The manuscript entitled "Advancements in Drug Discovery: Integrating CADD Tools and Drug Repurposing for PD-1/PD-L1 Axis Inhibition" is an all-encompassing research study related to strategies for identification of novel PD-1/PD-L1 Axis Inhibitors. The author worked on QSAR models to predict PD-1/PD-L1 inhibition, utilizing a diverse dataset of 29,197 molecules sourced from ChEMBL, PubChem, and recent literature. Machine learning techniques, including Random Forest, Support Vector Machine, and Convolutional Neural Network, were employed for benchmarking to assess model performance. Also used a drug repurposing strategy, leveraging the best in silico model for a virtual screening campaign involving 1,576 off-patent approved drugs. One of the proposed hits was sonidegib, an anticancer drug, featuring a biphenyl system. Sonidegib was subsequently validated for in vitro PD-1/PD-L1 binding modulation using ELISA and flow cytometry.
The study has been executed well. The findings could be significant to the scientific community and thus the manuscript can be accepted for publication.
Just a minor observation: The decimal points in each table to be reduced to only three significant places., e.g. 0.000.