Biology subjects

Schoeder, C. T.

Publications and source records attributed to Schoeder, C. T..

9 recordsLinked to original sources

bioRxiv2026

Latent generative search unlocks de novo design of untapped biomolecular interactions at scale

De novo protein design has advanced rapidly, yet designing binders to polar, solvent-exposed epitopes and small, flexible ligands remains challenging. Such hydrated surfaces and flexible molecules, including carbohydrates, provide few of the hydrophobic contacts favoured by current methods and have largely resisted de novo binders. To address this challenge, here we introduce latent generative search for binder design, a novel framework that uses reward-guided search at inference time to steer the Proteina-Complexa generative model. The model codesigns sequence and structure - generating them together in a continuous latent space - and thereby removes the inverse-folding step on which current methods rely. In a screen of more than one million designs by multiplexed phage display, latent generative search produced more validated binders than every other method tested, its codesigned sequences surpassing post hoc redesign. It delivered high-affinity binders across therapeutic receptors, a viral attachment protein and intracellular signalling targets. Our approach also accessed previously untapped biology, generating the first de novo proteins that bind a free carbohydrate, including one that discriminates between blood-group antigens - a polar, flexible target class beyond the reach of current design methods.

bioengineering↗

bioRxiv2026

A generalizable interface-seeded framework for de novo design of functional oligomers

Protein oligomers are ubiquitous in biological systems and essential for function. However, the de novo design of oligomers that controllably assemble in response to exogenous stimuli remains challenging. Here, we present an AI-based generative approach that leverages an interface-seeded strategy for designing responsive homo-oligomers from isolated interaction modules. Experimentally validated designs are highly accurate and explore new-to-nature topologies. We show that designs effectively respond to their chemical triggers with conditional oligomerization or to phosphorylation-driven conformational changes with reversible oligomerization. We further functionalized our responsive assemblies to build ligand-dependent membrane binding systems and phosphorylation-controlled gene regulatory switches. Our framework enables the generalizable design of responsive protein complexes, opening novel possibilities for the engineering of biosynthetic systems with sophisticated regulatory mechanisms.

biochemistry↗

bioRxiv2026

Assessment of Generative De Novo Peptide Design Methods for G Protein-Coupled Receptors

G protein-coupled receptors (GPCRs) play an ubiquitous role in the transduction of extracellular stimuli into intracellular responses and therefore represent a major target for the development of novel peptide-based therapeutics. In fact, approximately 30% of all non-sensory GPCRs are peptide-targeted, representing a blueprint for the design of de novo peptides, both as pharmacological tools and therapeutics. The recent advances of deep learning-based protein structure generation and structure prediction offer a multitude of peptide design stategies for GPCRs, yet confidence metrics rarely correlate with experimental success. In the context of peptides, this problem is exacerbated due to the lack of elaborate tertiary structures in peptides, raising the question of whether this is due to inadequate sampling or insufficient scoring. In this two-part benchmark, we addressed this question by first simulating the validation process of 124 unique known GPCR-peptide complexes using AlphaFold2 Initial Guess, Boltz-2 and RosettaFold3. We then assessed the peptide sampling capabilities of the respective generative methods BindCraft, BoltzGen and RFdiffusion3. Our results indicate that current design pipelines primarily suffer from significant confidence overestimation for misplaced peptides in the validation phase across all three prediction methods. We further highlight occurrences of significant memorization in both prediction as well as generation of peptides. While all generative methods sample backbone space sufficiently, their simultaneous sequence generation remains subpar and can be partially recovered through the use of ProteinMPNN. Taken together, our benchmark offers guidance for the design of peptides specifically using deep learning-based pipelines. Autor summaryDeep learning-based protein design is revolutionizing computational biology and development of such tools is progressing rapidly with increasing attention from both academic and non-academic institutions. Their applicability and performance is often assessed from an all-purpose objective, with implicit bias towards larger protein-protein interactions. Due to their size, peptides therefore present an edge case where performance is known to decrease compared to larger, more structured proteins. Here, we present a benchmark specifically for the deep learning-based design of peptides targeting G protein-coupled receptors (GPCRs), a major therapeutic drug target family, assessing the generation of novel GPCR-targeting peptides and the validation of these designs separately. Our results show that generative methods sample potential peptide placements and orientations sufficiently but validation fails to differentiate valid from invalid designs, indicating that the so-called scoring problem remains unsolved. Although focusing on a specific use-case, our results are generalizable to the broader field of protein design. Consequently, it can offer guidance for peptide-specific design applications and can contribute to the development and improvement of new methods.

bioinformatics↗

bioRxiv2026

FrustraMPNN: An ultra-fast deep learning tool for proteome-scale analysis of deep mutational single-residue local energetic frustration in proteins

Energetic frustration, characterized by conflicting local interactions, is a key determinant of protein dynamics, allostery, protein-protein interactions, enzyme catalysis, and overall protein function. Although the principle of minimal frustration describes an evolutionary bias towards a reduction of energetic conflicts for efficient protein folding, local violations are selectively embedded to encode the complex energy landscapes necessary for protein functions. However, the computational cost of traditional single-residue frustration analysis has prevented the calculation of these energetic conflicts at the proteome scale, creating a significant bottleneck in structural biology. Here, we introduce FrustraMPNN, a message-passing neural network retrained via transfer learning that predicts a complete per-residue frustration-index mutation profile (full mutational scanning per position for a whole protein) orders-of-magnitude faster than existing methods while maintaining high accuracy. This is demonstrated by calculating the single-residue frustration of the E. coli proteome, reducing the calculation time from years to 12 hours. We validate FrustraMPNN on a diverse external set of over 3,400 human protein structures, achieving a Spearman correlation of up to 0.80 across all frustration categories, demonstrating robust generalizability. By conducting a thorough dataset ablation study, we find that the models performance improves when trained on a broader range of protein sizes, highlighting an inherent limitation of datasets like Megascale. We provide FrustraMPNN as an open-source tool, expecting it will enable exploration of new hypotheses in fields such as personalized structural biology and allow analysis of local energetic frustration patterns involved in protein function at an unprecedented proteomic scale.

bioinformatics↗

bioRxiv2026

A novel dual-chimeric surrogate virus to select, test and characterize mutants of protease and glycoprotein inhibitors.

Antivirals are life-saving medications. However, pathogens develop resistances, mitigating their effect. Resistance studies (gain-of-function, GOF) are therefore paramount before and while antivirals are used clinically to characterize resistances and ideally study how to counteract them with other antivirals. The ability to switch from one antiviral, against which the virus is resistant, to another, depends on the availability of alternative drugs. Furthermore, targeting different parts of the virus synergistically with multiple drugs decreases the odds of simultaneous resistance. However, studying resistance, cross-resistance and remaining susceptibility requires working with the actual pathogen. Especially when introducing resistances into dangerous viruses, performing GOF studies would raise safety concerns and require firm biological containment. In this study we developed a novel dual-chimeric virus encoding both SARS-CoV-2 spike glycoprotein and main protease (Mpro) based on the model virus, vesicular stomatitis virus (VSV). We applied selection pressure with spike and Mpro inhibitors, generated mutants of these viruses and determined their resistance and cross-resistance profiles by performing dose-response experiments with live and pseudotyped viruses. Additionally, we conducted in silico calculations to determine the impact of spike substitutions to the observed resistance phenotype. With the new technology presented here, we demonstrate that it is possible to carry out GOF studies more safely and highlight their importance in the future challenges of virological research.

microbiology↗

bioRxiv2025

Computational design and experimental characterization of mini-protein binders targeting Nipah, Langya and Measles virus receptor-binding domains

A lack of reagents represents a major bottleneck in pandemic preparedness and rapid vaccine development. It is therefore important to enable the design of reagents for use in the treatment and diagnosis of emerging viral diseases. Ideally, the design and identification platform is fast, can be performed by testing only a small number of candidates and enables a generally applicable strategy. In this study, we assessed the ability of recently developed computational protein design tools to establish such a workflow for validating paramyxovirus receptor-binding protein de novo binders as such reagents. The family Paramyxoviridae includes various members that cause severe disease and exhibit re-occurring zoonotic spillover events, with documented human infections over the past decades. We successfully designed, identified, and characterized mini-proteins targeting the receptor binding proteins of Nipah virus, Langya virus, and Measles virus while screening as few as 10-16 designs per target. The resulting functional binders have moderate to low nanomolar affinities and display high on-target specificity. We further showed that our most promising Nipah virus receptor-binding protein binder is able to inhibit human receptor binding in vitro and competes for an epitope that overlaps with that of the neutralizing antibody HENV-117. However, despite these promising results, this Nipah binder is only weakly neutralizing, preventing therapeutic applications. Nevertheless, we established a platform, applicable to rapidly generate diagnostically relevant proteins from only a small number of candidates, and developed novel reagents for the Paramyxoviridae family.

bioinformatics↗

bioRxiv2024

HyperMPNN - A general strategy to design thermostable proteins learned from hyperthermophiles

Stability is a key factor to enable the use of recombinant proteins in therapeutic or biotechnological applications. Deep learning protein design approaches like ProteinMPNN have shown strong performance both in creating novel proteins or stabilizing existing ones. However, it is unlikely that the stability of the designs will significantly exceed that of the natural proteins in the training set, which are biophysically only marginally stable. Therefore, we collected predicted protein structures from hyperthermophiles, which differ substantially in their amino acid composition from mesophiles. Notably, ProteinMPNN fails to recover their unique amino acid composition. Here we show that a retrained network on predicted proteins from hyperthermophiles, termed HyperMPNN, not only recovers this unique amino acid composition but can also be applied to proteins from non-hyperthermophiles. Using this novel approach on a protein nanoparticle with a melting temperature of 65{degrees}C resulted in designs remaining stable at 95{degrees}C. In conclusion, we created a new way to design highly thermostable proteins through self-supervised learning on data from hyperthermophiles.

bioinformatics↗

bioRxiv2024

Self-supervised machine learning methods for protein design improve sampling, but not the identification of high-fitness variants

Machine learning (ML) is changing the world of computational protein design, with data- driven methods surpassing biophysical-based methods in experimental success rates. However, they are most often reported as case studies, lack integration and standardization across platforms, and are therefore hard to objectively compare. In this study, we established a streamlined and diverse toolbox for methods that predict amino acid probabilities inside the Rosetta software framework that allows for the side-by-side comparison of these models. Subsequently, existing protein fitness landscapes were used to benchmark novel self- supervised machine learning methods in realistic protein design settings. We focused on the traditional problems of protein sequence design: sampling and scoring. A major finding of our study is that novel ML approaches are better at purging the sampling space from deleterious mutations. Nevertheless, scoring resulting mutations without model fine-tuning showed no clear improvement over scoring with Rosetta. This study fills an important gap in the field and allows for the first time a comprehensive head-to-head comparison of different ML and biophysical methods. We conclude that ML currently acts as a complement to, rather than a replacement for, biophysical methods in protein design.

bioinformatics↗

bioRxiv2021

Structural comparative modeling of multi-domain ΔF508 CFTR

Cystic Fibrosis (CF) is a common genetic disease caused by mutations in the Cystic Fibrosis Transmembrane Conductance Regulator (CFTR), an epithelial anion channel expressed in several vital organs. Absence of functional CFTR results in imbalanced osmotic equilibrium and subsequent mucus build up in the lungs - which increases the risk of infection and eventually causes death. CFTR is an ATP binding cassette (ABC) transporter composed of two transmembrane domains (TMDs), two nucleotide binding domains (NBDs), and an unstructured regulatory domain. The most prevalent patient mutation is the deletion of F508 ({Delta}F508), making {Delta}F508 CFTR the primary target for current FDA approved CF therapies. However, no experimental multi-domain {Delta}F508 CFTR structure has been determined and few studies have modeled {Delta}F508 using multi-domain WT CFTR structures. Here, we used cryo-EM density data and Rosetta comparative modeling (RosettaCM) to compare a {Delta}F508 model with published experimental data on CFTR NBD1 thermodynamics. We then apply this modeling method to generate multi-domain WT and {Delta}F508 CFTR structural models. These models demonstrate the destabilizing effects of {Delta}F508 on NBD1 and the NBD1/TMD interface in both the closed and open conformation of CFTR. Furthermore, we modeled {Delta}F508/R1070W and {Delta}F508 bound to the CFTR corrector VX-809. Our models reveal the stabilizing effects of R1070W and VX-809 on multi-domain models of {Delta}F508 CFTR and pave the way for rational design of additional drugs that target {Delta}F508 CFTR for treatment of CF. Author SummaryProteins three-dimension shape determines their function, so when genetic mutation compromises the shape of vital proteins, it may cause disease. Such is the case in Cystic Fibrosis, a chronic genetic disease caused by mutations in the protein Cystic Fibrosis Transmembrane Conductance Regulator. Here, we work backwards from the shape of the wild-type protein - found in healthy people, to computationally model the shape of the most common Cystic Fibrosis mutant. Our computer models reveal distinct defects in the shape of the mutant Cystic Fibrosis Transmembrane Conductance Regulator protein in the area surrounding the mutation. We also model an important FDA approved Cystic Fibrosis drug, VX-809, into the mutant protein structure and show how VX-809 stabilizes the protein around the location of the mutation. The method we developed will pave the way for computational drug design for Cystic Fibrosis.

bioinformatics↗