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.
A virtual mouse model for training behavioral analysis algorithms
A 3D virtual mouse model was generated by adding features such as fur and whiskers to a mesh model of the body. After annotation with landmarks and generation of realistic videos of behavior, the virtual mouse model can be used to train behavioral analysis algorithms.
Single-molecule FRET is making stepwise progress toward the realization of its full potential: becoming the reference technique to monitor protein structural dynamics in live cells.
This Perspective describes advances that have enabled robust directed evolution in mammalian cells. These approaches are poised to improve the development of new generations of tools to probe or modulate mammalian biology.
An updated version of DIAMOND uses improved algorithmic procedures and a customized high-performance computing framework to make seemingly prohibitive large-scale protein sequence alignments feasible.
ROSE-Z achieves axial interference through an asymmetrical optical scheme, yielding 2 nm axial localization precision with ~3,000 photons and a single objective, which offers improved multicolor three-dimensional localization microscopy for cellular structures.
Bolaños et al. present a realistic three-dimensional virtual mouse model that can be animated and that facilitates the training of pose estimation algorithms.
Martini 3.0 is an updated and reparametrized force field for coarse-grained molecular dynamics simulations with new bead types and an expanded ability to model molecular packing and interactions.
A deep-learning-guided approach enables protein engineering using only a small number (‘low N’) of functionally characterized variants of target proteins.
A suite of generally applicable methods and tools, developed to enable single-molecule FRET-based studies of transmembrane proteins diffusing in the cell membrane of living cells, was used to study the oligomerization and dynamics of GPCRs.
Point-scanning super-resolution imaging uses deep learning to supersample undersampled images and enable time-lapse imaging of subcellular events. An accompanying ‘crappifier’ rapidly generates quality training data for robust performance.