Indexing and mining topological patterns for drug discovery

摘要: Increased availability of large repositories chemical compounds has created new challenges and opportunities for the application data-mining indexing techniques to problems in informatics. The primary goal analysis molecular databases is identify structural patterns that can predict biological activity. Two most popular approaches representing topologies are graphs 3D geometries. As a result, problem mining map from graph geometric databases.In this tutorial, we will first introduce drug discovery how computer science plays critical role process. We then proceed by introducing performing subgraph similarity searches on databases. Due NP-hardness problems, number heuristics have been designed recent years tutorial present an overview those techniques. Next, frequent along with some their limitations ignited interest statistically significant patterns. After presenting in-depth survey patterns, towards analyzing structures molecules. Finally, conclude two open impact field discovery.