APPLE server is capable of predicting a specific or pan-inhibitor of Bcl-2 and Bcl-xL proteins.Overexpression of Bcl-2 and Bcl-xL was reported in various cancers which is responsible for evading apoptosis.Targeting Bcl-2 anti-apoptotic proteins with high specificity is the need of the hour to combat cancer in which the existing tactics fail. We attempted to address the pitfalls of existing tactics by analyzing the descriptors accounting for activity and predicting the biological activity of user compounds.
We have considered a dataset of 1649 small molecule inhibitors of Bcl-2 and Bcl-xL. In silico models were generated for seven scaffolds namely Apogossypol, Quinazoline thione, Pyrazole, Benzothiohydrazine, Quinolone, Poly Quinoline, Thiomorpholin.The models showed a pearson correlation coefficient(PCC) of 0.97 and mean absolute error(MAE) of 0.75 µM which reflects the robustness of the models.
|IITM||Protein Bioinformatics Lab|