DOPtools: a Python platform for descriptor calculation and model optimization. Overview and usage guide

Pavel Sidorov,Said Byadi,Philippe Gantzer,Timur Gimadiev

DOI: https://doi.org/10.26434/chemrxiv-2024-23v3c

2024-07-16

Abstract:DOPtools (Descriptors and Optimization tools) platform is a Python library for calculation of chemical descriptors and hyperparameters optimization, building and validation of QSPR models. While a variety of existing tools and libraries can calculate various molecular de- scriptors, their output format is often unique, which complicates their connection to standard machine learning libraries. DOPtools provides a unified API for the calculated descriptors as an input for the scikit-learn library. Moreover, DOPtools has a command line interface for automatic calculation of various descriptors on server side and for eventual hyperparameters optimization of statistical models. The modular nature of the code allows easy additions of algorithms if required by the end user. The code for the platform is freely available at GitHub: https://github.com/POSidorov/DOPtools.

Chemistry