RosettaDDGPrediction for high-throughput mutational scans: from stability to binding

Abstract

Reliable prediction of free energy changes upon amino acidic substitutions (ΔΔGs) is crucial to investigate their impact on protein stability and protein-protein interaction. Moreover, advances in experimental mutational scans allow high-throughput studies thanks to sophisticated multiplex techniques. On the other hand, genomics initiatives provide a large amount of data on disease-related variants that can benefit from analyses with structure-based methods. Therefore, the computational field should keep the same pace and provide new tools for fast and accurate high-throughput calculations of ΔΔGs. In this context, the Rosetta modeling suite implements effective approaches to predict the change in the folding free energy in a protein monomer upon amino acid substitutions and calculate the changes in binding free energy in protein complexes. Their application can be challenging to users without extensive experience with Rosetta. Furthermore, Rosetta protocols for ΔΔG prediction are designed considering one variant at a time, making the setup of high-throughput screenings cumbersome. For these reasons, we devised RosettaDDGPrediction, a customizable Python wrapper designed to run free energy calculations on a set of amino acid substitutions using Rosetta protocols with little intervention from the user. RosettaDDGPrediction assists with checking whether the runs are completed successfully aggregates raw data for multiple variants, and generates publication-ready graphics. We showed the potential of the tool in selected case studies, including variants of unknown significance found in children who developed cancer, proteins with known experimental unfolding ΔΔGs values, interactions between target proteins and a disordered functional motif, and phospho-mimetic variants. RosettaDDGPrediction is available, free of charge and under GNU General Public License v3.0, at https://github.com/ELELAB/RosettaDDGPrediction.

MAVISp: A Modular Structure-Based Framework for Protein Variant Effects

“…In this step, the effect of the variants on the structural stability of the proteins is investigated using folding free energy calculations with FoldX5 suite and the foldx5 energy function 37 , the cartddg2020 protocol, and the ref2015 energy function of Rosettaddg 38,39 . These steps take advantage of the recently developed workflows for high-throughput in-silico mutagenesis scans, i.e., MutateX 40 and RosettaDDGPrediction 41 . MAVISp applies a consensus approach between the two methods.…”

“…In addition, we support variant lists from other studies, such as data on cohort-based or nationwide studies or other cancer genomic initiatives. For example, we recently analyzed variants found with low frequency in the healthy population on germline samples of childhood cancer patients 54 using one of the tools applied in the STABILITY module 41 . The assessment for MLH1 and BLM (see below) includes some of these variants, which have been reported in COSMIC, cBioPortal or ClinVar.…”

“…We identified 50 mutations for one representative of this class of enzymes (PPIA, also known as CypA), respectively. The methods used in the STABILITY module of MAVISp have been evaluated for their accuracy against experimentally determined unfolding free energy changes in FKB1A in one of our previous studies 41 .…”

“…We identified 50 mutations for one representative of this class of enzymes (PPIA, also known as CypA), respectively. The methods used in the STABILITY module of MAVISp have been evaluated for their accuracy against experimentally determined unfolding free energy changes in FKB1A in one of our previous studies 41 . Interestingly, we found only three destabilizing variants in terms of change of folding free energy for PPIA (S77F, G75V, and G12V) and a group of possibly allosteric mutations, of which P4L, S21F, V29I, and H92Y cause a longrange decrease of flexibility in the area where N129 is, which is in proximity of the catalytic site.…”

TRAP1S-nitrosylation as a model of population-shift mechanism to study the effects of nitric oxide on redox-sensitive oncoproteins

“…For TP53, we used Cancermuts to include also mutations from other sources 56 . We then applied the STABILITY module of MAVISp to classify the effects of the mutations in unknown, destabilizing, stabilizing, or neutral according to changes in the folding free energy as estimated with the MutateX protocol 96 with Foldx5 97 , as well as with RosettaDDGPrediction 98 with the cartddg2020 protocol 99 and the ref2015 energy function 100 . We retrieved the classification for the variants of interest as benign, pathogenic, or variants of unknown significance from ClinVar 58 .…”