Articles

Micro-kiss (μkiss) is a micropipette-based approach for delivering very small amounts of nanoparticles and small molecules to the cell surface with exquisite spatiotemporal control, enabling a wide range of biological investigations.

Transcript Imputation with Spatial Single-cell Uncertainty Estimation (TISSUE) offers a general framework for estimating uncertainty for spatial gene expression predictions, enabling improved downstream analysis of spatially resolved transcriptomics data.

CombFold is a combinatorial and hierarchical assembly algorithm for predicting structures of large protein complexes utilizing pairwise interactions between subunits predicted by AlphaFold2.

Exo-PE is an approach to improve prime editing efficacy for insertions while maintaining precision.

ALBATROSS is a deep-learning-based model for predicting ensemble properties of intrinsically disordered proteins and protein regions, such as radius of gyration, end-to-end distance, polymer-scaling exponent and ensemble asphericity, directly from sequences.

RNA family sequence generator (RfamGen) is a deep generative model for designing novel, functional RNA sequences. RfamGen is applicable to diverse RNA families and can yield ribozymes with higher enzymatic activity.

Content-aware frame interpolation (CAFI) improves the temporal resolution in time-lapse imaging by accurately predicting images in between image pairs. By allowing fewer frames to be imaged, CAFI also enables gentler live-cell imaging.

DoTA-seq leverages a microfluidic droplet system to isolate and lyse diverse microbes and amplify target genetic loci, enabling high-throughput single-cell sequencing of microbial populations.

Temporal analysis of relative distances (TARDIS) is a conceptually new alternative to traditional single-particle tracking methods that overcomes challenges associated with high particle density, emitter blinking and spurious localizations.

This work introduces ARTR-seq for in situ measurement of RNA-binding protein (RBP) binding sites, which has been demonstrated in a small number of cells and for capturing dynamic RBP binding within short timeframes.

CytoCommunity enables both supervised and unsupervised analyses of spatial omics data in order to identify complex tissue cellular neighborhoods based on cell phenotypes and spatial distributions.

SynapShot combines ddFPs with engineered synaptic adhesion molecules for real-time observation of the structural plasticity of synapses in cultured cells and animals.

SnapATAC2 uses a matrix-free spectral embedding algorithm for nonlinear dimension reduction of single-cell omics data, which shows an improved performance in capturing cellular heterogeneity and scalability for large datasets.

Real-time mid-infrared photothermal imaging of nitrile chameleons enables simultaneous, multiplexed measurement of enzymatic activity in living systems and is poised to reveal the spatiotemporal regulation of enzymes in health and disease.

InfraRed-mediated Image Restoration (IR²) uses deep learning to combine the benefits of deep-tissue imaging with NIR probes and the convenience of imaging with GFP for improved time-lapse imaging of embryogenesis.

mEnrich-seq leverages DNA methylation differences between bacteria to enrich taxa of interest from metagenomic samples for selective sequencing.

Single-cell structure probing of RNA transcripts enables simultaneous determination of transcript secondary structure and abundance in single cells, allowing new insights into RNA structural heterogeneity within and among cells.

The DeepMSA2 pipeline employs iterative alignment search against large genomic and metagenomic sequence databases to construct single- and multichain multiple-sequence alignment (MSA) for proteins. Use of these MSAs shows improvement for deep learning-based protein tertiary and quaternary structure predictions.

smartLLSM uses artificial intelligence-based instrument control to switch between epiflouorescence and lattice light-sheet microscopy to monitor cells at the population level while also capturing multicolor three-dimensional datasets of rare events of interest.

ANCOM-BC2 is developed to perform multigroup differential abundance analysis and allows modeling of covariates and longitudinal measures while controlling false discovery rate (FDR) or mixed directional FDR.