Highly curated Nav1.5 dataset of 1723 unique molecular compounds with corresponding potency values

This dataset was built during a research project, in the field of Computer-Aided Drug Discovery (CADD), funded by the Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery grant. The aim of the project was to build descriptor-based machine learning models for Nav1.5 cardiotoxicity liability predictions. The dataset includes a total of 1723 unique molecular compounds gathered from ChEMBL and PubChem publicly available bioactivity databases. The list is split into 2 sets, 1550 for training and 173 for testing. All molecular compounds are represented in their SMILE format with their corresponding PIC50 potency values.

To access the full original work, please visit the following link:  Manuscript 
Refer to a much larger and latest dataset here: link 

Note: Upon usage of this data, kindly cite both the dataset and the original manuscript describing the curation process as written below.

Arab I, Barakat K. ToxTree: Descriptor-based machine learning models for both hERG and Nav1. 5 cardiotoxicity liability predictions. arXiv preprint arXiv:2112.13467. 2021

Arab, Issar, & Barakat, Khaled. (2021). Highly curated Nav1.5 dataset of 1723 unique molecular compounds with corresponding potency values [Data set]. Zenodo. https://doi.org/10.5281/zenodo.5807731