Fig. 5: Predicting protein stability change upon mutations.
From: Deciphering protein evolution and fitness landscapes with latent space models
a, b Correlation between experimental stability change and VAE free energy change upon single-site mutations for fibronectin type III domain (a) and staphylococcal nuclease (b). \(\Delta \Delta {G}_{\exp }\) is experimental protein folding free energy change upon single-site mutations compared with the wild type protein. \(\Delta \Delta {G}_{{\rm{VAE}}}\) is VAE free energy change upon single-site mutations. \(\Delta \Delta {G}_{{\rm{VAE}}}\) is calculated as the change of negative log-likelihood of sequences when single-site mutations are introduced. Therefore, \(\Delta \Delta {G}_{{\rm{VAE}}}\) is an unitless quantity. Each point corresponds to a mutant sequence with one mutation compared with the wild-type sequence. \(r\) and \(\rho\) are Pearson’s correlation coefficients and Spearman’s rank correlation coefficients, respectively. c, d In addition to latent models trained with VAEs, similar analysis is conducted using sequence profile and DCA methods. Spearman’s rank correlation coefficients between experimental protein folding free energy change upon single-site mutations and free energy change calculated using the three methods are compared for the same two protein families: fibronectin type III domain (c) and staphylococcal nuclease (d).
