Research Briefing

Quantitative and sensitive methods for the detection of pseudouridine (Ψ) have been lacking. Now, a method termed 2-bromoacrylamide-assisted cyclization sequencing (BACS) has been developed that enables the accurate quantification of Ψ stoichiometry, precise identification of Ψ positions and robust detection of densely modified Ψ sites.

Similarly to CRISPR–Cas systems, TnpB proteins from bacterial transposons can be employed as RNA-guided endonucleases for genome editing. By combining rational protein design and machine learning, ISDra2 TnpB variants with enhanced editing efficiency and a broader targeting range were developed, along with a prediction tool to design effective guiding RNAs.

FICTURE software addresses a critical challenge in spatial omics analysis: making high-resolution inference with only a few molecules per square micron. This tool fully realizes the potential of contemporary spatial platforms by learning latent spatial factors from the whole transcriptome while preserving the resolution of each technology at scale.

Single-cell RNA-sequencing and spatial transcriptomics data enable the inference of how information is transmitted from one cell to another and how it modulates gene expression within cells. Now, a learning method infers networks describing how the inflow of one signal, mediated by intracellular gene modules, drives the outflow of other signals for intercellular communication.

Rapid advancements in transcriptomics have enabled the quantification of individual transcripts for thousands of genes in millions of single cells. By coupling a machine learning inference framework with biophysical models describing the RNA life cycle, we can explore the dynamics driving RNA production, processing and degradation across cell types.

Single-cell bisulfite sequencing enables the genome-wide quantification of DNA methylation at single-cell resolution, but methods to analyze the resulting data are lacking. The MethSCAn software accurately distinguishes cell types and states by scanning the genome for informative regions and providing a robust approach to quantifying methylation within these regions.

We developed PINNACLE, a graph-based AI model for learning protein representations across cell-type contexts. These contextualized protein representations enable the integration of 3D protein structure with single-cell genomic-based representations to enhance protein–protein interaction prediction, analysis of drug effects across cell-type contexts, and prediction of therapeutic targets in a cell type-specific manner.

A systematic comparison of 11 sequencing-based spatial transcriptomics methods reveals molecular diffusion as a critical variable that influences the effective resolution and data interpretation across platforms. Our benchmarking study should aid biologists in selecting the most appropriate method for their specific tissue.

SpatialGlue is a tool designed to decipher spatial domains from spatial multi-omics data acquired from a single tissue section. It employes graph neural networks with a dual-attention mechanism to accomplish within-omics integration of measured features and spatial information, followed by cross-omics integration.

We demonstrate CRISPRdelight, a robust CRISPR–Cas12a-based method for imaging non-repetitive genomic DNAs in a highly efficient way. This system is a powerful tool for studying functional links between gene dynamics, localization and regulation, and reveals heterogeneity in the expression of differently localized alleles in the same cells.

The annotation of structural glycomics data is a bottleneck in gaining insight into complex carbohydrates. An artificial intelligence model, CandyCrunch, has now been developed that accelerates the annotation process by orders of magnitude and achieves high accuracy in the prediction of glycan structures from tandem mass spectrometry data.

A fundamental mechanism for information processing in the brain is electrical activity. However, observing such activity at the single-cell level is challenging. We have developed an optical microscope that combines the advantages of targeted illumination and confocal gating to enable kilohertz-rate voltage imaging across large fields of view in thick tissue.

Comparing brain connectivity between chimpanzees and humans is a means of understanding human cognition and evolution. To address the scarcity of chimpanzee neuroimaging data, we introduce a high-quality MRI resource that reveals previously unseen anatomical details, offering valuable insights into human brain evolution.

uiPSF is a toolbox to measure point spread functions based on inverse modeling that improves single-molecule localization microscopy (SMLM) localization and microscope characterization, and that works for many microscopy technologies.

An experimental method to study how cells sense and react to external mechanical forces combines controlled mechanical stimulation using nanopipettes with fluorescence imaging of membrane tension. This approach facilitates the study of mechanosensitive ion channels and the propagation of cell membrane tension.

Spatial transcriptomics and mRNA splicing measurements encode rich spatiotemporal information for cell states and their transitions. We present a multiscale dynamical system method for reconstructing cell-state-specific dynamics and spatial state transitions. This theory-based approach reconciles short-timescale local tensor streamlines between cells with long-timescale transition paths that connect cell attractors.

Pebblescout navigates vast, rapidly growing nucleotide content in resources by providing indexing and search capabilities. We used Pebblescout to index a metagenomic subset of Sequence Read Archive and seven other resources into databases spanning over 3.7 petabases and searchable interactively at a pilot website using queries as short as 42 bases.

We developed a two-pronged strategy to functionally probe the enormous repertoire of noncoding DNA within genomes. Our approach markedly improved signal-to-noise ratio and successfully intersected single-cell genomics with reporter assays. The result delivers a multiplex and highly quantitative readout of regulatory sequences’ activity in dynamic and multicellular systems.

Combining post-translational modification site-centric base editing with phenotypic screens uncovers the function of phosphorylation sites in high throughput, enabling the study of expansive signaling networks at a speed comparable to that of functional genomics.

We created DELiVR, a deep-learning pipeline for 3D brain-cell mapping that is trained with virtual reality-generated reference annotations. It can be deployed via the user-friendly interface of the open-source software Fiji, which makes the analysis of large-scale 3D brain images widely accessible to scientists without computational expertise.