DiPCell is a specialized computational platform designed to accelerate the drug discovery process for pancreatic cancer, one of the most devastating diseases with a very poor prognosis. Unlike tools focused on peptides, this specific iteration of DiPCell is a web-bench for predicting and screening promiscuous inhibitors—small molecules capable of targeting multiple oncogenic pathways—to improve therapeutic outcomes in pancreatic cancer.
Web Server: https://webs.iiitd.edu.in/raghava/dipcell/
Kumar, R., Chaudhary, K., Singla, D. et al. Designing of promiscuous inhibitors against pancreatic cancer cell lines. Sci Rep 4, 4668 (2014). https://doi.org/10.1038/srep04668
This dataset can also be found on Zenodo at https://doi.org/10.5281/zenodo.20094719
The primary goal of this resource is to identify effective drug candidates by leveraging large-scale pharmacological data. The platform uses Quantitative Structure-Activity Relationship (QSAR) models to predict the efficacy of compounds against various pancreatic cancer cell lines.
-
Model Performance: The QSAR models achieved a maximum Pearson correlation coefficient of 0.86 during 10-fold cross-validation, indicating high predictive reliability.
-
Validation: The models successfully validated known drug-to-oncogene relationships, ensuring the computational predictions align with biological reality.
-
Experimental Testing: The researchers used these models to screen FDA-approved drugs, which were subsequently tested in vitro to confirm their effectiveness.
-
Small Molecule Focus: Specifically designed to predict the inhibition potential of small chemical compounds rather than peptides.
-
FDA-Drug Repurposing: Includes a module for screening existing FDA-approved drugs to identify new applications in treating pancreatic cancer.
-
Efficacy Profiling: The tool identifies the most and least effective drugs for specific pancreatic cancer cell lines.
-
Resistance Mapping: Highlights resistant pancreatic cancer cell lines that require further investigation to uncover the underlying mechanisms of drug resistance.
-
QSAR Prediction: Allows users to input chemical structures to predict their inhibitory activity against pancreatic cancer models.
-
Design Module: Facilitates the design of novel promiscuous drug molecules by analyzing how structural changes impact predicted efficacy.
-
Accelerated Drug Discovery: Reducing the time and cost of finding new leads by prioritizing high-potential candidates through virtual screening.
-
Precision Medicine: Understanding which cell lines respond to specific inhibitors to help tailor potential treatments.
-
Resistance Research: Providing a starting point for scientists to study why certain pancreatic cancers do not respond to standard therapies.
Prof. Gajendra P. S. Raghava (Corresponding Author)
Department of Computational Biology, Indraprastha Institute of Information Technology (IIIT Delhi), New Delhi, India.
This study and the development of DiPCell were supported by the Council of Scientific and Industrial Research (CSIR) and the Department of Biotechnology (DBT), Government of India.