Quantitative Structure-Activity Relationship of Fluorescent Probes and Their Intracellular Localizations

The development of organelle-specific fluorescent probes has been impeded by the absence a comprehensive understanding relationship between physicochemical properties and their selectivity towards specific organelles. Although few machine learning models have suggested several parameters that control target organelle attempted to predict organelles, they challenged low accuracy limited range applicable Herein, we report multi-organelle prediction QSAR model is capable predicting destination among nine categories, including cytosol, endoplasmic reticulum, Golgi body, lipid droplet, lysosome, mitochondria, nucleus, plasma membrane, no entry. trained using Random Forest algorithm with dataset 350 786 descriptors, it able organelles an 75%. MDI analysis identifies 38 key significant impact on probes, LogD, pKa, hydrophilic-lipophilic balance (HLB), topological polar surface area (TPSA). This may be useful in developing new providing crucial variables determine probes.