, Marketing or Sales, Operations

Revenue model

1. Platform Access (API/Dashboard): CellMetrix Predict provides REST API and web dashboard access for compound screening. Currently in private beta, the revenue model is likely subscription or usage-based for the prediction engine.
2. Atlas Subscription: CellMetrix Atlas provides queryable reference data access, likely offered as a subscription-based service for researchers and pharma partners.
3. Lab Copilot (Early Access): Wet-lab workflow assistant offered under early access, potentially as a subscription or seat-based model for bench scientists.

Go-to-market motion1 record

Distribution channels2 records

Marketing channels2 records

Core offering

CellMetrix AI builds foundation models trained directly on multi-omic single-cell data (transcriptomic, epigenomic, and proteomic) to predict how compounds behave in real human cells. It sells a connected software stack of three products: CellMetrix Atlas (a queryable 40+ tissue-type single-cell reference), CellMetrix Predict (an API and dashboard returning compound toxicity, mechanism of action, and off-target scores with confidence intervals in under 10 minutes), and CellMetrix Lab Copilot (a wet-lab workflow assistant for experiment design and interpretation).

Product overview

CellMetrix AI is an AI-native drug discovery platform built as a connected stack of three core products: CellMetrix Atlas (queryable single-cell reference atlas), CellMetrix Predict (REST API and dashboard for compound prediction), and CellMetrix Lab Copilot (wet-lab workflow assistant). The platform trains foundation models directly on multi-omic single-cell data to predict compound behavior in human cells, offering toxicity scoring, mechanism of action prediction, and off-target analysis with confidence intervals.

Differentiator

Problem solved

Functional benefit

Products and services

• CellMetrix Atlas A queryable single-cell reference atlas spanning 40+ tissue types that unifies public and partner datasets with AI-predicted perturbation responses. Provides a harmonized multi-omic reference with predicted and observed states shown side by side, exportable to notebook or pipeline. Targeted at pharma, biotech, and academic research partners.
• CellMetrix Predict REST API and web dashboard that takes a compound structure and target cell context, returning predicted toxicity, mechanism of action, and off-target scores in minutes. Includes confidence intervals on every score and supports batch screening for compound libraries. Targeted at pharmaceutical and biotechnology drug discovery teams.
• CellMetrix Lab Copilot Assistant embedded in the wet-lab workflow that helps scientists design perturbation experiments, choose sequencing panels, and interpret output without a bioinformatics detour. Offers experiment design suggestions, panel and assay recommendations, and plain-language result interpretation. Targeted at bench scientists in pharma, biotech, and academic labs.

Quantifiable outcome

• Compound to prediction turnaround in under 10 minutes instead of months-long assay pipelines
• +4 more outcomes

Segments2 records

Ideal customer profiles2 records

Technology focussed Yes

API detail

Has API

Yes

API docs

API detail

Core technology

AI maturity

App detail

AI capability4 records

Feature5 records

Scale indicators4 records

Recent moves4 records

Expansion highlights4 records

Emerging players

• Owkin: AI platform for biomedical research applying ML to multi-modal patient and biological data for pharma R&D. Adjacent in combining AI with biological data for pharma partners, with a broader multi-modal data focus.
• Immunai: Applies single-cell genomics and ML to immunology and drug discovery. Comparable in using single-cell data as the substrate for ML-driven pharma insights, with a narrower focus on immune cell states.

Direct peers

• BenevolentAI: AI drug discovery company using knowledge graphs and ML to identify novel targets and drug candidates. Comparable in targeting pharma/biotech buyers with AI-augmented discovery workflows.
• Insitro: Machine learning-driven drug discovery platform that integrates high-throughput cellular data with predictive models. Direct comparable to CellMetrix's foundation-model-on-cellular-data approach and its pharma partnership-led go-to-market.
• CytoReason: Builds cell-level computational models of the immune system and disease biology for pharma R&D. Comparable in its cellular-state modeling focus and pharma partner model, with similar target customers and use cases.
• Valo Health: AI-native drug discovery and development platform combining human-centric data with computational chemistry. Comparable in pursuing a vertically integrated, AI-native drug discovery stack with pharma partnerships.
• Exscientia: AI-driven drug discovery company using computational platforms to design and prioritize novel therapeutics. Overlaps with CellMetrix on AI-native compound prediction, though biased toward ligand-based and structure-based chemistry approaches.
• Atomwise: AI-based small molecule drug design company using structure-based convolutional neural networks. Comparable in providing AI-driven compound screening to pharma, with a different (structure-based) technical foundation.
• Recursion Pharmaceuticals: AI-native drug discovery company that combines high-throughput experimental biology with machine learning to build predictive models of cellular biology. Closely comparable to CellMetrix in its use of cellular-state ML to drive compound prediction, though significantly more mature and capitalized.

Broad incumbents

• Schrödinger: Established computational chemistry and molecular simulation platform serving pharma and materials science. Broader incumbent with a wider portfolio; overlaps with CellMetrix on AI/computational compound prediction but does not specialize in single-cell foundation models.

Market position

Strengths4 records

Weaknesses5 records

Competitive moat4 records

Key risks6 records

Key highlights6 records

Customer concentration

Revenue estimate

Valuation estimate

Number of profiles

Profiles1 record

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about CellMetrix AI