Publications by authors named "Anup Sarma"
Article Synopsis
- Modern drug discovery is costly and slow, with current computational methods being less accurate and high-latency due to reliance on traditional docking software.
- Recent machine learning methods improve protein-ligand binding affinity evaluation but still depend on conventional approaches for pose sampling, leading to long execution times.
- The new framework, MedusaGraph, uses a graph neural network to directly generate docking poses, offering a 10 to 100 times speed increase and slightly better accuracy than existing methods.
The high-performance computational techniques have brought significant benefits for drug discovery efforts in recent decades. One of the most challenging problems in drug discovery is the protein-ligand binding pose prediction. To predict the most stable structure of the complex, the performance of conventional structure-based molecular docking methods heavily depends on the accuracy of scoring or energy functions (as an approximation of affinity) for each pose of the protein-ligand docking complex to effectively guide the search in an exponentially large solution space.
View Article and Find Full Text PDF