Validation of helical symmetry parameters in the EMDB

This article reports a systematic validation of the consistency between the deposited helical parameters and the density maps of helical structures in the EMDB. Multiple parameter errors were identified and corrected.

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

Estimating structural heterogeneity in cryo-EM remains difficult due to noise, misclassification and algorithmic biases that obscure subtle conformational differences. This work introduces a statistical framework based on p-values to assess whether a given classification reflects meaningful structural variation or is indistinguishable from random partitioning.

Off-target structural insights: ArnA and AcrB in bacterial membrane-protein cryo-EM analysis

This study highlights that cryo-EM can reveal structural information beyond the intended target, and that overexpression of membrane proteins involved in quality control may lead to the co-purification of proteins such as ArnA and AcrB.

Categorizing prediction modes within low-pLDDT regions of AlphaFold2 structures: near-predictive, pseudostructure and barbed wire

Behavioral modes of AlphaFold2 predictions at low pLDDT are identified and described, ranging from unprotein-like `barbed wire' to near-predictive folds, and a tool to automate this identification is provided. Associations between prediction modes and features of interest, such as regions of conditional folding and signal peptides, are discussed.

Room-temperature X-ray data collection enabled the structural determination of statin-bound CYP105A1

The comparison of CYP105A1–statin complex structures obtained from data collected at cryogenic and room temperature revealed release of the statin from the active site under cryogenic conditions. It was found that shrinkage of the c axis caused a cis–trans conversion of Pro142, resulting in changes to the active-site conformation.

Perspective on structure predictions of disorder

Low-confidence regions in computational protein models have been identified to represent a largely untapped resource that may contain valuable structural information.

Biophysical Chemistry, Second Edition