Biosimulation Platform
Biology Matters
Cellworks predictions are powered by mechanistic biology models, biostatistics and machine learning.
The Cellworks Platform performs computational biosimulation of protein-protein interactions, using mechanistic biology-based models. It enables in silico modeling of cellular tissue and disease with multi-omic data, evaluating how personalized treatments interact with the patient’s unique disease molecular network.
One Platform Powers Pharma and Clinical Solutions
Best-in-Class PTRS* for Pharma
* Probability of Technical and Regulatory Success PRECISION
PATIENT
SELECTION
- Identify patients who will respond to a specific therapy.
THERAPY
OPTIMIZATION
- Advance the line of therapy.
- Discover new therapeutic uses for existing drugs with 20,000,000+ drug combinations.
- Optimize drug dosages to maximize efficacy while minimizing toxicity.
IN SILICO
FIRST-IN-HUMAN
DRUG EFFICACY
STUDIES
- Predict an investigational agent’s clinical efficacy as early as the drug discovery stage.
Personalized Therapy Decision Support
SINGULA™
Predicts personalized response to Standard of Care therapies and therapies matching a patient’s biomarkers.
VENTURA™
Ranks the most efficacious personalized combinations of FDA-approved drugs for refractory patients.
Computational Biology Model (CBM)
The CBM is a highly curated and validated mechanistic network of human genes, molecular species and molecular interactions. This model along with associated drug models are used to biosimulate the impact of specific compounds or combinations of drugs on a patient or a cohort and produce therapy response predictions.
Mechanistic
Network
- 6,000+ Human genes
- 55,000+ Molecular species
- 600,000+ Molecular interactions
- 80+ Hallmarks of Disease
- 350+ Drugs
- 20,000,000 Unique Therapies
Validated
Cell Types
- Cancer cell
- Fibroblast
- B cell
- T cell
- Macrophage
- Dendritic cell
- Endothelial cell
- Adipocyte
Personalized Disease Model
A personalized disease model is generated from the mechanistic biology model using multi-omic data for each individual patient. The personalized disease model is then used to run ‘what-if’ experiments with any desired therapy and to predict patient-specific drug response.
Mechanistic Learning and AI
The Cellworks Platform employs a mechanistic learning approach that leverages both biological knowledge and data. Mechanistic biology models are used to simulate patient-specific drug responses to produce highly accurate predictions of individual therapy responses. AI and machine learning techniques are utilized to build and train the models on the platform.
Explainable Predictions
The Cellworks Platform provides detailed explanations of its therapy response predictions on a per-patient basis at a molecular level. These predictions are based on the underlying biological mechanisms involved in disease and treatment response.
20+ Million Combinations Modeled
There are more than 350 approved drug models within the Cellworks Platform, including a quantitative model of drug targets, binding affinities, and reaction rate constants coupled with pharmacokinetic and drug economic data. These drugs can be combined into 20+ million combination therapies for modeling.
Scalable, Cloud-based Infrastructure
The Cellworks cloud-based infrastructure can run millions of prediction biosimulations in parallel and produce reports for an unlimited number of patients per day. The platform can scale differential equations and other mathematical structures at the rate of millions per second per biosimulation.