Towards Generalizable Predictions for the Effects of Mutations on G-Protein Coupled Receptor Expression

Missense mutations that compromise the plasma membrane expression (PME) of integral membrane proteins (MPs) are the root cause of numerous genetic diseases. Differentiation of this class of mutations from those that specifically modify the activity of the folded protein has proven useful for the development and targeting of precision therapeutics. Nevertheless, it remains challenging to predict the effects of mutations on the stability and/ or expression of MPs. In this work, we utilize deep mutational scanning data to train a series of artificial neural networks to predict the effects of mutations on the PME of the G-protein coupled receptor (GPCR) rhodopsin from structural and/ or evolutionary features. We show that our best performing network, which we term PMEpred, can differentiate pathogenic rhodopsin variants that induce misfolding from those that primarily compromise signaling. This network also generates statistically significant predictions for the effects of mutations on the PME of another GPCR (Beta-2 adrenergic receptor) but not for an unrelated voltage-gated potassium channel (KCNQ1). Notably, our analyses of these networks suggest structural features alone are generally sufficient to recapitulate the observed mutagenic trends. Moreover, our findings imply that networks trained in this manner may be generalizable to proteins that share a common fold. Implications of our findings for the design of mechanistically specific genetic predictors are discussed.