Browse Articles

Science needs diversity, but barriers such as housing insecurity can hinder access to education and training. To lobby for change, students put their own experiences to work.

CryoREAD automatically builds DNA–RNA atomic structure from cryo-EM maps. Backbone accuracy is typically >85% and the method is applicable for maps with RNA-only, DNA-only and DNA–RNA–protein complex structures. CryoREAD uses deep learning to identify structure information and subsequently construct the 3D structure of nucleic acids.

We conducted a comprehensive long-read RNA sequencing (RNA-seq) benchmarking experiment by combining spike-ins and in silico mixtures to establish a ground-truth dataset. We used long- and short-read RNA-seq technology to deeply sequence samples and compared the performance of a range of analysis tools on these data.

Few methods for three-dimensional structure modeling of nucleic acids from cryo-EM data exist. CryoREAD, a fully automated DNA/RNA atomic structure modeling method based on deep learning, was developed to fill this gap.

A comprehensive redevelopment of the ribosome profiling workflow involves improved nuclease treatment and sequencing library preparation, enabling richer and more accurate translatome profiling with lower input and fewer technical hurdles.

Single-cell Deep Visual Proteomics integrates imaging, cell segmentation, laser microdissection and multiplexed mass spectrometry for spatial single-cell proteomics measurements in complex tissues.

This analysis leverages experimentally sequenced data and in silico mixtures to simulate transcript expression differences, which enables a performance assessment of long-read tools developed for isoform detection, differential transcript expression analysis and differential transcript usage analysis.

An updated version of the Waxholm Space atlas of the rat brain includes more detailed annotations of several brain regions, including the cortex, striatopallidal region, midbrain and thalamus, expanding the previous version with 112 new and 57 revised structures.

We developed CellOT, a tool that integrates optimal transport with input convex neural networks to predict molecular responses of individual cells to various perturbations. By learning a map between the unpaired distributions of unperturbed and perturbed cells, CellOT outperforms current methods and generalizes the inference of treatment outcomes in unobserved cell types and patients.

GSFA is a statistical model to automatically detect latent factors (or gene modules) in single-cell CRISPR screening.

CellOT combines the benefits of optimal transport and input convex neural architectures to directly learn and uncover maps between control and perturbed cell states at the single-cell level.

A deep learning approach bypasses iterative trials associated with sensorless adaptive optics to compensate for wavefront deformations when imaging biological specimens, enabling improved deep tissue localization microscopy.

Guided sparse factor analysis (GSFA) is a powerful statistical framework to detect changes in gene expression as a result of perturbations in single-cell CRISPR screening.

D-LMBmap is a fully automated pipeline for mesoscale connectomics including deep-learning modules for axon segmentation, brain region segmentation and whole-brain registration. D-LMBmap works accurately across cell types and modalities.

AlteredPQR is a software tool, available as an R package, to infer remodeling of protein functional modules from whole-cell or tissue lysate proteomic measurements.

A Ni²⁺-modified MspA nanopore construct can unambiguously discriminate the 20 proteogenic amino acids as well as several post-translational modifications.

Droplet-based microfluidics enable rapid mixing with millisecond dead times and allow single-molecule measurements of non-equilibrium binding kinetics on even challenging, strongly adsorptive samples, such as intrinsically disordered proteins.

Modern high-throughput metagenomics is producing hundreds of thousands of metagenome-assembled genomes (MAGs), which is overwhelming traditional sequence-similarity search methods. We present a computational method, skani, that efficiently compares MAGs on a terabyte scale while being robust to the inherent noise in MAGs, enabling larger and more accurate analyses.

skani achieves fast calculation of average nucleotide identity (ANI) between metagenome-assembled genomes (MAGs), with improved robustness against incomplete and fragmented MAGs.

EzMechanism is a tool for automated prediction of the catalytic mechanisms of enzymes using their three-dimensional structures and chemical reactions as input.