Importantly, three different contexts of bioactivity similarity were considered.
The use of the three contexts of bioactivity similarity allows for organizing bioactivity data of molecules tested in a single assay, as well as those scattered across multiple assays that are targeting the same protein or pathway.
Although any database can have unique identifiers (UIDs) to organize its records, the term "UID" is specifically reserved in the present study for any of AID, GI, and BSID (depending on the context) to represent the three contexts of bioactivity similarity, but not for CID (the unique identifier used in the PubChem Compound database).
The resulting clusters can be broadly classified into three types, according to the context of the bioactivity similarity considered.
Overall, the application of ensemble modeling with a model library trained with either the same algorithm but different parameter values (homo-ensemble), or with different algorithms (hetero-ensemble) constitutes a promising alternative to single models in the context of predictive bioactivity modeling.
Pre-computation of all 3-D similarity scores and the clustering of fifteen different contexts of structural and bioactivity similarities for all PubChem data can take months on hundreds of CPU cores; however, if the biological activity data is largely static, most of this computation is a one-time cost and the results need only be stored.
For a recent and detailed discussion of the application of the student t-test to assess the statistical significance of bioactivity differences in the context of Matched Molecular Pair Analysis (MMPA), the reader is referred to Kramer et al. [89].
Sur, S., Matson, J. B., Webber, M. J., Newcomb, C. J. & Stupp, S. I. Photodynamic control of bioactivity in a nanofiber matrix.
Boekhoven, J., Rubert Pérez, C. M., Sur, S., Worthy, A. & Stupp, S. I. Dynamic display of bioactivity through host-guest chemistry.
Titanium possesses good mechanical property, but a lack of bioactivity.
Our platform addresses the challenge of controlling the dynamics of bioactivity on synthetic scaffolds, an essential feature of extracellular matrix biology.
