Newest papers

Learning the Shared Latent Structure of Cryo-EM Inverse Problems

Nithin Chalapathi, Theo Sternlieb, Ethan Han, Sonya M. Hanson, Pilar Cossio, Aditi S. Krishnapriyan, Ritwik Gupta · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.22523 · cited by 0 works (OpenAlex) · open-access PDF

German

Das Verständnis der biomolekularen Teilchenfunktion erfordert die Charakterisierung der dreidimensionalen Struktur und der konformationalen Variabilität, für die die cryogenic Electron Microscopy (CR-EM) ein leistungsfähiges Werkzeug ist. Cryo-EM erzeugt laute zweidimensionale Teilchenbilder, in denen die molekulare Struktur, die Teilchenposition und die Bildbedingungen gemeinsam kodiert werden. Die Wiederherstellung von biologischen Informationen aus diesen Beobachtungen umfasst umgekehrte Probleme, einschließlich Positionsschätzung, Heterogenitätsanalyse und Rekonstruktion. Diese Aufgaben werden normalerweise unabhängig angegangen, obwohl die gemeinsame Proteinstruktur und Bildbildung eine gemeinsame Darstellung über umgekehrte Probleme motiviert. Da biologisch bedeutende Unterschiede räumlich

Machine translation (Meta M2M100); the English original follows.

Abstract

Understanding biomolecular function requires characterizing three-dimensional structure and conformational variability, for which cryogenic electron microscopy (cryo-EM) is a powerful tool. Cryo-EM produces noisy two-dimensional particle images in which molecular structure, particle pose, and imaging conditions are jointly encoded. Recovering biological information from these observations spans inverse problems including pose estimation, heterogeneity analysis, and reconstruction. These tasks are usually addressed independently despite shared protein structure and image formation, motivating a common representation across inverse problems. Since biologically meaningful differences may be spatially localized, we study self-distillation with complementary global and masked patch objectives. Because experimental dataset discrimination can reflect acquisition-specific cues, we develop controlled synthetic and experimental evaluations and find that structural discrimination benefits from dense spatial features. We introduce the EMber family of vision transformers, pre-trained on >50 million experimental particle images from 214 EMPIAR datasets together with simulated projections from 268,000 molecular assemblies. EMber improves upon existing methods at conformational and compositional discrimination by 14.85 and 14.26 percentage points, estimates conformational populations to within 1.59pp error with a linear probe, and produces particle assignments yielding reference-quality reconstructions in fewer epochs. We demonstrate how to learn representations that capture biological information for use across multiple cryo-EM inverse problems.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Representation (politics), Inverse problem, Inverse, Computer science, Artificial intelligence, Function (biology)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange