Articles

Keypoint-MoSeq is an unsupervised behavior segmentation algorithm that extracts behavioral modules from keypoint tracking data acquired with diverse algorithms, as demonstrated in mice, rats and fruit flies. The extracted modules faithfully reflect human-annotated behaviors even though they are obtained in an unsupervised fashion.

This work presents ORFtag that enables proteome-wide functional screens by tagging and overexpression of endogenously encoded proteins via randomly integrated cassettes.

This work introduces CRISPRdelight, a dCas12a-based DNA-imaging tool that facilitates the imaging of non-repetitive loci and the tracking of DNA dynamics.

Gapr is an efficient platform for reconstructing neurons in large-scale light microscopy datasets. It enables various proofreading modes as well as collaboration among many annotators.

CandyCrunch is a deep learning-based tool for predicting glycan structures from tandem mass spectrometry data. The paper also introduces CandyCrumbs that automatically annotates fragment ions in higher-order tandem mass spectrometry spectra.

Lightning Pose is an efficient pose estimation approach that requires few labeled training data owing to its semi-supervised learning strategy and ensembling.

SpatialGlue is a graph neural network-based approach for integrating multimodal spatial omics data. Combining complementary data modalities improves the discovery of spatial domains as well as the identification of cell subpopulations across tissues.

Bessel-droplet two-photon fluorescence microscopy offers high-contrast and high-resolution volumetric imaging in vivo and can be used for high-throughput mapping of functional synaptic organization in the mouse brain.

Mellon accurately estimates cell-state densities using high-dimensional single-cell data, unveiling regulatory dynamics and mechanisms in cell differentiation.

Permittivity tensor imaging is a label-free computational microscopy approach that enables the three-dimensional measurement of molecular permittivity tensors, revealing information about a biomolecule’s dry mass and orientation in cells and tissues.

This work describes scEdU-seq for studying replication fork speed in single cells, which enables researchers to investigate variability in replication speed along S phase and its associations with transcription and DNA damage.

DART (drug acutely restricted by tethering) enables the manipulation of native receptors on genetically defined neurons. This work describes second-generation DART reagents for manipulating GABA_A and AMPA receptors with higher cellular specificity than previously achieved.

Although not widely appreciated, measured smFRET efficiencies can vary depending on illumination intensity. Self-healing fluorophores offer robust triplet state suppression to enable accurate smFRET measurements under broad illumination regimes.

CellRank 2 is a comprehensive framework of cell fate analysis using large-scale multiview single-cell data.

Blush regularization makes use of a neural network pre-trained on a diverse set of high-resolution cryo-EM half-maps to improve image alignment, effectively lowering the size barrier, during cryo-EM structure determination.

Deconwolf is a computationally efficient and user-friendly software tool for fluorescence microscopy image deconvolution that improves the analysis of diverse fluorescence in situ hybridization methods and can handle large datasets.

scFoundation, with 100 million parameters covering about 20,000 genes, pretrained on over 50 million single-cell transcriptomics profiles, is a foundation model for diverse tasks of single-cell analysis.

Integrative and reference-informed tissue segmentation (IRIS) harnesses single-cell RNA sequencing data for the accurate identification of spatial domains in spatially resolved transcriptomics. IRIS is computationally efficient and uniquely suited for analyzing large datasets.

In targeted-illumination confocal microscopy, the combination of targeted illumination with confocal detection allows one-photon voltage imaging across large fields of view with a high signal-to-noise ratio.

uiPSF (universal inverse modeling of point spread functions) offers a versatile solution for inferring accurate PSF models from images of beads and blinking fluorophores, enabling improved imaging and image processing in localization microscopy.