BACKGROUND: Cardiac differentiation of human-induced pluripotent stem (hiPS) cells consistently produces a mixed population of cardiomyocytes and non-cardiac cell types, even when using well-characterized protocols. We sought to determine whether different cell types might result from intrinsic differences in hiPS cells prior to the onset of differentiation. RESULTS: By associating individual differentiated cells that share a common hiPS cell precursor, we tested whether expression variability is predetermined from the hiPS cell state. In a single experiment, cells that shared a progenitor were more transcriptionally similar to each other than to other cells in the differentiated population. However, when the same hiPS cells were differentiated in parallel, we did not observe high transcriptional similarity across differentiations. Additionally, we found that substantial cell death occurs during differentiation in a manner that suggested all cells were equally likely to survive or die, suggesting that there is no intrinsic selection bias for cells descended from particular hiPS cell progenitors. We thus wondered how cells grow spatially during differentiation, so we labeled cells by expression of marker genes and found that cells expressing the same marker tended to occur in patches. Our results suggest that cell type determination across multiple cell types, once initiated, is maintained in a cell-autonomous manner for multiple divisions. CONCLUSIONS: Altogether, our results show that while substantial heterogeneity exists in the initial hiPS cell population, it is not responsible for the variability observed in differentiated outcomes; instead, factors specifying the various cell types likely act during a window that begins shortly after the seeding of hiPS cells for differentiation.

BACKGROUND: Promoters are sites of transcription initiation that harbour a high concentration of phenotype-associated genetic variation. The evolutionary gain and loss of promoters between species (collectively, termed turnover) is pervasive across mammalian genomes and may play a prominent role in driving human phenotypic diversity. RESULTS: We classified human promoters by their evolutionary history during the divergence of mouse and human lineages from a common ancestor. This defined conserved, human-inserted and mouse-deleted promoters, and a class of functional-turnover promoters that align between species but are only active in humans. We show that promoters of all evolutionary categories are hotspots for substitution and often, insertion mutations. Loci with a history of insertion and deletion continue that mode of evolution within contemporary humans. The presence of an evolutionary volatile promoter within a gene is associated with increased expression variance between individuals, but only in the case of human-inserted and mouse-deleted promoters does that correspond to an enrichment of promoter-proximal genetic effects. Despite the enrichment of these molecular quantitative trait loci (QTL) at evolutionarily volatile promoters, this does not translate into a corresponding enrichment of phenotypic traits mapping to these loci. CONCLUSIONS: Promoter turnover is pervasive in the human genome, and these promoters are rich in molecularly quantifiable but phenotypically inconsequential variation in gene expression. However, since evolutionarily volatile promoters show evidence of selection, coupled with high mutation rates and enrichment of QTLs, this implicates them as a source of evolutionary innovation and phenotypic variation, albeit with a high background of selectively neutral expression variation.

The precise spatiotemporal gene expression is orchestrated by enhancers that lack general sequence features and thus are difficult to be computationally identified. By nascent RNA sequencing combined with epigenome profiling, we detect active transcription of enhancers from the complex bread wheat genome. We find that genes associated with transcriptional enhancers are expressed at significantly higher levels, and enhancer RNA is more precise and robust in predicting enhancer activity compared to chromatin features. We demonstrate that sub-genome-biased enhancer transcription could drive sub-genome-biased gene expression. This study highlights enhancer transcription as a hallmark in regulating gene expression in wheat.

BACKGROUND: Patient-derived organoid culture is a powerful system for studying the molecular mechanisms of cancers, especially colorectal cancer (CRC), one of the most prevalent cancers worldwide. There are two main types of 3D culture methods for colonic cells, but the similarities and differences between gene expression patterns in different culture media remain largely unexplored. RESULTS: Here, we establish patient-derived organoids from colorectal cancer patients and perform single-cell RNA-Seq for pairwise samples from seven patients for both organoids and their corresponding tumor and normal tissues in vivo. We find that organoids derived from tumor tissues faithfully recapitulate the main gene expression signatures of cancer cells in vivo. On the other hand, organoids derived from normal tissues exhibited some tumor-like features at the whole transcriptome level but retained normal genomic features, such as CNVs, point mutations, and normal global DNA methylation levels, for both cultural media. More importantly, we show that conditioned medium outperforms chemical-defined medium in long-term culture of tumor epithelial cells. Finally, we mutually exchange the culture medium for the organoids and find that after interchanging the medium, the organoid cells basically maintain the transcriptome characteristics of the original medium. CONCLUSIONS: Our work gives a thorough evaluation of both the cultural conditions and the biological features of organoids of CRC patients.

Despite recent improvements in basecalling accuracy, nanopore sequencing still has higher error rates on short-tandem repeats (STRs). Instead of using basecalled reads, we developed DeepRepeat which converts ionic current signals into red-green-blue channels, thus transforming the repeat detection problem into an image recognition problem. DeepRepeat identifies and accurately quantifies telomeric repeats in the CHM13 cell line and achieves higher accuracy in quantifying repeats in long STRs than competing methods. We also evaluate DeepRepeat on genome-wide or candidate region datasets from seven different sources. In summary, DeepRepeat enables accurate quantification of long STRs and complements existing methods relying on basecalled reads.

BACKGROUND: Revealing the gene targets of distal regulatory elements is challenging yet critical for interpreting regulome data. Experiment-derived enhancer-gene links are restricted to a small set of enhancers and/or cell types, while the accuracy of genome-wide approaches remains elusive due to the lack of a systematic evaluation. We combined multiple spatial and in silico approaches for defining enhancer locations and linking them to their target genes aggregated across >500 cell types, generating 1860 human genome-wide distal enhancer-to-target gene definitions (EnTDefs). To evaluate performance, we used gene set enrichment (GSE) testing on 87 independent ENCODE ChIP-seq datasets of 34 transcription factors (TFs) and assessed concordance of results with known TF Gene Ontology annotations, and other benchmarks. RESULTS: The top ranked 741 (40%) EnTDefs significantly outperform the common, naïve approach of linking distal regions to the nearest genes, and the top 10 EnTDefs perform well when applied to ChIP-seq data of other cell types. The GSE-based ranking of EnTDefs is highly concordant with ranking based on overlap with curated benchmarks of enhancer-gene interactions. Both our top general EnTDef and cell-type-specific EnTDefs significantly outperform seven independent computational and experiment-based enhancer-gene pair datasets. We show that using our top EnTDefs for GSE with either genome-wide DNA methylation or ATAC-seq data is able to better recapitulate the biological processes changed in gene expression data performed in parallel for the same experiment than our lower-ranked EnTDefs. CONCLUSIONS: Our findings illustrate the power of our approach to provide genome-wide interpretation regardless of cell type.

Recent progress in deep learning has greatly improved the prediction of RNA splicing from DNA sequence. Here, we present Pangolin, a deep learning model to predict splice site strength in multiple tissues. Pangolin outperforms state-of-the-art methods for predicting RNA splicing on a variety of prediction tasks. Pangolin improves prediction of the impact of genetic variants on RNA splicing, including common, rare, and lineage-specific genetic variation. In addition, Pangolin identifies loss-of-function mutations with high accuracy and recall, particularly for mutations that are not missense or nonsense, demonstrating remarkable potential for identifying pathogenic variants.

Integrative analysis of large-scale single-cell RNA sequencing (scRNA-seq) datasets can aggregate complementary biological information from different datasets. However, most existing methods fail to efficiently integrate multiple large-scale scRNA-seq datasets. We propose OCAT, One Cell At a Time, a machine learning method that sparsely encodes single-cell gene expression to integrate data from multiple sources without highly variable gene selection or explicit batch effect correction. We demonstrate that OCAT efficiently integrates multiple scRNA-seq datasets and achieves the state-of-the-art performance in cell type clustering, especially in challenging scenarios of non-overlapping cell types. In addition, OCAT can efficaciously facilitate a variety of downstream analyses.

Scaling scRNA-seq to profile millions of cells is crucial for constructing high-resolution maps of transcriptional manifolds. Current analysis strategies, in particular dimensionality reduction and two-phase clustering, offer only limited scaling and sensitivity to define such manifolds. We introduce Metacell-2, a recursive divide-and-conquer algorithm allowing efficient decomposition of scRNA-seq datasets of any size into small and cohesive groups of cells called metacells. Metacell-2 improves outlier cell detection and rare cell type identification, as shown with human bone marrow cell atlas and mouse embryonic data. Metacell-2 is implemented over the scanpy framework for easy integration in any analysis pipeline.

Reproducibility is a significant challenge in (epi)genomic research due to the complexity of experiments composed of traditional biochemistry and informatics. Recent advances have exacerbated this as high-throughput sequencing data is generated at an unprecedented pace. Here, we report the development of a Platform for Epi-Genomic Research (PEGR), a web-based project management platform that tracks and quality controls experiments from conception to publication-ready figures, compatible with multiple assays and bioinformatic pipelines. It supports rigor and reproducibility for biochemists working at the bench, while fully supporting reproducibility and reliability for bioinformaticians through integration with the Galaxy platform.

BACKGROUND: Genome-wide association studies (GWAS) aim to correlate phenotypic changes with genotypic variation. Upon transcription, single nucleotide variants (SNVs) may alter mRNA structure, with potential impacts on transcript stability, macromolecular interactions, and translation. However, plant genomes have not been assessed for the presence of these structure-altering polymorphisms or "riboSNitches." RESULTS: We experimentally demonstrate the presence of riboSNitches in transcripts of two Arabidopsis genes, ZINC RIBBON 3 (ZR3) and COTTON GOLGI-RELATED 3 (CGR3), which are associated with continentality and temperature variation in the natural environment. These riboSNitches are also associated with differences in the abundance of their respective transcripts, implying a role in regulating the gene's expression in adaptation to local climate conditions. We then computationally predict riboSNitches transcriptome-wide in mRNAs of 879 naturally inbred Arabidopsis accessions. We characterize correlations between SNPs/riboSNitches in these accessions and 434 climate descriptors of their local environments, suggesting a role of these variants in local adaptation. We integrate this information in CLIMtools V2.0 and provide a new web resource, T-CLIM, that reveals associations between transcript abundance variation and local environmental variation. CONCLUSION: We functionally validate two plant riboSNitches and, for the first time, demonstrate riboSNitch conditionality dependent on temperature, coining the term "conditional riboSNitch." We provide the first pan-genome-wide prediction of riboSNitches in plants. We expand our previous CLIMtools web resource with riboSNitch information and with 1868 additional Arabidopsis genomes and 269 additional climate conditions, which will greatly facilitate in silico studies of natural genetic variation, its phenotypic consequences, and its role in local adaptation.

Multiplex assays of variant effect (MAVEs) are a family of methods that includes deep mutational scanning experiments on proteins and massively parallel reporter assays on gene regulatory sequences. Despite their increasing popularity, a general strategy for inferring quantitative models of genotype-phenotype maps from MAVE data is lacking. Here we introduce MAVE-NN, a neural-network-based Python package that implements a broadly applicable information-theoretic framework for learning genotype-phenotype maps-including biophysically interpretable models-from MAVE datasets. We demonstrate MAVE-NN in multiple biological contexts, and highlight the ability of our approach to deconvolve mutational effects from otherwise confounding experimental nonlinearities and noise.

The analysis of microbiome data has several technical challenges. In particular, count matrices contain a large proportion of zeros, some of which are biological, whereas others are technical. Furthermore, the measurements suffer from unequal sequencing depth, overdispersion, and data redundancy. These nuisance factors introduce substantial noise. We propose an accurate and robust method, mbDenoise, for denoising microbiome data. Assuming a zero-inflated probabilistic PCA (ZIPPCA) model, mbDenoise uses variational approximation to learn the latent structure and recovers the true abundance levels using the posterior, borrowing information across samples and taxa. mbDenoise outperforms state-of-the-art methods to extract the signal for downstream analyses.

Differential abundance analysis is at the core of statistical analysis of microbiome data. The compositional nature of microbiome sequencing data makes false positive control challenging. Here, we show that the compositional effects can be addressed by a simple, yet highly flexible and scalable, approach. The proposed method, LinDA, only requires fitting linear regression models on the centered log-ratio transformed data, and correcting the bias due to compositional effects. We show that LinDA enjoys asymptotic FDR control and can be extended to mixed-effect models for correlated microbiome data. Using simulations and real examples, we demonstrate the effectiveness of LinDA.

The advancement of highly multiplexed spatial technologies requires scalable methods that can leverage spatial information. We present MISTy, a flexible, scalable, and explainable machine learning framework for extracting relationships from any spatial omics data, from dozens to thousands of measured markers. MISTy builds multiple views focusing on different spatial or functional contexts to dissect different effects. We evaluated MISTy on in silico and breast cancer datasets measured by imaging mass cytometry and spatial transcriptomics. We estimated structural and functional interactions coming from different spatial contexts in breast cancer and demonstrated how to relate MISTy's results to clinical features.

Update of bioRxiv. 2021 Jun 11;:

BACKGROUND: Biases of DNA repair can shape the nucleotide landscape of genomes at evolutionary timescales. The molecular mechanisms of those biases are still poorly understood because it is difficult to isolate the contributions of DNA repair from those of DNA damage. RESULTS: Here, we develop a genome-wide assay whereby the same DNA lesion is repaired in different genomic contexts. We insert thousands of barcoded transposons carrying a reporter of DNA mismatch repair in the genome of mouse embryonic stem cells. Upon inducing a double-strand break between tandem repeats, a mismatch is generated if the break is repaired through single-strand annealing. The resolution of the mismatch showed a 60-80% bias in favor of the strand with the longest 3' flap. The location of the lesion in the genome and the type of mismatch had little influence on the bias. Instead, we observe a complete reversal of the bias when the longest 3' flap is moved to the opposite strand by changing the position of the double-strand break in the reporter. CONCLUSIONS: These results suggest that the processing of the double-strand break has a major influence on the repair of mismatches during a single-strand annealing.

Prime editing can induce a desired base substitution, insertion, or deletion in a target gene using reverse transcriptase after nick formation by CRISPR nickase. In this study, we develop a technology that can be used to insert or replace external bases in the target DNA sequence by linking reverse transcriptase to the Francisella novicida Cas9, which is a CRISPR-Cas9 ortholog. Using FnCas9(H969A) nickase, the targeting limitation of existing Streptococcus pyogenes Cas9 nickase [SpCas9(H840A)]-based prime editing is dramatically extended, and accurate prime editing is induced specifically for the target genes in human cell lines.

BACKGROUND: Interactions of chromatin with the nuclear lamina via lamina-associated domains (LADs) confer structural stability to the genome. The dynamics of positioning of LADs during differentiation, and how LADs impinge on developmental gene expression, remains, however, elusive. RESULTS: We examined changes in the association of lamin B1 with the genome in the first 72 h of differentiation of adipose stem cells into adipocytes. We demonstrate a repositioning of entire stand-alone LADs and of LAD edges as a prominent nuclear structural feature of early adipogenesis. Whereas adipogenic genes are released from LADs, LADs sequester downregulated or repressed genes irrelevant for the adipose lineage. However, LAD repositioning only partly concurs with gene expression changes. Differentially expressed genes in LADs, including LADs conserved throughout differentiation, reside in local euchromatic and lamin-depleted sub-domains. In these sub-domains, pre-differentiation histone modification profiles correlate with the LAD versus inter-LAD outcome of these genes during adipogenic commitment. Lastly, we link differentially expressed genes in LADs to short-range enhancers which overall co-partition with these genes in LADs versus inter-LADs during differentiation. CONCLUSIONS: We conclude that LADs are predictable structural features of adipose nuclear architecture that restrain non-adipogenic genes in a repressive environment.