eLife Assessment
The authors present a set of wrappers around previously developed software and machine-learning toolkits, and demonstrate their use in identifying endogenous sterols binding to a GPCR. The resulting pipeline is potentially useful for molecular pharmacology researchers due to its accessibility and ease of use. However, the evidence supporting the GPCR-related findings remains incomplete, as the machine-learning model shows indications of overfitting, and no direct ligand-binding assays are provided for validation.