Biology subjects

Felbinger, N.

Publications and source records attributed to Felbinger, N..

4 recordsLinked to original sources

bioRxiv2026

Activation pathway of the human voltage-gated proton channel

Many protein functions are coupled to proton transfer, yet the underlying mechanisms remain extremely difficult to resolve by experiment or conventional molecular dynamics (MD) simulations. The human voltage-gated proton channel HV1 (hHV1) plays key roles in human physiology, but how voltage and pH together drive its activation remains unresolved. Experimental structures of hHV1 are limited to the X-ray structure of a mouse chimera and NMR models of a truncated hHV1, both representing an intermediate resting state. Here we determine the activation pathway of hHV1, together with structural models of its resting and activated states, using an approach that integrates all-atom continuous constant-pH MD (CpHMD), weighted-ensemble (WE) path sampling, and AlphaFold modeling. Starting from an AlphaFold model selected for consistency with resting-state experimental data, over 100 s of simulations reveal a mechanism in which transmembrane voltage, assisted by proton titration, drives an outward displacement of the transmembrane segment S4 accompanied by a one-click reorganization of countercharge pairings. These events are enabled by a delicately balanced electrostatic landscape and pH-modulated breathing motions in the resting state that prime the channel for protonation-state switches upon voltage sensing, a principle that may underlie other proton-coupled voltage-gating processes. The simulations also uncover lipid-binding sites at which POPG selectively stabilizes the S4-up conformation, whereas POPE can stabilize either state. These predictions are validated against extensive electrophysiology, mutagenesis data, and new smFRET experiments, which support the one-click S4 displacement and the lipid effects. WE-CpHMD, implemented in the AMBER package, is applicable to other proton-coupled processes that remain elusive.

biophysics↗

bioRxiv2026

Benchmarking Deep Learning Predictions of Mutation-Induced Fold Switching

Many proteins are known to adopt multiple distinct folded states which are often associated with key functional behavior. A predictive understanding of the properties of such fold-switching or metamorphic proteins can provide insights into protein dynamics and energetics, and enable the design of complex protein functions and molecular machines. Recently developed deep learning modeling tools, including AlphaFold, have led to dramatic increases in accuracy for prediction of protein structures from sequence, but their performance for prediction of point mutant effects or fold switching is unclear. Here we present a systematic NMR-characterized dataset of mutants of the GA/GB model fold-switching system and use it to evaluate whether current structure prediction and design methods can predict mutation-induced changes in fold state. We measured fold-state populations for variants at three key positions that differentially stabilize the 3, 4{beta}+, mixed, or unfolded states, generating a quantitative experimental benchmark for mutation-level fold switching. Using this benchmark to assess and compare a panel of deep learning and physics-based modeling and design algorithms, we found that this benchmark revealed variable and position-dependent performance across methods, with certain AlphaFold2-based algorithms were able to predict mutant effects at individual sites, indicating some understanding of physical effects of residue substitutions. Additional comparisons of predictions with experimentally measured stability changes further highlighted position-dependent success and general challenges for predictive algorithms. Together, this study provides a new benchmark for mutation-induced fold switching and reveals the current capabilities and limitations of deep learning models for predicting mutation-dependent protein conformational states.

biophysics↗

bioRxiv2026

Benchmarking AlphaFold and related deep learning approaches for modeling antibody and TCR antigen recognition

Determining the structural basis of antigen recognition by antibodies and T cell receptors (TCRs) provides critical insights into effective immune targeting and can inform design of biotherapeutics and vaccines. Accurate computational modeling of antibodies and TCRs in complex with their targets poses a major challenge for predictive methods, including AlphaFold, which is generally accurate for modeling protein complexes but has shown limited success for immune recognition. In this study we assessed the performance of AlphaFold2, AlphaFold3, increased sampling protocols, and related deep learning methods for modeling antibody-protein, antibody-peptide, and TCR-peptide-major histocompatibility complex (pMHC) recognition. We show that increased sampling and AlphaFold3 generally improve performance relative to default sampling and AlphaFold2, however predictive accuracy and improvement levels varied considerably among interface classes, with antibody-peptide complexes representing a challenge despite their small antigen size. Comparing per-case success across methods showed some complementarity, indicating opportunities for increased success through model pooling approaches, for instance increasing antibody-peptide near-native success from 41% to 59%. Analysis of AlphaFold confidence scores and modeling of a noncanonical complex provided further insights into predictive performance. These results highlight considerations for predictive antibody and TCR complex modeling efforts, while revealing key distinctions among protocols, scoring, and immune complex classes.

bioinformatics↗

bioRxiv2022

Epitope Mapping of SARS-CoV-2 Spike Protein Reveals Distinguishable Antibody Binding Activity of Vaccinated and Infected Individuals.

Previous studies have attempted to characterize the antibody response of individuals to the SARS-CoV-2 virus on a linear peptide level by utilizing peptide microarrays. These studies have helped to identify epitopes that have potential to be used for diagnostic tests to identify infected individuals, however, the immunological responses of individuals who have received the currently available Moderna mRNA-1273 or Pfizer BNT162b2 mRNA vaccines have not been characterized. We aimed to identify linear peptides of the SARS-CoV-2 spike protein that elicited high IgG or IgA binding activity and to compare the immunoreactivity of infected individuals to those who received both doses of either vaccines by utilizing peptide microarrays. Our results revealed peptide epitopes of significant IgG binding among recently infected individuals. Some of these peptides are located near functional domains implicated in the high infectivity of SARS-CoV-2. Vaccinated individuals lacked these distinct markers despite overall binding activity being similar.

immunology↗