LiveDesign Learning
Your complete solution for rapid AI/ML molecular property predictions
Efficiently integrate AI/ML modeling into your drug discovery workflows
LiveDesign Learning is a module in LiveDesign for training and deploying state-of-the-art AI/ML models for small molecule drug discovery. By treating datasets as dynamic information feeds that evolve as scientists explore new chemistry, LiveDesign Learning delivers AI/ML models that are optimized to the most relevant project compounds and provides users with the critical information needed to determine if the models are sufficiently accurate for prospective use. The technology allows teams to streamline the optimization and validation of AI/ML models, expanding their application within drug discovery programs.
Democratize AI/ML model generation with a fully automated workflow to build and monitor molecular property and multi-parameter optimization (MPO) predictions
Deploy models prospectively with confidence via data visualizations and performance metrics that monitor a model’s predictive power in evolving chemistry
Easily integrate models into team decision-making and design workflows via a centralized collaborative platform
Benefit from highly scalable computational pipelines based on a modern cloud infrastructure to model hundreds of properties and make millions of predictions
Seamless integration of cutting-edge AI/ML models into a centralized design platform
Examples of properties predicted with LiveDesign Learning:
- Experimental data: Potency (on- and off-target), ADME, DMPK, and safety endpoints
- In silico physics data: Any properties stored in LiveDesign, e.g. FEP+
Key Features
Rapid property predictions in an interactive design environment to triage compound ideas in minutes for large library screening
Simple user interface for generating AI/ML models and retraining them as the chemistry evolves
Built-in transfer learning across AI/ML models to improve low-data performance
Leverages DeepAutoQSAR to train and tune state-of-the art model ensembles
Visual representation of atom-level attribution
Integrated model management to track ML performance over time
Case Studies & Webinars
Discover how Schrödinger technology is being used to solve real-world research challenges.
LiveDesign Learning: Complete solution for rapid AI/ML molecular property predictions
Empowering scientists with integrated AI/ML modeling for rapid molecular property predictions
Benchmark study of DeepAutoQSAR, ChemProp, and DeepPurpose on the ADMET subset of the Therapeutic Data Commons
Related Products
DeepAutoQSAR
Automated, scalable solution for the training and application of predictive machine learning models
Training & Resources
Online certification courses
Level up your skill set with hands-on, online molecular modeling courses. These self-paced courses cover a range of scientific topics and include access to Schrödinger software and support.
Tutorials
Learn how to deploy the technology and best practices of Schrödinger software for your project success. Find training resources, tutorials, quick start guides, videos, and more.
Other Resources