Volume 19 Issue 6, June 2022

MicroED for structure determination without a model and with heavy metal guitar riffs.

How to gauge academic culture and celebrate being an LGBT+ scientist.

The release of the first telomere-to-telomere (T2T) human genome sequence marks a milestone for human genomics research and holds promise of complete genomes for evolutionary genomic studies. Here we describe the advances that this new human genome assembly represents and explore the potential insights that the complete genome sequence could bring to evolutionary genomics. We also discuss the potential challenges to be faced in applying this new sequencing strategy to a broad spectrum of extant species.

A chemical cocktail reprograms human somatic cells into pluripotent stem cells.

TALE-fused deaminases enable C-to-T or A-to-G base editing in human mitochondria.

To recapitulate some of the complexities of the human physiology, engineered tissues can be linked by vascular flow in a multi-organ chip.

A newly described fluorescent protein, StayGold, is bright and extremely photostable, enabling extended time-lapse imaging.

A coarse-grained model enables simulating droplet formation with low-complexity RNA sequences.

Widespread adoption of new embryo models hinges on how faithfully they mirror real embryos.

A repeat-aware pipeline corrects structural assembly errors and polishes assembly sequences in large repeats without overcorrection.

DiMeLo-seq leverages immunotethered DNA methyltransferases with long-read sequencing to map the locations of chromatin proteins in their natural context.

New data collection methods enable ab initio protein structure determination by micro-electron diffraction for the first time.

A novel bright near-infrared fluorescent protein inserted into a nanobody enables visualization of native proteins inside living cells and specific manipulation of cell function, including Boolean protein-based operators.

Tangram, gimVI and SpaGE outperformed other integration methods for predicting the spatial distributions of RNA transcripts, while Cell2location, SpatialDWLS and RCTD were the top-performing methods for the cell type deconvolution of spots in histological sections.

Repository-scale analysis of hundreds of millions to billions of mass spectra is a challenging endeavor due to the complexity and volume of associated data. A deep neural network embedding method is presented that enables large-scale investigation of repeatedly observed yet consistently unidentified mass spectra.

Determining the functional properties of a protein from its structure is challenging. This study presents an interpretable deep learning model that directly learns function-bearing structural motifs from raw data, allowing accurate mapping of protein binding sites and antibody epitopes onto a protein structure.

This work presents a comprehensive benchmarking analysis of computational methods that integrates spatial and single-cell transcriptomics data for transcript distribution prediction and cell type deconvolution.

This work presents hifiasm-meta, a software tool designed for assembling metagenomes using high-accuracy long-read data.

GLEAMS, a deep learning-based algorithm, embeds mass spectra such that spectra related to the same peptide are close to each other, enabling unknown spectra to be identified on a massive scale.

ColabFold is a free and accessible platform for protein folding that provides accelerated prediction of protein structures and complexes using AlphaFold2 or RoseTTAFold.

ASLPrep is a software suite for reproducible processing of arterial spin labeled magnetic resonance imaging data.

The work describes the validation and polishing strategies developed by the telomere-to-telomere consortium for evaluating and improving the first complete human genome assembly.

Merfin is a computational tool for variant filtering that improves accuracy in genotyping and genome assembly polishing.

Winnowmap2 enables better long-read mapping and more accurate variant calling in repetitive regions of the genome.

DiMeLo-seq uses native long-read sequencing to examine protein–DNA interactions by mapping exogenous methylation marks generated by a nonspecific DNA methyltransferase, as well as profile endogenous CpG methylation simultaneously.

This article reports sub- and near-atomic structures of triclinic lysozyme and serine protease proteinase K, respectively, providing first demonstrations of ab initio phasing using electron counted MicroED data to solve macromolecular structures.

ScanNet is an end-to-end, interpretable geometric deep learning model that learns spatio-chemical and atomic features directly from protein 3D structures and can be employed for functional site prediction tasks.

miRFP670nano3 offers improved near-infrared imaging and was used to develop fluorescent nanobodies whose stability and fluorescence strongly depend on antigen binding, with broad implications for detecting and manipulating cellular targets.

Nanofluidic scattering microscopy enables label-free, quantitative measurements of the molecular weight and hydrodynamic radius of biological molecules and nanoparticles freely diffusing inside a nanofluidic channel.

CELESTA identifies cell types in multiplexed imaging datasets based on the expression profiles of cells and their spatial information.