Embedded Files
Molecular Descriptor Calculator for QSAR and Machine Learning
PyDescriptorC*: A Brief idea
PyDescriptorC*: A Brief idea
PyDescriptorC* is a Python-based extension of PyDescriptor (https://doi.org/10.1016/j.chemolab.2017.08.003)
Provides interpretable and reproducible descriptors directly in CSV format.
Validated on diverse datasets, PyDescriptorC* provides interpretable descriptors and deep information often missed by traditional descriptor tools.
Broad descriptor coverage: constitutional, topological, geometric/spatial, circular fingerprints, quantum chemical, and chirality-specific descriptors. It computes 112,194 molecular descriptors, including 15,150 chirality-aware descriptors (~13.5%).
Just upload your .zip or .rar file containing all molecules in mol2 format using the Google form. Size limit is 100 MB for zip or rar file. Use a separate mol2 file for each molecule. Please fill the following Google form to upload your molecules- no other data required:
Upload your molecules here to use PyDescriptorC*.
Google form link: https://forms.gle/oJv1pgD8TmH8Td6R7
The image on the right represents the 3D depiction of two molecular descriptors selected from the 112,194 descriptors calculated by PyDescriptorC*. For more details about the image, please refer to the following reference: Vijay H. Masand et al, Journal of Molecular Structure 1175 (2019) 481e487 (DOI: https://doi.org/10.1016/j.molstruc.2018.07.080).
Comparative/Benchmark testing with other descriptor calculators:
Comparative/Benchmark testing with other descriptor calculators:
T.B. Kimber, S. Engelke, I.V. Tetko, E. Bruno, G. Godin, Synergy effect between convolutional neural networks and the multiplicity of SMILES for improvement of molecular prediction, arXiv preprint arXiv:1812.04439, (2018).
S. Sosnin, D. Karlov, I.V. Tetko, M.V. Fedorov, Comparative study of multitask toxicity modeling on a broad chemical space, Journal of chemical information and modeling, 59 (2018) 1062-1072.
Developers:
Developers:
The platform is jointly developed by researchers from India, Saudi Arabia, and Croatia.
Vidya Bharati Mahavidyalaya, India
Dr. D. Y. Patil Institute of Technology, India
Dr. Sami A. Al-Hussain
Imam Mohammad Ibn Saud Islamic University, Suadi Arabia
Dr. Rahul D. Jawarkar
R. Gode Institute of Pharmacy, India
J.J.S. University of Osijek, Croatia
Dr. Magdi E.A. Zaki
Imam Mohammad Ibn Saud Islamic University, Suadi Arabia
Reference:
Reference:
If you find the platform useful in your research, please cite the following article:
Masand, VH, Masand, GS, Al-Hussain, SA, Jawarkar, RD, Rastija, V. and Zaki, MEA (2025) PyDescriptorC*: A Descriptor Calculation Tool for Decoding Chirality Cliffs and Revealing Hidden Patterns in Drug Discovery. RHAZES: Green and Applied Chemistry, 21, 32 - 51. https://doi.org/10.26434/chemrxiv-2025-w3k4n
Users' opinion on PyDescriptorC*:
Users' opinion on PyDescriptorC*:
"I take the opportunity to provide feedback on PyDescriptorC*. In QSAR studies, descriptors are the heart of model building, & essential for generating accurate and robust equations. What stands out about PyDescriptorC* is that which covers all types of properties, especially stereochemical, this feature makes it valuable for comprehensive QSAR studies. Additionally, descriptors are facilitating mechanistic interpretations without requiring prior extensive knowledge about chemistry, making it user-friendly. The prompt responses and excellent support provided further enhance the overall experience. I believe this tool can play a crucial role for researchers in the field. I look forward to incorporating it into my projects and seeing how it continues to evolve. "
Regards,
Dr. Somdatta Y. Chaudhari, Modern College of Pharmacy, Nigdi, Pune, India
Page updated
Google Sites
Report abuse