Reviews & Analysis

CheckM2 is a tool that applies machine learning to evaluate the quality of genomes from metagenomic data. CheckM2 is faster and more accurate than existing methods, and it outperforms them when applied to novel lineages and lineages with reduced genome sizes, such as Patescibacteria and the DPANN superphylum.

We developed, characterized and validated nLight sensors, a new family of genetically encoded green and red fluorescent norepinephrine indicators based on an alpha-1 adrenergic receptor. nLight probes can detect norepinephrine in living animals with superior sensitivity, ligand specificity and temporal resolution as compared with previous tools.

We developed CREST (CRISPR editing-based lineage-specific tracing) to enable high-throughput mapping of single-cell lineages in any Cre lineage of interest in mice. In addition, we delineated a comprehensive lineage landscape of the developing mouse ventral midbrain, revealing novel differentiation trajectories and molecular programs underlying neural specification.

A decade ago, the first bioinformatics pipelines to detect circular RNA molecules based on short-read sequencing data were published. Here, we show that dozens of such circular RNA detection tools differ vastly in their sensitivity but not in their specificity.

We developed LIONESS, a technology that leverages improvements to optical super-resolution microscopy and prior information on sample structure via machine learning to overcome the limitations (in 3D-resolution, signal-to-noise ratio and light exposure) of optical microscopy of living biological specimens. LIONESS enables dense reconstruction of living brain tissue and morphodynamics visualization at the nanoscale.

Three groundbreaking studies have created a new generation of genetically encoded voltage indicators, empowering us to tackle a host of questions on our path toward understanding the brain.

This Perspective introduces the Minimum Information About Disorder Experiments guidelines, which provide a community consensus on the minimum information required to appropriately describe metadata on experimentally and computationally derived structural state(s) of intrinsically disordered proteins or regions.

This Review provides an overview of computational methods recently developed for detecting and analyzing structural variants using long-read sequencing data.

Genome architecture mapping (GAM) enables understanding of 3D genome structure in the nucleus. We directly compared multiplex-GAM and Hi-C data and found that local chromatin interactions were generally detected by both methods, but active genomic regions rich in enhancers that established higher-order contacts were preferentially detected by GAM.

Cell type-specific transgene expression in mice has broad utility in biomedical research. We developed a versatile system for in vivo transgene delivery using adeno-associated virus (AAV). Efficient and tissue-specific transgene expression is achieved by regulating the expression of the gene encoding the AAV receptor, thereby precisely targeting AAV to the cell type of interest.

Our study introduces conditional autoencoder for multiplexed pixel analysis (CAMPA), a deep-learning framework that uses highly multiplexed imaging to identify consistent subcellular landmarks across heterogeneous cell populations and experimental perturbations. Generating interpretable cellular phenotypes revealed links between subcellular organization and perturbations of RNA production, RNA processing and cell size.

This Perspective discusses how machine learning can help in studying rare diseases using various emerging approaches.

Nano-DMS-MaP focuses in on the structures of individual RNA isoforms, enabling direct examination of the structural diversity of different RNAs inside cells.

A deep learning algorithm maps out the continuous conformational changes of flexible protein molecules from single-particle cryo-electron microscopy images, allowing the visualization of the conformational landscape of a protein with improved resolution of its moving parts.

We developed EmbryoNet, a deep learning tool that can automatically identify and classify developmental defects caused by perturbations of signaling pathways in vertebrate embryos. The tool could help to elucidate the mechanisms of action of pharmaceuticals, potentially transforming the drug discovery process.

Prime editing systems hold tremendous promise for the precise correction of pathogenic mutations. We developed a method to tag sequences modified by a prime editor to evaluate its genome-wide precision for therapeutic applications.

A new mutagenesis platform enables the fast, cost-efficient and automatable production of defined multi-site sequence variants for a wide range of applications. Demonstrations of this method included the generation of SARS-CoV-2 spike gene variants, DNA fragments for large-scale genome engineering, and adeno-associated virus 2 (AAV2) cap genes with improved packaging capacity.

Photoselective sequencing is a new method for genomic and epigenomic profiling within specific regions of a biological specimen that are chosen using light microscopy. This combination of spatial and sequencing information preserves the connections between genomic and environmental properties and deepens our understanding of structure–function relationships in cells and tissues.

We highlight the BUDDY software, which was developed to accurately determine the molecular formulae of unknown chemicals in mass spectrometry data. BUDDY is a bottom-up approach that shows superior annotation performance on reference spectra and experimental datasets. Incorporation of global peak annotation could enable BUDDY to refine formula annotations and reveal feature interrelationships.

Unlike cell surface proteins, secreted proteins are difficult to quantify and trace back to individual cells. We show that the capture of secreted proteins onto their source cell surfaces using an affinity matrix enables simultaneous measurement of protein secretion, cell surface proteins and transcriptomics in thousands of cells at single-cell resolution.