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 compact adaptive optics module using a deformable mirror compensates for optical aberrations and enables synaptic-resolution imaging of neuronal structures in deep layers of the mouse brain.
Community-driven initiatives are proposing standards to improve the reporting and reproducibility of machine learning in biology. We support these developments, some of which are described in this month’s special issue.
DOME is a set of community-wide recommendations for reporting supervised machine learning–based analyses applied to biological studies. Broad adoption of these recommendations will help improve machine learning assessment and reproducibility.
We present the AIMe registry, a community-driven reporting platform for AI in biomedicine. It aims to enhance the accessibility, reproducibility and usability of biomedical AI models, and allows future revisions by the community.
To make machine-learning analyses in the life sciences more computationally reproducible, we propose standards based on data, model and code publication, programming best practices and workflow automation. By meeting these standards, the community of researchers applying machine-learning methods in the life sciences can ensure that their analyses are worthy of trust.
Deep learning algorithms are powerful tools for analyzing, restoring and transforming bioimaging data. One promise of deep learning is parameter-free one-click image analysis with expert-level performance in a fraction of the time previously required. However, as with most emerging technologies, the potential for inappropriate use is raising concerns among the research community. In this Comment, we discuss key concepts that we believe are important for researchers to consider when using deep learning for their microscopy studies. We describe how results obtained using deep learning can be validated and propose what should, in our view, be considered when choosing a suitable tool. We also suggest what aspects of a deep learning analysis should be reported in publications to ensure reproducibility. We hope this perspective will foster further discussion among developers, image analysis specialists, users and journal editors to define adequate guidelines and ensure the appropriate use of this transformative technology.
A deep-learning-based tool and a large ground truth dataset enable spike inference from calcium imaging data acquired in a variety of experimental conditions.
Technological innovations in optical object recognition and high-throughput ultrasensitive mass spectrometry are enabling subcellular metabolomics and peptidomics, providing unprecedented opportunities to study small-molecule mediators of cellular function with important implications in health and disease.
Dynamic mass photometry, a method based on optical imaging of unlabeled proteins, enables direct observation and tracking of single-protein interactions on lipid membranes.
This Perspective describes advances in computer science that enable the integration of deep learning with traditional knowledge-based modeling in biological sciences, and discusses how such integration might overcome the challenges of modeling sparse, incomplete and noisy experimental data.
DeepImageJ offers a user-friendly solution in ImageJ to run trained deep learning models for biomedical image analysis. It includes guiding tools for reliable analyses, contributing to the democratization of deep learning in microscopy.
By using a new deep learning architecture, Enformer leverages long-range information to improve prediction of gene expression on the basis of DNA sequence.
This work describes inCITE-seq that jointly measures intranuclear protein levels and the transcriptome in single nuclei, which is applied to mouse brain tissue to relate quantitative protein levels of TFs to gene expression programs.
This work describes the generation of a modification-free RNA library that resembles endogenous transcriptome sequence and expression level, which can be used as a negative control in epitranscriptomic sequencing methods to obtain high-confidence and quantitative maps of various RNA modifications.
SEAM is a platform for the analysis of high-resolution secondary ion mass spectrometry imaging that allows spatially resolved nuclear metabolomic profiling at the single-cell level.
An iSCAT image processing and analysis strategy enables mass-sensitive particle tracking (MSPT) of single unlabeled biomolecules on a supported lipid bilayer. MSPT was used to observe the (dis-)assembly of membrane complexes in real-time.
Dynamic mass photometry allows label-free tracking and mass measurement of individual membrane-associated proteins diffusing on supported lipid bilayers. The approach can be used to monitor dynamic (dis)assembly of protein complexes.
Three-photon microscopy in combination with adaptive optics-based aberration correction and ECG-triggered gating allows high-resolution imaging of neurons and astrocytes up to a depth of 1.4 mm in the mouse brain.
A compact adaptive optics module corrects aberrations in two-photon and three-photon microscopy, enabling structural and functional imaging deep in the mouse brain, the mouse spinal cord and the zebrafish larva.