Biology subjects

Thiede, E. H.

Publications and source records attributed to Thiede, E. H..

3 recordsLinked to original sources

bioRxiv2026

Multi-Peptide Prompting Enables In-Context Learning in Protein Language

Protein language models (PLMs) are trained primarily on individual protein sequences, yet many peptide-discovery problems require inference from only a small number of labeled examples. Here, we show that single-sequence PLMs can perform in-context peptide learning without gradient updates, task-specific retraining, or architectural modification. We introduce multi-peptide example prompts (MPEPs), in which demonstration peptides are concatenated with glycine spacers and used as context for scoring query peptides by their prompted probability. We evaluate this approach across a synthetic pattern-completion task, secondary-structure classification, and MHC-II binder prediction using both encoder-only ESM-2 models and decoder-only ProGen2 models. Across tasks, performance improves with the number of peptide examples and with model scale, indicating that PLMs can extract shared sequence-level properties from prompted examples. We further introduce a difference score that contrasts positive-example and negative-example MPEPs, reducing compositional biases in raw PLM probabilities and substantially improving classification. On MHC-II binder prediction, MPEP-based classification with larger ESM-2 models matches or exceeds low-data classifiers trained on frozen ESM-2 embeddings, while requiring no training. These results reveal an unexpected in-context inference capability in single-sequence PLMs and establish MPEP conditioning as a lightweight strategy for low-data peptide classification.

biochemistry↗

bioRxiv2025

Cryo-electron microscopy ensemble optimization using individual particles and physical constraints

AO_SCPLOWBSTRACTC_SCPLOWBiomolecules are inherently dynamic, and understanding their conformational ensemble distributions is essential for understanding their dynamics and biological roles. Cryo-electron microscopy (cryo-EM), a technique that images individual biomolecules frozen in a thin layer of amorphous ice, has emerged as a leading method for determining the structure of biomolecules at atomic resolution. Recent advances in cryo-EM reconstruction have made significant progress in determining structure in heterogeneous conformational landscapes. In contrast to reconstruction, a different class of techniques has been used to infer population weights, referred to as ensemble reweighting. These methods have yet to be generalized to infer structural heterogeneity simultaneously. Here, we present a method for cryo-EM ensemble optimization that directly infers the optimal set of structures and their associated population weights from cryo-EM images using Bayesian optimization techniques. Our method iterates between optimizing the structures and weights using a likelihood defined in terms of cryo-EM particle images (not reconstructions) and projecting onto the domain of a physical prior through an approach inspired by projected gradient descent. We test the method on several systems, ranging from a four-atom toy model to a large protein system with real cryo-EM data. We find that our approach successfully recovers the structures and their associated weights across a wide range of experimental conditions, even when the number of structures does not match the actual number of metastable states. Our method paves the way for cryo-EM ensemble optimization of flexible biomolecules exhibiting complex, multimodal conformational landscapes.

biophysics↗

bioRxiv2020

Insulin dissociates by diverse mechanisms of coupled unfolding and unbinding

The protein hormone insulin exists in various oligomeric forms, and a key step in binding its cellular receptor is dissociation of the dimer. This dissociation process and its corresponding association process have come to serve as a paradigms of coupled (un)folding and (un)binding more generally. Despite its fundamental and practical importance, the mechanism of insulin dimer dissociation remains poorly understood. Here, we use molecular dynamics simulations, leveraging recent developments in umbrella sampling, to characterize the energetic and structural features of dissociation in unprecedented detail. We find that the dissociation is inherently multipathway with limiting behaviors corresponding to conformational selection and induced fit, the two prototypical mechanisms of coupled folding and binding. Along one limiting path, the dissociation leads to detachment of the C-terminal segment of the insulin B chain from the protein core, a feature believed to be essential for receptor binding. We simulate IR spectroscopy experiments to aid in interpreting current experiments and identify sites where isotopic labeling can be most effective for distinguishing the contributions of the limiting mechanisms.

biophysics↗