STAT3In: Computer-Aided Prediction of Inhibitors Against STAT3 for Managing COVID-19 Associated Cytokine Storm

No description provided. STAT3In: Computer-Aided Prediction of Inhibitors Against STAT3 for Managing COVID-19 Associated Cytokine Storm STAT3In is a computational web server developed for the prediction and design of STAT3 inhibitors. STAT3 is an important transcription factor involved in inflammation, cancer progression, angiogenesis, and cytokine storm. Since IL6-mediated STAT3 activation is associated with COVID-19-related cytokine storm, STAT3In was developed to identify small chemical molecules that may inhibit the STAT3 signaling pathway. Web Server: https://webs.iiitd.edu.in/raghava/stat3in/ Citation Dhall, A., Patiyal, S., Sharma, N., Devi, N. L., and Raghava, G. P. S. Computer-aided prediction of inhibitors against STAT3 for managing COVID-19 associated cytokine storm. Computers in Biology and Medicine, 137, 104780, 2021. https://doi.org/10.1016/j.compbiomed.2021.104780 About the Research Proinflammatory cytokines are strongly associated with disease severity in COVID-19 patients. IL6-mediated activation of STAT3 can promote inflammatory signaling and contribute to cytokine storm. STAT3 is a cytoplasmic transcription factor involved in several biological processes such as cell proliferation, differentiation, angiogenesis, inflammation, and apoptosis. However, abnormal activation of STAT3 is associated with cancer, pulmonary fibrosis, acute lung injury, and COVID-19-related cytokine storm. STAT3In was developed to predict small chemical molecules that may act as STAT3 inhibitors and help in identifying potential candidates for managing IL6/STAT3-mediated inflammatory responses. Data Compilation: STAT3 inhibitors and non-inhibitors were collected from PubChem BioAssay AID 862. The final dataset contained 1565 STAT3 inhibitors and 1671 non-inhibitors. Methodology: STAT3In uses machine learning models trained on chemical descriptors and molecular fingerprints. The descriptors were calculated using PaDEL software, and models were developed using classifiers such as Random Forest, Decision Tree, Logistic Regression, Support Vector Classifier, Gaussian Naive Bayes, K-Nearest Neighbour, and eXtreme Gradient Boosting.