Abstract
Radiopharmaceuticals have emerged as essential tools in modern medicine, playing a pivotal role in disease diagnosis and treatment. This paper explores the application of Quantitative Structure-Property Relationships (QSPR) for predicting logP of a dataset comprising 121 molecules. Utilizing a hybrid optimal descriptor that combines SMILES and the hydrogen suppressed graph, QSPR models are constructed. The QSPR model established in splits 2, 6, and 8, demonstrating the highest R2 values for the validation set (0.8859, 0.8816, and 0.8144, respectively), is identified as the leading model. Moreover, these leading models provide insights into the promoters influencing the increase and decrease of logP.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
