DeepAutoQSAR

Automated, scalable solution for the training and application of predictive machine learning models

DeepAutoQSAR

Create high-performing machine learning models using state-of-the-art methods

DeepAutoQSAR is a machine learning (ML) solution that allows users to predict molecular properties based on chemical structure. The automated supervised learning pipeline enables both novice and experienced users to train and inference best-in-class quantitative structure activity/property relationship (QSAR/QSPR) models.

Key Capabilities

Streamline model building with fully automated workflows

Automatically compute descriptors and fingerprints, create models with multiple machine learning architectures, and evaluate model performance.

Customize models to your project with unique project-specific descriptors

Provide your own descriptors in CSV format to be used in addition to or instead of those generated by DeepAutoQSAR for a wide range of applications beyond small molecules, such as polymers, organic electronics, catalysis, and more.

Ensure model optimization using best practices

Employ QSAR/QSPR best practices to minimize the likelihood of overfitting or misrepresenting a model’s performance while ensuring maximum predictive model performance.

Understand the domain of applicability using model confidence estimates

DeepAutoQSAR provides uncertainty estimates alongside model predictions to help determine how much confidence should be placed on predictions generated for candidate molecules which may lie beyond the model’s training set.

Visualize and analyze results to gain further insights 

Visualize color-coded atomic contributions towards target property facilitating ideation of novel chemistry. Visualize and analyze DeepAutoQSAR metrics reports and plots in Maestro to enable further experiments — quickly learn what model architectures are most effective and how models generalize on holdout sets.  

Scalable training to support small or large datasets

Use classical ML methods like boosted trees on smaller datasets while also supporting the largest scale QSAR/QSPR models using graph neural networks and other modern deep learning approaches.

Case studies & resources

Discover how Schrödinger technology is being used to solve real-world research challenges.

De novo design of hole-conducting molecules for organic electronics

Accelerating the design and optimization of OLED materials using active learning

Machine-learned force Fields for improved materials modeling

Related Products

Learn more about the related computational technologies available to progress your research projects.

Active Learning Applications

Accelerate discovery with machine learning

FEP+

High-performance free energy calculations for drug discovery

Glide

Industry-leading ligand-receptor docking solution

De Novo Design Workflow

Fully-integrated, cloud-based design system for ultra-large scale chemical space exploration and refinement

Jaguar

Quantum mechanics solution for rapid and accurate prediction of molecular structures and properties

Maestro

Complete modeling environment for your molecular discovery

LiveDesign

Your complete digital molecular design lab

MS Informatics

Automated machine learning tools for materials science applications

Quantum ESPRESSO Interface

Integrated graphical user interface for nanoscale quantum mechanical simulations

Publications

Browse the list of peer-reviewed publications using Schrödinger technology in related application areas.

Materials Science
Leveraging High-throughput Molecular Simulations and Machine Learning for Formulation Design
Materials Science
Development of Scalable and Generalizable Machine Learned Force Field for Polymers
Life Science
Pathfinder-Driven Chemical Space Exploration and Multiparameter Optimization in Tandem with Glide/IFD and QSAR-Based Active Learning Approach to Prioritize Design Ideas for FEP+ Calculations of SARS-CoV-2 PLpro Inhibitors
Materials Science
Benchmarking Machine Learning Descriptors for Crystals
Materials Science
Machine Learning for the Design of Novel OLED Materials
Life Science
A Descriptor Set for Quantitative Structure-Property Relationship Prediction in Biologics
Materials Science
Active Learning Accelerates Design and Optimization of Hole-Transporting Materials for Organic Electronics
Materials Science
Design of organic electronic materials with a goal-directed generative model powered by deep neural networks and high-throughput molecular simulations
Materials Science
Design and Synthesis of Novel Oxime Ester Photoinitiators Augmented by Automated Machine Learning
Life Science
Digitalisierung: molekulares Design plattformisieren

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.