The judicious design of libraries that can be synthesised rapidly in our High Throughput Chemistry lab can accelerate SAR exploration at various stages of the discovery process and help find breakthrough structures in lead optimisation.
- We apply in silico methods to library design of focused arrays of compounds for parallel synthesis. Curated sets of available reagents are used to enumerate compounds accessible through the current chemistry and then clustering and filtering (based on desired physicochemical properties, docking score or pharmacophore fit) are applied to prioritize compounds for synthesis.
In addition, our use of multiple computational techniques to carefully curate and optimize virtual and physical libraries of compounds can provide you with the best starting point for in vitro or in silico screening depending on your project’s needs and increase your chances of finding new hits.
- We curate an 8 million-compound in silico library for high-throughput virtual screening derived from databases of commercially available compounds by the application of physicochemical property filters and the removal of compounds with undesired functionalities.
- When a more focused library design approach is desired, we can design project-specific targeted libraries for in vitro or in silico screening, enriched with chemotypes potentially active against specific target classes or with compounds bearing desired properties and functionalities.
- From Sygnature’s significant contribution to the European Lead Factory (ELF) initiative we have generated a library of novel cores synthetized at Sygnature that offer unique and synthetically tractable starting points for virtual screening or scaffold hopping approaches.
- Physicochemical property filters, fingerprint clustering and diversity analysis are used to curate and update our proprietary fragment library to increase its quality for fragment-based drug discovery.
- Computational chemistry can also support the workup of in vitro screening results during hit finding, with clustering, profiling and close neighbour search, helping to identify the most promising series to progress.