Explore the vast range of titles available.

As the number of single‐cell transcriptomics datasets grows, the natural next step is to integrate the accumulating data to achieve a common ontology of cell types and states. However, it is not straightforward to compare gene expression levels across datasets and to automatically assign cell type labels in a new dataset based on existing annotations. In this manuscript, we demonstrate that our previously developed method, scVI, provides an effective and fully probabilistic approach for joint representation and analysis of scRNA‐seq data, while accounting for uncertainty caused by biological and measurement noise. We also introduce single‐cell ANnotation using Variational Inference (scANVI), a semi‐supervised variant of scVI designed to leverage existing cell state annotations. We demonstrate that scVI and scANVI compare favorably to state‐of‐the‐art methods for data integration and cell state annotation in terms of accuracy, scalability, and adaptability to challenging settings. In contrast to existing methods, scVI and scANVI integrate multiple datasets with a single generative model that can be directly used for downstream tasks, such as differential expression. Both methods are easily accessible through scvi‐tools. SYNOPSIS This study demonstrates the ability of scVI to integrate single‐cell RNA‐seq datasets in a variety of settings and presents scANVI, a new development based on scVI for automated annotation of cell types and states. In scVI, datasets from different labs and technologies are integrated in a joint latent space. In scANVI, cell type annotations are transferred between datasets and across different scenarios. Uncertainties of differential gene expression in multiple samples are quantified. The performance of scVI and scANVI in data integration and cell state annotation is superior to other related methods. Graphical Abstract This study demonstrates the ability of scVI to integrate single‐cell RNA‐seq datasets in a variety of settings and presents scANVI, a new development based on scVI for automated annotation of cell types and states.

RNA velocity has enabled the recovery of directed dynamic information from single‐cell transcriptomics by connecting measurements to the underlying kinetics of gene expression. This approach has opened up new ways of studying cellular dynamics. Here, we review the current state of RNA velocity modeling approaches, discuss various examples illustrating limitations and potential pitfalls, and provide guidance on how the ensuing challenges may be addressed. We then outline future directions on how to generalize the concept of RNA velocity to a wider variety of biological systems and modalities. Graphical Abstract This Review discusses the emerging challenges and potential pitfalls of current RNA velocity modeling approaches and provides guidance on how to address them.

β‐Catenin (Armadillo in Drosophila ) is a multitasking and evolutionary conserved molecule that in metazoans exerts a crucial role in a multitude of developmental and homeostatic processes. More specifically, β‐catenin is an integral structural component of cadherin‐based adherens junctions, and the key nuclear effector of canonical Wnt signalling in the nucleus. Imbalance in the structural and signalling properties of β‐catenin often results in disease and deregulated growth connected to cancer and metastasis. Intense research into the life of β‐catenin has revealed a complex picture. Here, we try to capture the state of the art: we try to summarize and make some sense of the processes that regulate β‐catenin, as well as the plethora of β‐catenin binding partners. One focus will be the interaction of β‐catenin with different transcription factors and the potential implications of these interactions for direct cross‐talk between β‐catenin and non‐Wnt signalling pathways. Konrad Basler and colleagues survey and interpret the vast literature on armadillo/β‐catenin. The result is a very broad and informative picture of this evolutionary‐conserved, versatile protein.

To deal with the huge number of novel protein‐coding variants identified by genome and exome sequencing studies, many computational variant effect predictors (VEPs) have been developed. Such predictors are often trained and evaluated using different variant data sets, making a direct comparison between VEPs difficult. In this study, we use 31 previously published deep mutational scanning (DMS) experiments, which provide quantitative, independent phenotypic measurements for large numbers of single amino acid substitutions, in order to benchmark and compare 46 different VEPs. We also evaluate the ability of DMS measurements and VEPs to discriminate between pathogenic and benign missense variants. We find that DMS experiments tend to be superior to the top‐ranking predictors, demonstrating the tremendous potential of DMS for identifying novel human disease mutations. Among the VEPs, DeepSequence clearly stood out, showing both the strongest correlations with DMS data and having the best ability to predict pathogenic mutations, which is especially remarkable given that it is an unsupervised method. We further recommend SNAP2, DEOGEN2, SNPs&GO, SuSPect and REVEL based upon their performance in these analyses. Synopsis Data from deep mutational scans is used to benchmark computational protein variant effect predictors using fully independent data. The performance of deep mutational scanning is also compared to computational predictors for identifying pathogenic variants. DeepSequence is the method that correlates the best with deep mutational scanning data for human proteins. Predictor performance depends heavily on the protein and fitness metric. For this reason, using results from multiple predictors is recommended. Other recommended predictors include SNAP2, DEOGEN2, SNPs&GO, SuSPect and REVEL. Deep mutational scanning is generally superior to variant effect predictors for distinguishing pathogenic from benign variants. Graphical Abstract Data from deep mutational scans is used to benchmark computational protein variant effect predictors using fully independent data. The performance of deep mutational scanning is also compared to computational predictors for identifying pathogenic variants.

The nuclear function of the heterodimeric NF‐κB transcription factor is regulated in part through reversible acetylation of its RelA subunit. We now demonstrate that the p300 and CBP acetyltransferases play a major role in the in vivo acetylation of RelA, principally targeting lysines 218, 221 and 310 for modification. Analysis of the functional properties of hypoacetylated RelA mutants containing lysine‐to‐arginine substitutions at these sites and of wild‐type RelA co‐expressed in the presence of a dominantly interfering mutant of p300 reveals that acetylation at lysine 221 in RelA enhances DNA binding and impairs assembly with IκBα. Conversely, acetylation of lysine 310 is required for full transcriptional activity of RelA in the absence of effects on DNA binding and IκBα assembly. Together, these findings highlight how site‐specific acetylation of RelA differentially regulates distinct biological activities of the NF‐κB transcription factor complex.

High‐throughput ‐omics techniques have revolutionised biology, allowing for thorough and unbiased characterisation of the molecular states of biological systems. However, cellular decision‐making is inherently a unicellular process to which “bulk” ‐omics techniques are poorly suited, as they capture ensemble averages of cell states. Recently developed single‐cell methods bridge this gap, allowing high‐throughput molecular surveys of individual cells. In this review, we cover core concepts of analysis of single‐cell gene expression data and highlight areas of developmental biology where single‐cell techniques have made important contributions. These include understanding of cell‐to‐cell heterogeneity, the tracing of differentiation pathways, quantification of gene expression from specific alleles, and the future directions of cell lineage tracing and spatial gene expression analysis. Graphical Abstract Single‐cell genomic techniques have advanced our understanding of several developmental processes. This Review summarises advances related to generating and analyzing single‐cell transcriptome data and discusses areas of developmental biology that benefited from such technologies.

Automated cell type identification is a key computational challenge in single‐cell RNA‐sequencing (scRNA‐seq) data. To capitalise on the large collection of well‐annotated scRNA‐seq datasets, we developed scClassify, a multiscale classification framework based on ensemble learning and cell type hierarchies constructed from single or multiple annotated datasets as references. scClassify enables the estimation of sample size required for accurate classification of cell types in a cell type hierarchy and allows joint classification of cells when multiple references are available. We show that scClassify consistently performs better than other supervised cell type classification methods across 114 pairs of reference and testing data, representing a diverse combination of sizes, technologies and levels of complexity, and further demonstrate the unique components of scClassify through simulations and compendia of experimental datasets. Finally, we demonstrate the scalability of scClassify on large single‐cell atlases and highlight a novel application of identifying subpopulations of cells from the Tabula Muris data that were unidentified in the original publication. Together, scClassify represents state‐of‐the‐art methodology in automated cell type identification from scRNA‐seq data. Synopsis scClassify is a multiscale classification framework based on ensemble learning and cell type hierarchies, enabling sample size estimation required for accurate cell type classification and joint classification of cells using multiple references. scClassify performs multiscale cell type classification based on cell type hierarchies constructed from single or multiple reference datasets. It implements a post‐hoc clustering procedure for discovering novel cell types from cells that are unassigned due to the absence of their types in the reference data. It enables the estimation of the number of cells required in a reference dataset to accurately discriminate a given cell type in a cell type hierarchy. Application to large atlas datasets such as Tabula Muris demonstrates its ability to refine cell types and identify cells from sub‐populations. Graphical Abstract scClassify is a multiscale classification framework based on ensemble learning and cell type hierarchies, enabling sample size estimation required for accurate cell type classification and joint classification of cells using multiple references.

Polygenic risk scores have identified that genetic variants without genome‐wide significance still add to the genetic risk of developing Alzheimer's disease (AD). Whether and how subthreshold risk loci translate into relevant disease pathways is unknown. We investigate here the involvement of AD risk variants in the transcriptional responses of two mouse models: APPswe/PS1 L166P and Thy‐TAU22. A unique gene expression module, highly enriched for AD risk genes, is specifically responsive to Aβ but not TAU pathology. We identify in this module 7 established AD risk genes ( APOE , CLU , INPP5D , CD33, PLCG2 , SPI1, and FCER1G ) and 11 AD GWAS genes below the genome‐wide significance threshold ( GPC2, TREML2, SYK, GRN, SLC2A5, SAMSN1, PYDC1, HEXB, RRBP1, LYN, and BLNK), that become significantly upregulated when exposed to Aβ. Single microglia sequencing confirms that Aβ, not TAU, pathology induces marked transcriptional changes in microglia, including increased proportions of activated microglia. We conclude that genetic risk of AD functionally translates into different microglia pathway responses to Aβ pathology, placing AD genetic risk downstream of the amyloid pathway but upstream of TAU pathology. Synopsis It is unknown how genetic risk of Alzheimer's disease (AD) manifests itself at the molecular and the cellular level in the brain. Analysis of a TAUtg and an APPtg mouse models show that genetic risk of AD is mainly reflected in the transcriptional responses of microglia to amyloid‐β pathology. GWAS risk genes are enriched among the transcriptomic response to amyloid‐β pathology, but not to TAU pathology. APPtg/Amyloid plaques forming mice show massive transcriptional deregulation with aging, mainly increasing expression of neuroinflammatory genes, while TAUtg/tangle forming mice display mainly downregulation in expression of neuronal genes. Many AD GWAS risk genes above and below genome wide significance are co‐regulated in a large gene expression module involved in neuroinflammation. 18 AD GWAS risk genes are prioritized, which are all expressed in microglia and may regulate their function (in red in the synopsis figure). Single microglia sequencing confirms the expression of 15 risk genes in microglia and confirms that many more microglia adopt an activated phenotype when facing amyloid‐β than TAU pathology. Graphical Abstract It is unknown how genetic risk of Alzheimer's disease (AD) manifests itself at the molecular and the cellular level in the brain. Analysis of a TAUtg and an APPtg mouse models show that genetic risk of AD is mainly reflected in the transcriptional responses of microglia to amyloid‐β pathology.

Limited therapy options due to antibiotic resistance underscore the need for optimization of current diagnostics. In some bacterial species, antimicrobial resistance can be unambiguously predicted based on their genome sequence. In this study, we sequenced the genomes and transcriptomes of 414 drug‐resistant clinical Pseudomonas aeruginosa isolates. By training machine learning classifiers on information about the presence or absence of genes, their sequence variation, and expression profiles, we generated predictive models and identified biomarkers of resistance to four commonly administered antimicrobial drugs. Using these data types alone or in combination resulted in high (0.8–0.9) or very high (> 0.9) sensitivity and predictive values. For all drugs except for ciprofloxacin, gene expression information improved diagnostic performance. Our results pave the way for the development of a molecular resistance profiling tool that reliably predicts antimicrobial susceptibility based on genomic and transcriptomic markers. The implementation of a molecular susceptibility test system in routine microbiology diagnostics holds promise to provide earlier and more detailed information on antibiotic resistance profiles of bacterial pathogens and thus could change how physicians treat bacterial infections. Synopsis The spread of antibiotic resistance complicates infection treatment, requiring an optimization of current diagnostics. In this study, a machine learning approach identified a set of biomarkers suitable for the development of a molecular test system to determine antibiotic resistance profiles. Genome and transcriptome data of 414 clinical isolates was combined for biomarker identification using information on gene expression, gene presence or absence, and single nucleotide variations. For some antibiotics, transcriptome information greatly improves resistance prediction. Depending on the antibiotic, 37–93 biomarkers are sufficient to obtain high (0.8–0.9) or very high (> 0.9) sensitivity and predictive values. Biomarkers include known resistance conferring genes (e.g. gyrA, oprD, ampC, efflux pumps) as well as unexpected and potential novel candidates. Graphical Abstract The spread of antibiotic resistance complicates infection treatment, requiring an optimization of current diagnostics. In this study, a machine learning approach identified a set of biomarkers suitable for the development of a molecular test system to determine antibiotic resistance profiles.

Mammalian chromosomes fold into arrays of megabase‐sized topologically associating domains (TADs), which are arranged into compartments spanning multiple megabases of genomic DNA. TADs have internal substructures that are often cell type specific, but their higher‐order organization remains elusive. Here, we investigate TAD higher‐order interactions with Hi‐C through neuronal differentiation and show that they form a hierarchy of domains‐within‐domains (metaTADs) extending across genomic scales up to the range of entire chromosomes. We find that TAD interactions are well captured by tree‐like, hierarchical structures irrespective of cell type. metaTAD tree structures correlate with genetic, epigenomic and expression features, and structural tree rearrangements during differentiation are linked to transcriptional state changes. Using polymer modelling, we demonstrate that hierarchical folding promotes efficient chromatin packaging without the loss of contact specificity, highlighting a role far beyond the simple need for packing efficiency. Synopsis Genome‐wide mapping of chromatin architecture reveals a hierarchical folding of chromatin that involves higher‐order domains interactions across the whole chromosomes, reflects epigenomic features and reorganizes upon differentiation‐induced gene expression changes. Chromatin architecture is mapped genome‐wide using Hi‐C and a neuronal differentiation model from mESC to post‐mitotic neurons. Mammalian chromosomes fold hierarchically in a manner that reflects epigenomic features and involves higher‐order domains (metaTADs) up to the chromosome scale. metaTAD topologies are relatively conserved through differentiation, and their reorganization is related to gene expression changes. Polymer modelling shows that hierarchical chromatin folding promotes efficient packaging without the loss of contact specificity. Graphical Abstract Genome‐wide mapping of chromatin architecture reveals a hierarchical folding of chromatin that involves higher‐order domains interactions across the whole chromosomes, reflects epigenomic features and reorganizes upon differentiation‐induced gene expression changes.