Predictive Toxicology
Improve the safety profile of your drugs with accurate, highly predictive in silico toxicity screening
De-risk your drug discovery early with in silico toxicity screening
Many drug discovery programs fail due to ADMET liabilities, such as binding to anti-targets associated with adverse patient outcomes. While the use of in vitro safety panels has substantially improved drug safety, the approach is slow and costly and only used on the most promising compounds in the later stages of drug discovery. Schrödinger’s predictive toxicology solution leverages physics-based computational approaches with AI/ML technologies to enable a broadly predictive, in silico, scalable solution that offers fast ligand evaluation and early de-risking.
Benefits
Save more than 2X in costs
by incorporating in silico toxicity screening early to de-risk projects with accurate structure-based predictions
Reduce evaluation time from weeks to a single day
with cloud computing resources and get affinities and all-atom 3D structures in real time
Screen virtual compounds
with structure-based computational modeling to rapidly mature your chemical matter before committing to synthesis
Enable rational design with actionable readouts detailing molecular level insights to design out liabilities
Current Predictive Toxicology Solutions
Cloud-based in silico predictive toxicology screening
-
Capabilities:Schrödinger presently supports in silico screening of 55 kinases using our new cloud-based Predictive Toxicology Kinase Panel -
Requirements: No prior experimental data is required to run the Predictive Toxicology Kinase Panel
-
How to access: • Available through Schrödinger web services
• Run on Schrödinger-managed secure cloud resources, includes all licenses and compute fees
Capabilities:Schrödinger presently supports in silico screening of 55 kinases using our new cloud-based Predictive Toxicology Kinase Panel 
Structure enablement of ADMET anti-targets with IFD-MD and FEP+
-
Capabilities:
• IFD-MD technology within Maestro supports the generation of multiple anti-target/ligand poses – for several commonly encountered antitargets: hERG, PXR, CYP3A4, CYP2D6, and CYP2C9
• FEP+ is used to prosecute IFD-MD generated models using user supplied data -
Requirements: A small congeneric series with assay data against the anti-target of interest – a minimum of 9 assayed ligands are typically required
-
How to access:
• Available through Schrödinger software licenses
– Run on your local in-house cluster or cloud resources
• Available through Schrödinger modeling services
– Run on Schrödinger compute resources, includes all license, compute, and service fees
Featured Webinar
- Life Science
- Webinar
Predictive toxicology solutions: Actionable, structure-based insights to dial-out tox liabilities early
-
Date & Time:
December 16th, 2025 | 11:00AM EST
-
Location:
Virtual
Date & Time: 
Location: Schedule a demo with predictive toxicology solutions
Start screening off-targets with computational predictive models—contact us today to discuss how you can start using Schrödinger’s solutions on your programs.
Don’t see your off-target of interest in the current lists above? Reach out so we can help.
Form submitted
Thank you, we’ll be in touch soon.