How to speed up the preparation of data sets for Structure-Based Drug Design? – Webinar Recap
The process of Structure-Based Drug Design (SBDD) refers to the systematic use of 3D structural data. The crucial starting point, in most cases, is the preparation of curated structural data sets for the subsequent creation of predictive models. For instance, 3D-QSAR models, structure-based (as well as ligand-based) 3D pharmacophore models, predictive ML / AI and ensemble docking models are just some of the most common. For each of them, a specific set of 3D structures and a specific set of 3D information is required. Therefore, if we want to trust our predictions, we need to make sure that the collected data sets are accurate, consistent, and complete.
Some of the issues when working on the construction of the sets are due to the lack of tools for automation. Selecting structures manually might take a lot of time and is often inefficient. Moreover, the landscape of available data sources is increasing at high-speed. Therefore, how can we be sure that our list of structures for the specific target is complete? Depending on the needs, other questions might arise as well - How can I quickly retrieve all the ligands binding to a binding site of interest? How can I collect protein structures of a certain conformation and ligands binding to it? How can I spot all relevant mutations in a glimpse? - and the list goes on.
3decision® can provide proper data organization and storage to facilitate the preparation of accurate data sets, and thus reduce the timelines of drug discovery projects. In our latest webinar, we explained how to overcome the challenges of data sets preparation in SBDD. For a better understanding of how our tool helps, we presented 3 common use cases - have a look at the webinar replay here.