README.md

Table of Contents

This tutorial tries to cover the most important topics on the various features of QSPRpred, but it is not exhaustive. For more detailed information on the features of the package, please refer to the documentation.The tutorial data is available through OneDrive (just unzip and place the two datasets A2A_LIGANDS.tsv and AR_LIGANDS.tsv in the tutorial_data folder) or recreate the dataset yourself by running tutorial_data/create_tutorial_data.py after you have installed QSPRpred.

The Quick Start tutorial is designed to get you up and running with QSPRpred as quickly as possible while the rest dedicates more time to explain each feature in more detail. The Basics cover the most commonly used functionality of QSPRpred. The Advanced tutorials cover more advanced topics and are designed for users who are already familiar with QSPRpred more in depth or are looking for more niche features. For detailed description of all QSPRpred classes and functions, as well as examples of how to use the command line interface, see the documentation pages.

• Quick Start: A quick start guide to using QSPRpred.
• Basics
  • Data
    • Data Collection with Papyrus: How to collect data with Papyrus.
    • Data Preparation: How to prepare data for QSPRpred.
    • Data Representation: How data is represented in QSPRpred (MolTable, QSPRDataset, etc.).
    • Data Splitting: How to split data into training, validation, and test sets.
    • Descriptors: How to calculate descriptors for molecules.
    • Searching, Filtering and Plotting: How to search and filter data.
    • Applicability Domain: How to calculate the applicability domain of a model.
  • Modelling
    • Classification: How to train a classification model.
    • Logging: How to set-up logging.
    • Model Assessment: How to assess the performance of a model.
  • Other
    • Benchmarking: How to benchmark QSPRpred.
    • Serialization: How to save and load datasets and models.
• Advanced
  • Data
    • Parallelization: How to parallelize data functions across data sets.
    • Custom descriptors: How to use custom descriptors.
    • Custom data splitting: How to use custom data splitting.
  • Modelling
    • Custom models: How to use custom models.
    • Deep learning models: How to use deep learning models.
    • Hyperparameter optimization: How to optimize model hyperparameters.
    • Monitoring: How to monitor model training.
    • Multi-task learning: How to train a multi-task model.
    • PCM modelling: How to prepare data for and train a proteochemometric model.
    • Chemprop models: How to use Chemprop models in QSPRpred.