查看文章

The training dataset used for model development consisted of 102 drug release profiles of 13 drugs. Some of this data has been previously collected by the Allen lab, and all data has been previously published. Data taken exclusively from the literature was obtained through a “Web of Science” database search using the keyword combinations of “polymeric microparticle” and “drug delivery”. Data was extracted from published release profiles using the “GetData Graph Digitizer” application. The external validation dataset used for the pseudo-prospective study consists of 79 release profiles collected from the literature in a similar way. Both of these datasets are available as supplementary information files and also online at the Aspuru-Guzik Group’s GitHub page (https://github. com/aspuru-guzik-group/to-be-released-upon-acceptance).