Ligand-Based AI-Assisted Drug Discovery for Lung Cancer Using Swiss-Similarity and In-Silico Validation

Abstract

Lung cancer still ranks among the deadliest cancers everywhere, pushing demand for treatments that work better and reach more people. Old ways of finding new medicines usually take too long, cost a lot, yet frequently fail, so scientists now lean heavier on computer methods to speed up initial phases. Here, researchers turned to digital-only, molecule-focused techniques powered by AI tools to track down tiny compounds that might fight lung tumors. Instead of lab tests, they ran simulations using Swiss-Similarity, which scanned vast collections of chemicals by matching their traits to existing lung cancer medications acting as blueprints. Later on, researchers looked at chemicals that resemble each other by using computer models to see how well they stick to proteins linked to lung cancer. Instead of moving forward blindly, they ran detailed checks on absorption, distribution, metabolism, and toxicity to spot red flags early. Some molecules held onto targets just as tightly - sometimes even more so - than current medications, yet still appeared safe based on forecasts. From start to finish, every step happened inside a machine, showing what digital methods can do though real-world testing remains essential down the line.