Barbara J. Wold

Mapping and quantifying mammalian transcriptomes by RNA-Seq.

We have mapped and quantified mouse transcriptomes by deeply sequencing them and recording how frequently each gene is represented in the sequence sample (RNA-Seq). This provides a digital measure of the presence and prevalence of transcripts from known and previously unknown genes. We report reference measurements composed of 41–52 million mapped 25-base-pair reads for poly(A)-selected RNA from adult mouse brain, liver and skeletal muscle tissues. We used RNA standards to quantify transcript prevalence and to test the linear range of transcript detection, which spanned five orders of magnitude. Although >90% of uniquely mapped reads fell within known exons, the remaining data suggest new and revised gene models, including changed or additional promoters, exons and 3′ untranscribed regions, as well as new candidate microRNA precursors. RNA splice events, which are not readily measured by standard gene expression microarray or serial analysis of gene expression methods, were detected directly by mapping splice-crossing sequence reads. We observed 1.45 × 105 distinct splices, and alternative splices were prominent, with 3,500 different genes expressing one or more alternate internal splices.

Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation

High-throughput mRNA sequencing (RNA-Seq) promises simultaneous transcript discovery and abundance estimation. However, this would require algorithms that are not restricted by prior gene annotations and that account for alternative transcription and splicing. Here we introduce such algorithms in an open-source software program called Cufflinks. To test Cufflinks, we sequenced and analyzed >430 million paired 75-bp RNA-Seq reads from a mouse myoblast cell line over a differentiation time series. We detected 13,692 known transcripts and 3,724 previously unannotated ones, 62% of which are supported by independent expression data or by homologous genes in other species. Over the time series, 330 genes showed complete switches in the dominant transcription start site (TSS) or splice isoform, and we observed more subtle shifts in 1,304 other genes. These results suggest that Cufflinks can illuminate the substantial regulatory flexibility and complexity in even this well-studied model of muscle development and that it can improve transcriptome-based genome annotation.High-throughput mRNA sequencing (RNA-Seq) promises simultaneous transcript discovery and abundance estimation. However, this would require algorithms that are not restricted by prior gene annotations and that account for alternative transcription and splicing. Here we introduce such algorithms in an open-source software program called Cufflinks. To test Cufflinks, we sequenced and analyzed >430 million paired 75-bp RNA-Seq reads from a mouse myoblast cell line over a differentiation time series. We detected 13,692 known transcripts and 3,724 previously unannotated ones, 62% of which are supported by independent expression data or by homologous genes in other species. Over the time series, 330 genes showed complete switches in the dominant transcription start site (TSS) or splice isoform, and we observed more subtle shifts in 1,304 other genes. These results suggest that Cufflinks can illuminate the substantial regulatory flexibility and complexity in even this well-studied model of muscle development and that it can improve transcriptome-based genome annotation.

Landscape of transcription in human cells

Eukaryotic cells make many types of primary and processed RNAs that are found either in specific subcellular compartments or throughout the cells. A complete catalogue of these RNAs is not yet available and their characteristic subcellular localizations are also poorly understood. Because RNA represents the direct output of the genetic information encoded by genomes and a significant proportion of a cell’s regulatory capabilities are focused on its synthesis, processing, transport, modification and translation, the generation of such a catalogue is crucial for understanding genome function. Here we report evidence that three-quarters of the human genome is capable of being transcribed, as well as observations about the range and levels of expression, localization, processing fates, regulatory regions and modifications of almost all currently annotated and thousands of previously unannotated RNAs. These observations, taken together, prompt a redefinition of the concept of a gene.

Genome-wide mapping of in vivo protein-DNA interactions.

In vivo protein-DNA interactions connect each transcription factor with its direct targets to form a gene network scaffold. To map these protein-DNA interactions comprehensively across entire mammalian genomes, we developed a large-scale chromatin immunoprecipitation assay (ChIPSeq) based on direct ultrahigh-throughput DNA sequencing. This sequence census method was then used to map in vivo binding of the neuron-restrictive silencer factor (NRSF; also known as REST, for repressor element–1 silencing transcription factor) to 1946 locations in the human genome. The data display sharp resolution of binding position [±50 base pairs (bp)], which facilitated our finding motifs and allowed us to identify noncanonical NRSF-binding motifs. These ChIPSeq data also have high sensitivity and specificity [ROC (receiver operator characteristic) area ≥ 0.96] and statistical confidence (P <10–4), properties that were important for inferring new candidate interactions. These include key transcription factors in the gene network that regulates pancreatic islet cell development.

ChIP-seq guidelines and practices of the ENCODE and modENCODE consortia

Chromatin immunoprecipitation (ChIP) followed by high-throughput DNA sequencing (ChIP-seq) has become a valuable and widely used approach for mapping the genomic location of transcription-factor binding and histone modifications in living cells. Despite its widespread use, there are considerable differences in how these experiments are conducted, how the results are scored and evaluated for quality, and how the data and metadata are archived for public use. These practices affect the quality and utility of any global ChIP experiment. Through our experience in performing ChIP-seq experiments, the ENCODE and modENCODE consortia have developed a set of working standards and guidelines for ChIP experiments that are updated routinely. The current guidelines address antibody validation, experimental replication, sequencing depth, data and metadata reporting, and data quality assessment. We discuss how ChIP quality, assessed in these ways, affects different uses of ChIP-seq data. All data sets used in the analysis have been deposited for public viewing and downloading at the ENCODE (http://encodeproject.org/ENCODE/) and modENCODE (http://www.modencode.org/) portals.

Extensive Promoter-centered Chromatin Interactions Provide a Topological Basis for Transcription Regulation

Higher-order chromosomal organization for transcription regulation is poorly understood in eukaryotes. Using genome-wide Chromatin Interaction Analysis with Paired-End-Tag sequencing (ChIA-PET), we mapped long-range chromatin interactions associated with RNA polymerase II in human cells and uncovered widespread promoter-centered intragenic, extragenic, and intergenic interactions. These interactions further aggregated into higher-order clusters, wherein proximal and distal genes were engaged through promoter-promoter interactions. Most genes with promoter-promoter interactions were active and transcribed cooperatively, and some interacting promoters could influence each other implying combinatorial complexity of transcriptional controls. Comparative analyses of different cell lines showed that cell-specific chromatin interactions could provide structural frameworks for cell-specific transcription, and suggested significant enrichment of enhancer-promoter interactions for cell-specific functions. Furthermore, genetically-identified disease-associated noncoding elements were found to be spatially engaged with corresponding genes through long-range interactions. Overall, our study provides insights into transcription regulation by three-dimensional chromatin interactions for both housekeeping and cell-specific genes in human cells.

Know Your Neighbors: Three Phenotypes in Null Mutants of the Myogenic bHLH Gene MRF4

The data from the MRF4, Hox, and globin studies suggest that many other knockout experiments may be subject to similar effects. The examples we have discussed involve loci that contain two or more related genes, and these may be especially prone to the evolution of joint locus control elements and other interrelated cis-regulatory interactions. It is equally plausible that our current picture is skewed toward these cases by investigator knowledge of the identity of neighboring genes and that similar long-range effects will be seen for unrelated adjacent genes. We therefore feel that until substantially more is understood about cis-regulation over substantial regions of the genome it is prudent to consider these effects in design and in interpretation. One might think that direct complementation testing would clarify which knockouts are subject to neighborhood complications, but in the mouse this solution is not straightforward. It has proved frustratingly difficult to capture and reintroduce into mice pieces of DNA that faithfully recapitulate the full developmental pattern and correct level of expression of the wildtype gene. The culprits appear to be the very same aspects of gene structure that are at work in generating the neighborhood problem: the dispersal of pertinent regulatory elements over very large stretches of DNA that may also include additional genes. Moreover, minigenes that might be induced to express at will complementing protein coding sequences, in the manner of heat shock regulated constructions in Drosophila, are unreliable in current mouse transgenic formats. For these reasons the application of improved, if somewhat more complex, knockout strategies provides a more immediate solution to neighborhood uncertainties. To avoid deleting sequences that may have regulatory effects on adjacent genes, one can instead introduce an effectively positioned stop codon. This can be coupled with removal of the selection cassette by site-specific recombinases such as yeast Flp or phage P1 Cre. Alternatively, the so-called “hit-and-run” strategy, which provides for simultaneous introduction of subtle nonsense or missense mutations together with the elimination of all selection cassette residue (Ramirez-Solis et al. 1993xRamirez-Solis, R, Zheng, II, Whiting, J, Krumlauf, R, and Bradley, A. Cell. 1993; 73: 279–294Abstract | Full Text PDF | PubMed | Scopus (227)See all ReferencesRamirez-Solis et al. 1993), can be used. This design achieves the cleanest introduction of small and specific mutations, but calls for somewhat greater effort in establishing and identifying the desired ES cell line. Finally, for many existing knockouts, an immediate challenge is to consider their phenotypes with attention to the identities and activities of neighboring genes.

Computation for ChIP-seq and RNA-seq studies

Genome-wide measurements of protein-DNA interactions and transcriptomes are increasingly done by deep DNA sequencing methods (ChIP-seq and RNA-seq). The power and richness of these counting-based measurements comes at the cost of routinely handling tens to hundreds of millions of reads. Whereas early adopters necessarily developed their own custom computer code to analyze the first ChIP-seq and RNA-seq datasets, a new generation of more sophisticated algorithms and software tools are emerging to assist in the analysis phase of these projects. Here we describe the multilayered analyses of ChIP-seq and RNA-seq datasets, discuss the software packages currently available to perform tasks at each layer and describe some upcoming challenges and features for future analysis tools. We also discuss how software choices and uses are affected by specific aspects of the underlying biology and data structure, including genome size, positional clustering of transcription factor binding sites, transcript discovery and expression quantification.

Culture in reduced levels of oxygen promotes clonogenic sympathoadrenal differentiation by isolated neural crest stem cells.

Isolated neural crest stem cells (NCSCs) differentiate to autonomic neurons in response to bone morphogenetic protein 2 (BMP2) in clonal cultures, but these neurons do not express sympathoadrenal (SA) lineage markers. Whether this reflects a developmental restriction in NCSCs or simply inappropriate culture conditions was not clear. We tested the growth and differentiation potential of NCSCs at ∼5% O2, which more closely approximates physiological oxygen levels. Eighty-three percent of p75+P0− cells isolated from embryonic day 14.5 sciatic nerve behaved as stem cells under these conditions, suggesting that this is a nearly pure population. Furthermore, addition of BMP2 plus forskolin in decreased oxygen cultures elicited differentiation of thousands of cells expressing tyrosine hydroxylase, dopamine-β-hydroxylase, and the SA lineage marker SA-1 in nearly all colonies. Such cells also synthesized and released dopamine and norepinephrine. These data demonstrate that isolated mammalian NCSCs uniformly possess SA lineage capacity and further suggest that oxygen levels can influence cell fate. Parallel results indicating that reduced oxygen levels can also promote the survival, proliferation, and catecholaminergic differentiation of CNS stem cells (Studer et al., 2000) suggests that neural stem cells may exhibit a conserved response to reduced oxygen levels.

Dynamic DNA methylation across diverse human cell lines and tissues

As studies of DNA methylation increase in scope, it has become evident that methylation has a complex relationship with gene expression, plays an important role in defining cell types, and is disrupted in many diseases. We describe large-scale single-base resolution DNA methylation profiling on a diverse collection of 82 human cell lines and tissues using reduced representation bisulfite sequencing (RRBS). Analysis integrating RNA-seq and ChIP-seq data illuminates the functional role of this dynamic mark. Loci that are hypermethylated across cancer types are enriched for sites bound by NANOG in embryonic stem cells, which supports and expands the model of a stem/progenitor cell signature in cancer. CpGs that are hypomethylated across cancer types are concentrated in megabase-scale domains that occur near the telomeres and centromeres of chromosomes, are depleted of genes, and are enriched for cancer-specific EZH2 binding and H3K27me3 (repressive chromatin). In noncancer samples, there are cell-type specific methylation signatures preserved in primary cell lines and tissues as well as methylation differences induced by cell culture. The relationship between methylation and expression is context-dependent, and we find that CpG-rich enhancers bound by EP300 in the bodies of expressed genes are unmethylated despite the dense gene-body methylation surrounding them. Non-CpG cytosine methylation occurs in human somatic tissue, is particularly prevalent in brain tissue, and is reproducible across many individuals. This study provides an atlas of DNA methylation across diverse and well-characterized samples and enables new discoveries about DNA methylation and its role in gene regulation and disease.