How many (distinguishable) classes can we identify in single-particle analysis?

Authors	LAUZIRIKA Oier PERNICA Martin HERREROS David RAMÍREZ-APORTELA Erney KRIEGER James GRAGERA Marcos ICETA Mikel CONESA Pablo FONSECA Yunior JIMÉNEZ Jorge FILIPOVIČ Jiří CARAZO Jose Maria SORZANO Carlos Oscar
Year of publication	2025
Type	Article in Periodical
Magazine / Source	ACTA CRYSTALLOGRAPHICA SECTION D
MU Faculty or unit	Institute of Computer Science
Citation
web	URL
Doi	https://doi.org/10.1107/S2059798325007831
Keywords	cryo-electron microscopy; 3D classification; structural heterogeneity; statistical significance; reproducibility analysis
Description	Heterogeneity in cryoEM is essential for capturing the structural variability of macromolecules, reflecting their functional states and biological significance. However, estimating heterogeneity remains challenging due to particle mis{\-}classification and algorithmic biases, which can lead to reconstructions that blend distinct conformations or fail to resolve subtle differences. Furthermore, the low signal-to-noise ratio inherent in cryo-EM data makes it nearly impossible to detect minute structural changes, as noise often obscures subtle variations in macromolecular projections. In this paper, we investigate the use of {\it p}-values associated with the null hypothesis that the observed classification differs from a random partition of the input data set, thereby providing a statistical framework for determining the number of distinguishable classes present in a given data set.