ADVERTISEMENT RETURN TO ISSUEPREVArticleGlass-forming microemulsions: vitrification of simple liquids and electron microscope probing of droplet-packing modesJ. Dubochet, C. M. Alba, D. R. MacFarlane, C. A. Angell, R. K. Kadiyala, M. Adrian, and J. TeixeiraCite this: J. Phys. Chem. 1984, 88, 26, 6727–6732Publication Date (Print):December 1, 1984Publication History Published online1 May 2002Published inissue 1 December 1984https://pubs.acs.org/doi/10.1021/j150670a042https://doi.org/10.1021/j150670a042research-articleACS PublicationsRequest reuse permissionsArticle Views270Altmetric-Citations72LEARN ABOUT THESE METRICSArticle Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated. Share Add toView InAdd Full Text with ReferenceAdd Description ExportRISCitationCitation and abstractCitation and referencesMore Options Share onFacebookTwitterWechatLinked InRedditEmail Other access options Get e-Alerts
Summary Cryoelectron microsopy is a widely used technique to observe biological material in an almost physiological, fully hydrated state. The sample is prepared for electron microsopy observation by quickly reducing its temperature to −180 °C. The high‐speed cooling induces the formation of vitreous water, which preserves the sample conformation. However, the way vitrification occurs is still poorly understood. In order to better understand the phenomenon, we have used a stroboscopic device to visualize the interaction between the electron microscopy grid and the cryogen. By blocking the free fall of the plunger once the grid has penetrated the coolant by half its diameter, we have elucidated the way in which vitrification propagates. The findings were confirmed by numerical simulation. In addition, according to our observations, we now present an alternative way to prepare vitreous specimens. This new method, with the grid parallel to the liquid cryogen surface, decreases evaporation from the sample during its free fall towards the coolant and at the same time achieves a more uniform vitrification over the entire surface of the specimen.
Predicting ADMET (absorption, distribution, metabolism, excretion, and toxicity) properties of small molecules is a key task in drug discovery. A major challenge in building better ADMET models is the experimental error inherent in the data. Furthermore, ADMET predictors are typically regression tasks due to the continuous nature of the data. This makes it difficult to apply existing methods as most focus on classification tasks. Here, we develop denoising schemes based on deep learning to address this. We find that the training error can be used to identify the noise in regression tasks while ensemble-based and forgotten event-based metrics fail to detect the noise. The most significant performance increase occurs when the original model is finetuned with the denoised data using training error as the noise detection metric. Our method has the ability to improve models with medium noise and does not degrade the performance of models with noise outside this range. To our knowledge, our denoising scheme is the first to improve model performance for ADMET data and has implications for improving models for experimental assay data in general.
Abstract Transmission electron microscopy and scanning force microscopy of negative‐stained, carbon‐coated replica and mica‐adsorbed preparations of 200 μM poly r(A‐U) and 50 μM ethidium bromide/200 μM poly r(A‐U) have been employed to evaluate ethidium‐induced changes in poly r(A‐U) topology. Poly r(A‐U) alone exhibits elongated conformations 85–115 nm in length that possess a number of hairpin loops as well as single‐stranded domains. While the double‐stranded domains are found predominately at the base of the hairpin loops (diameter = 5–30 nm), other rod‐like (presumably double‐stranded) regions ranging from 25–80 nm in length are present in other portions of the poly r(A‐U). In contrast with the poly r(A‐U) alone, the EB/poly r(A‐U) combination appears as a heterogeneous population of condensed structures whose lengths and widths vary from 12–88 nm and 15–45 nm, respectively. These conformational changes are due to a number of factors, including the displacement of ordered water surrounding the poly r(A‐U) and charge shielding of the phosphate groups of the poly r(A‐U) upon the binding of the ethidium.
In this study, we present the scanning force and electron microscopic visualization of single molecules of fibronectin either frozen hydrated or adsorbed onto metallic and polymeric surfaces with different solid surface tensions. The surfaces were characterized by dynamic contact angle measurements, X-ray photo emission spectroscopy (XPS or ESCA) and scanning force microscopy. The proteins were prepared by fast protein liquid chromatography (FPLC) and characterized by gel electrophoresis. Protein films on surfaces were investigated by surface plasmon resonance spectroscopy and directly imaged by scanning force microscopy. The spreading of the adsorbed fibronectin revealed dependence on the chemical composition and the solid surface tension. Structure of fibronectin in solution as well as on solid interface appeared as an extended straight strand as obtained by imaging with electron and scanning probe microscopies. Imaging of DNA was performed by scanning force microscopy to test the accuracy and reproducibility of our measurements. The measured contour lengths were accurate and the larger widths were caused by convolution of the tip shape and sample. Frictional forces during the scan have been of significant contribution in the imaging mechanism. Moreover, this work demonstrated that scanning force microscopy can be used for mapping the orientation and organization of protein film adsorbed onto various surfaces at the nanoscale.