Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Protein interaction fingerprinting using deep learning
An artistic representation of molecular surface interaction fingerprinting (MaSIF). Information flows in living systems, such as when proteins interact with each other. The watercolor medium expresses that flow, which is further highlighted by the neural network, identifying their connection.
This Perspective highlights open-source software for single-cell analysis released as part of the Bioconductor project, providing an overview for users and developers.
Comprehensive evaluation of algorithms for inferring gene regulatory networks using synthetic and experimental single-cell RNA-seq datasets finds heterogeneous performance and suggests recommendations to users.
Wtdbg2 assembles genomes with comparable contiguity and accuracy to existing tools using long-read sequencing data, and is several times faster, especially for large genomes.
NicheNet uses expression data, in combination with a previous model built on known signaling and gene regulatory networks, to predict ligand–target links in cell-to-cell communications.
An adaptive excitation source enables two- and three-photon imaging of the awake mouse brain with high spatial and temporal resolution at 30-fold-reduced laser power relative to conventional approaches.
An optogenetic strategy enables selection of proteases with improved catalytic rates. The developed TEV protease variants are well suited for biotechnology applications, including FLARE assays with substantially improved temporal resolution.
Protein–peptide interactions that underpin cell signaling are accurately predicted by wedding the strengths of machine learning with the interpretability of biophysical theory, facilitating detailed mechanistic analyses at the proteome scale.
MaSIF, a deep learning-based method, finds common patterns of chemical and geometric features on biomolecular surfaces for predicting protein–ligand and protein–protein interactions.
A statistical method called SPARK for analyzing spatially resolved transcriptomic data can efficiently identify spatially expressed genes with effective control of type I errors and high statistical power.
The unique advantages of single-particle cryo-electron microscopy and cryo-electron tomography are combined in a method called TYGRESS, here applied to determine the structure of the intact ciliary axoneme at a resolution of 12 Å.
A template-free image processing approach automatically detects and classifies membrane-bound protein complexes in cryo-electron tomograms of isolated endoplasmic reticulum and in intact cells.
Advances in MINFLUX nanoscopy enable multicolor imaging over large fields of view, bringing true nanometer-scale fluorescence imaging to labeled structures in fixed and living cells.
A new method of autophagy measurement is based on the detection of phospho-ATG16L1, a conserved early marker of autophagy. Sensitive detection can be achieved in multiple biological systems and assays with advantages over standard methods.