DesertSci has developed a new High Throughput Screening (HTS) analysis tool able to handle very large datasets with the accuracy of MOS methods. We have called this software, BMOS (Beyond Maximum Overlapping Set). With BMOS you can achieve a 10 fold increase on traditional MOS clustering numbers, with negligible loss of accuracy, and reasonable job times.
BMOS uses graph-based chemical similarity methods to cluster large datasets from HTS campaigns. It combines a fast Maximum Common Substructure (MCS) approach and our fine-grained Maximum Overlapping Set (MOS/Spinifex) approach. BMOS produces results comparable in quality to Spinifex, for datasets an order of magnitude larger, in the same amount of time.
The extended functionality within Spinifex – incorporating 3D conformation data and identifying cluster centroids at different levels of the similarity cutoff – is also available in BMOS.
BMOS can identify important relationships in structural data at much lower similarity levels, results that were previously unattainable with large data sets. Being able to identify important structural activity relationships that may impact the design of synthetic candidates, in seemingly dissimilar molecules, can have a significant impact on the drug discovery process.
BMOS is designed to run in a high performance computing (HPC) environment using SLURM (https://slurm.schedmd.com/ ).
For more information or to organise your obligation free evaluation please contact us at firstname.lastname@example.org