SpliceWiz: interactive analysis and visualization of alternative splicing in R

Alternative splicing (AS) is a crucial mechanism for regulating gene expression and isoform diversity in eukaryotes. However, the analysis and visualization of AS events from RNA sequencing data remains challenging. Most tools require a certain level of computer literacy and the available means of visualizing AS events, such as coverage and sashimi plots, have limitations and can be misleading. To address these issues, we present SpliceWiz, an R package with an interactive Shiny interface that allows easy and efficient AS analysis and visualization at scale. A novel normalization algorithm is implemented to aggregate splicing levels within sample groups, thereby allowing group differences in splicing levels to be accurately visualized. The tool also offers downstream gene ontology enrichment analysis, highlighting ASEs belonging to functional pathways of interest. SpliceWiz is optimized for speed and efficiency and introduces a new file format for coverage data storage that is more efficient than BigWig. Alignment files are processed orders of magnitude faster than other R-based AS analysis tools and on par with command-line tools. Overall, SpliceWiz streamlines AS analysis, enabling reliable identification of functionally relevant AS events for further characterization. SpliceWiz is a Bioconductor package and is also available on GitHub (https://github.com/alexchwong/SpliceWiz).

Mapping the splicing landscape of the human immune system

Most human genes code for more than one transcript. Different ratios of transcripts of the same gene can be found in different cell types or states, indicating differential use of transcription start sites or differential splicing. Such differential transcript use (DTUs) events provide an additional layer of regulation and protein diversity. With the exceptions of PTPRC and CIITA, there are very few reported cases of DTU events in the immune system. To rigorously map DTUs between different human immune cell types, we leveraged four publicly available RNA sequencing datasets. We identified 282 DTU events between five human healthy immune cell types that appear in at least two datasets. The patterns of the DTU events were mostly cell-type-specific or lineage-specific, in the context of the five cell types tested. DTUs correlated with the expression pattern of potential regulators, namely, splicing factors and transcription factors. Of the several immune related conditions studied, only sepsis affected the splicing of more than a few genes and only in innate immune cells. Taken together, we map the DTUs landscape in human peripheral blood immune cell types, and present hundreds of genes whose transcript use changes between cell types or upon activation.

Computational approaches for detecting disease-associated alternative splicing events

Alternative splicing (AS) is a key transcriptional regulation pathway. Recent studies have shown that AS events are associated with the occurrence of complex diseases. Various computational approaches have been developed for the detection of disease-associated AS events. In this review, we first describe the metrics used for quantitative characterization of AS events. Second, we review and discuss the three types of methods for detecting disease-associated splicing events, which are differential splicing analysis, aberrant splicing detection and splicing-related network analysis. Third, to further exploit the genetic mechanism of disease-associated AS events, we describe the methods for detecting genetic variants that potentially regulate splicing. For each type of methods, we conducted experimental comparison to illustrate their performance. Finally, we discuss the limitations of these methods and point out potential ways to address them. We anticipate that this review provides a systematic understanding of computational approaches for the analysis of disease-associated splicing.

A Database of Accurate Electrophoretic Migration Patterns for Human Proteins

Native molecular weight (MW) is one of the defining features of proteins. Denaturing gel electrophoresis (SDS-PAGE) is a very popular technique for separating proteins and determining their MW. Coupled with antibody-based detection, SDS-PAGE is widely applied for protein identification and quantitation. Yet, electrophoresis is poorly reproducible and the MWs obtained are often inaccurate. This hampers antibody validation and negatively impacts the reliability of western blot data, resulting worldwide in a considerable waste of reagents and labour. We argue that, to alleviate these problems there is a need to establish a database of reference MWs measured by SDS-PAGE. Using mass spectrometry as an orthogonal detection method, we acquired electrophoretic migration patterns for approximately 10'000 human proteins in five commonly used cell lines. We applied a robust internal calibration of migration to determine accurate and reproducible molecular weights. This in turn allows merging replicates to increase accuracy, but also enables comparing different cell lines. Mining of the data obtained highlights structural factors that affect migration of distinct classes of proteins. When combined with peptide coverage, the data produced recapitulates known post-translational modifications and differential splicing and can be used to formulate hypotheses on new or poorly known processing events. The full information is freely accessible as a web resource through a user friendly graphical interface (https://pumba.dcsr.unil.ch/). We anticipate that this database will be useful to investigators worldwide for troubleshooting western blot experiments, but could also contribute to the characterization of human proteoforms.

Deep Conservation of Hid-Like RHG Gene Family Homologs in Winged Insects Revealed by "Taxon Hopping" BLAST

Together with sickle (skl), the Drosophila paralogs reaper (rpr), head involution defective (hid), and grim (RHG) control a critical switch in the induction of programmed cell death. RHG homologs have been identified in other dipteran and lepidopteran species but not beyond. Revisiting this issue with a "taxon hopping" BLAST search strategy in current genome and transcriptome resources, I detected high confidence RHG homologs in Coleoptera, Hymenoptera, Hemiptera, and Dictyoptera. Analyses of gene structure and protein sequence conservation revealed aconserved splicing pattern and highly conserved amino acid residues at both the N- and C-terminal ends that identify hid as the most ancestrally organized RHG gene family member in Drosophila. hid-like RHG homologs were also detected in mosquitoes, redefining their michelob_x (mx) genes as an expansion of derived RHG homologs. Only singleton homologs were detected in the large majority of other insect clades. Lepidopteran RHG homologs, however, stand out by producing an evolutionarily-derived splice isoform, identified in previous work, in addition to the newly detected hid-like isoform. Exceptional sequence diversification of select RHG homologs at the family- and genus-level explain their previous elusiveness in important insect genome model species like the red flour beetle Tribolium castaneum and the pea aphid Acyrthosiphon pisum. Combined, these findings expand the minimal age of the RHG gene family by about 100 million years and open new avenues for molecular cell death studies in insects.

Alternative splicing reverses the cell-intrinsic and cell-extrinsic pro-oncogenic potentials of YAP1

In addition to acting as a transcriptional co-activator, YAP1 directly mediates translocalization of the pro-oncogenic phosphatase SHP2 from the cytoplasm to nucleus. In the cytoplasm, SHP2 potentiates RAS-ERK signaling, which promotes cell proliferation and cell motility, whereas in the nucleus, it mediates gene regulation. As a result, elucidating the details of SHP2 trafficking is important for understanding its biological roles, including in cancer. YAP1 comprises multiple splicing isoforms defined in part by the presence (as in YAP1-2γ) or absence (as in YAP1-2α) of a γ-segment encoded by exon 6 that disrupts a critical leucine zipper. Although the disruptive segment is known to reduce co-activator function, it is unclear how this element impacts the physical and functional relationships between YAP1 and SHP2. To explore this question, we first demonstrated that YAP1-2γ cannot bind SHP2. Nevertheless, YAP1-2γ exhibits stronger mitogenic and motogenic activities than does YAP1-2α because the YAP1-2α-mediated delivery of SHP2 to the nucleus weakens cytoplasmic RAS-ERK signaling. However, YAP1-2γ confers less in vivo tumorigenicity than does YA1-2α by recruiting tumor-inhibitory macrophages. Mechanistically, YAP1-2γ transactivates and the YAP1-2α-SHP2 complex transrepresses the monocyte/macrophage chemoattractant CCL2 Thus, cell-intrinsic and cell-extrinsic pro-oncogenic YAP1 activities are inversely regulated by alternative splicing of exon 6. Notably, oncogenic KRAS down-regulates the SRSF3 splicing factor that prevents exon 6 skipping, thereby creating a YAP1-2α-dominant situation that supports a "cold" immune microenvironment.

A Generative Neural Network for Maximizing Fitness and Diversity of Synthetic DNA and Protein Sequences

Engineering gene and protein sequences with defined functional properties is a major goal of synthetic biology. Deep neural network models, together with gradient ascent-style optimization, show promise for sequence design. The generated sequences can however get stuck in local minima and often have low diversity. Here, we develop deep exploration networks (DENs), a class of activation-maximizing generative models, which minimize the cost of a neural network fitness predictor by gradient descent. By penalizing any two generated patterns on the basis of a similarity metric, DENs explicitly maximize sequence diversity. To avoid drifting into low-confidence regions of the predictor, we incorporate variational autoencoders to maintain the likelihood ratio of generated sequences. Using DENs, we engineered polyadenylation signals with more than 10-fold higher selection odds than the best gradient ascent-generated patterns, identified splice regulatory sequences predicted to result in highly differential splicing between cell lines, and improved on state-of-the-art results for protein design tasks.

Systematic evaluation of differential splicing tools for RNA-seq studies

Differential splicing (DS) is a post-transcriptional biological process with critical, wide-ranging effects on a plethora of cellular activities and disease processes. To date, a number of computational approaches have been developed to identify and quantify differentially spliced genes from RNA-seq data, but a comprehensive intercomparison and appraisal of these approaches is currently lacking. In this study, we systematically evaluated 10 DS analysis tools for consistency and reproducibility, precision, recall and false discovery rate, agreement upon reported differentially spliced genes and functional enrichment. The tools were selected to represent the three different methodological categories: exon-based (DEXSeq, edgeR, JunctionSeq, limma), isoform-based (cuffdiff2, DiffSplice) and event-based methods (dSpliceType, MAJIQ, rMATS, SUPPA). Overall, all the exon-based methods and two event-based methods (MAJIQ and rMATS) scored well on the selected measures. Of the 10 tools tested, the exon-based methods performed generally better than the isoform-based and event-based methods. However, overall, the different data analysis tools performed strikingly differently across different data sets or numbers of samples.

The detection of PAX8 in human upper urinary tract urothelial carcinoma

Upper urinary tract urothelial carcinomas (UUT-UCs) are defined as malignant neoplasms of the urothelium from the upper urinary tract, including renal calyces, the renal pelvis and the distal ureter. The natural attributes of UUT-UCs differ from those of bladder cancer. The aim of the present study was to investigate PAX8 expression in the normal urothelium and in urothelial carcinomas (UCs) of the upper urinary tract. Immunohistochemistry was conducted in 35 cases of renal pelvic and 30 cases of ureteral papillary UCs and the adjacent normal urothelium respectively. PAX8 mRNA expression was evaluated by RT-PCR in a different set of normal urothelial mucosa of the urinary tract and UUT-UCs. In immunohistochemical studies, the positive rates of PAX8 staining in UCs of the renal pelvis and ureter were 17% and 6.6% respectively, presenting focally positive in most cases, while the positive rates in the adjacent normal epithelia of the pelvis and ureter were 100% and 93% respectively. PAX8 mRNA was detected in all of the tumors and adjacent normal urothelial mucosa specimens of the upper urinary tract. 4 types of PAX8 isoforms, PAX8a, PAX8b, PAX8c and PAX8e, were detected in UUT-UCs in this study. As in bladder cancer, PAX8 expression was highly heterogeneous in terms of the splicing mRNA isoforms, with the different isoforms differentially expressed in the UUT-UCs. Among the 4 types of PAX8 isoforms, the PAX8e isoform was found in almost all UUT-UCs tumor tissues, but the PAX8d isoform was not detected in UUT-UCs that were different from the transcriptional splicing patterns of PAX8 in bladder cancer reported in the literature. In addition, the above 4 types of PAX8 splicing isoforms were simultaneously detected in almost all of the normal mucosal epithelia of the upper urinary tract, which was very different from that of bladder mucosa. Further studies are suggested to reveal whether or not the differences in natural attributes between UCs of the upper and lower urinary tracts are related to their PAX8 transcriptional splicing patterns.

Solubility and subcellular localization of the three Drosophila RDGC phosphatase variants are determined by acylation

Protein phosphorylation is an abundant molecular switch that regulates a multitude of cellular processes. In contrast to other subfamilies of phosphoprotein phosphatases, the PPEF subfamily is only poorly investigated. Drosophila retinal degeneration C (RDGC) constitutes the founding member of the PPEF subfamily. RDGC dephosphorylates the visual pigment rhodopsin and the ion channel TRP.However, rdgC null mutant flies exhibit rhodopsin and TRP hyperphosphorylation, altered photoreceptor physiology, and retinal degeneration. Here, we report the identification of a third RDGC protein variant and show that the three RDGC isoforms harbor different N-termini that determine solubility and subcellular targeting due to fatty acylation. Taken together, solubility and subcellular targeting of RDGC splice variants are determined by their N-termini.

Computer Analysis of Glioma Transcriptome Profiling: Alternative Splicing Events

Here we present the analysis of alternative splicing events on an example of glioblastoma cell culture samples using a set of computer tools in combination with database integration. The gene expression profiles of glioblastoma were obtained from cell culture samples of primary glioblastoma which were isolated and processed for RNA extraction. Transcriptome profiling of normal brain samples and glioblastoma were done by Illumina sequencing. The significant differentially expressed exon-level probes and their corresponding genes were identified using a combination of the splicing index method. Previous studies indicated that tumor-specific alternative splicing is important in the regulation of gene expression and corresponding protein functions during cancer development. Multiple alternative splicing transcripts have been identified as progression markers, including generalized splicing abnormalities and tumor- and stage-specific events. We used a set of computer tools which were recently applied to analysis of gene expression in laboratory animals to study differential splicing events. We found 69 transcripts that are differentially alternatively spliced. Three cancer-associated genes were considered in detail, in particular: APP (amyloid beta precursor protein), CASC4 (cancer susceptibility candidate 4) and TP53. Such alternative splicing opens new perspectives for cancer research.

Whole-transcriptome brain expression and exon-usage profiling in major depression and suicide: evidence for altered glial, endothelial and ATPase activity

Brain gene expression profiling studies of suicide and depression using oligonucleotide microarrays have often failed to distinguish these two phenotypes. Moreover, next generation sequencing approaches are more accurate in quantifying gene expression and can detect alternative splicing. Using RNA-seq, we examined whole-exome gene and exon expression in non-psychiatric controls (CON, N=29), DSM-IV major depressive disorder suicides (MDD-S, N=21) and MDD non-suicides (MDD, N=9) in the dorsal lateral prefrontal cortex (Brodmann Area 9) of sudden death medication-free individuals post mortem. Using small RNA-seq, we also examined miRNA expression (nine samples per group). DeSeq2 identified 35 genes differentially expressed between groups and surviving adjustment for false discovery rate (adjusted P<0.1). In depression, altered genes include humanin-like-8 (MTRNRL8), interleukin-8 (IL8), and serpin peptidase inhibitor, clade H (SERPINH1) and chemokine ligand 4 (CCL4), while exploratory gene ontology (GO) analyses revealed lower expression of immune-related pathways such as chemokine receptor activity, chemotaxis and cytokine biosynthesis, and angiogenesis and vascular development in (adjusted P<0.1). Hypothesis-driven GO analysis suggests lower expression of genes involved in oligodendrocyte differentiation, regulation of glutamatergic neurotransmission, and oxytocin receptor expression in both suicide and depression, and provisional evidence for altered DNA-dependent ATPase expression in suicide only. DEXSEq analysis identified differential exon usage in ATPase, class II, type 9B (adjusted P<0.1) in depression. Differences in miRNA expression or structural gene variants were not detected. Results lend further support for models in which deficits in microglial, endothelial (blood-brain barrier), ATPase activity and astrocytic cell functions contribute to MDD and suicide, and identify putative pathways and mechanisms for further study in these disorders.

QoRTs: a comprehensive toolset for quality control and data processing of RNA-Seq experiments

BACKGROUND

High-throughput next-generation RNA sequencing has matured into a viable and powerful method for detecting variations in transcript expression and regulation. Proactive quality control is of critical importance as unanticipated biases, artifacts, or errors can potentially drive false associations and lead to flawed results.

RESULTS

We have developed the Quality of RNA-Seq Toolset, or QoRTs, a comprehensive, multifunction toolset that assists in quality control and data processing of high-throughput RNA sequencing data.

CONCLUSIONS

QoRTs generates an unmatched variety of quality control metrics, and can provide cross-comparisons of replicates contrasted by batch, biological sample, or experimental condition, revealing any outliers and/or systematic issues that could drive false associations or otherwise compromise downstream analyses. In addition, QoRTs simultaneously replaces the functionality of numerous other data-processing tools, and can quickly and efficiently generate quality control metrics, coverage counts (for genes, exons, and known/novel splice-junctions), and browser tracks. These functions can all be carried out as part of a single unified data-processing/quality control run, greatly reducing both the complexity and the total runtime of the analysis pipeline. The software, source code, and documentation are available online at http://hartleys.github.io/QoRTs.

A self-enforcing CD44s/ZEB1 feedback loop maintains EMT and stemness properties in cancer cells

Invasion and metastasis of carcinomas are often activated by induction of aberrant epithelial-mesenchymal transition (EMT). This is mainly driven by the transcription factor ZEB1, promoting tumor-initiating capacity correlated with increased expression of the putative stem cell marker CD44. However, the direct link between ZEB1, CD44 and tumourigenesis is still enigmatic. Remarkably, EMT-induced repression of ESRP1 controls alternative splicing of CD44, causing a shift in the expression from the variant CD44v to the standard CD44s isoform. We analyzed whether CD44 and ZEB1 regulate each other and show that ZEB1 controls CD44s splicing by repression of ESRP1 in breast and pancreatic cancer. Intriguingly, CD44s itself activates the expression of ZEB1, resulting in a self-sustaining ZEB1 and CD44s expression. Activation of this novel CD44s-ZEB1 regulatory loop has functional impact on tumor cells, as evident by increased tumor-sphere initiation capacity, drug-resistance and tumor recurrence. In summary, we identified a self-enforcing feedback loop that employs CD44s to activate ZEB1 expression. This renders tumor cell stemness independent of external stimuli, as ZEB1 downregulates ESRP1, further promoting CD44s isoform synthesis.

Reproducible probe-level analysis of the Affymetrix Exon 1.0 ST array with R/Bioconductor

The presence of different transcripts of a gene across samples can be analysed by whole-transcriptome microarrays. Reproducing results from published microarray data represents a challenge owing to the vast amounts of data and the large variety of preprocessing and filtering steps used before the actual analysis is carried out. To guarantee a firm basis for methodological development where results with new methods are compared with previous results, it is crucial to ensure that all analyses are completely reproducible for other researchers. We here give a detailed workflow on how to perform reproducible analysis of the GeneChip®Human Exon 1.0 ST Array at probe and probeset level solely in R/Bioconductor, choosing packages based on their simplicity of use. To exemplify the use of the proposed workflow, we analyse differential splicing and differential gene expression in a publicly available dataset using various statistical methods. We believe this study will provide other researchers with an easy way of accessing gene expression data at different annotation levels and with the sufficient details needed for developing their own tools for reproducible analysis of the GeneChip®Human Exon 1.0 ST Array.

Effects of NKCC2 isoform regulation on NaCl transport in thick ascending limb and macula densa: a modeling study

This study aims to understand the extent to which modulation of the Na(+)-K(+)-2Cl(-) cotransporter NKCC2 differential splicing affects NaCl delivery to the macula densa. NaCl absorption by the thick ascending limb and macula densa cells is mediated by apical NKCC2. A recent study has indicated that differential splicing of NKCC2 is modulated by dietary salt (Schieβl IM, Rosenauer A, Kattler V, Minuth WW, Oppermann M, Castrop H. Am J Physiol Renal Physiol 305: F1139-F1148, 2013). Given the markedly different ion affinities of its splice variants, modulation of NKCC2 differential splicing is believed to impact NaCl reabsorption. To assess the validity of that hypothesis, we have developed a mathematical model of macula densa cell transport and incorporated that cell model into a previously applied model of the thick ascending limb (Weinstein AM, Krahn TA. Am J Physiol Renal Physiol 298: F525-F542, 2010). The macula densa model predicts a 27.4- and 13.1-mV depolarization of the basolateral membrane [as a surrogate for activation of tubuloglomerular feedback (TGF)] when luminal NaCl concentration is increased from 25 to 145 mM or luminal K(+) concentration is increased from 1.5 to 3.5 mM, respectively, consistent with experimental measurements. Simulations indicate that with luminal solute concentrations consistent with in vivo conditions near the macula densa, NKCC2 operates near its equilibrium state. Results also suggest that modulation of NKCC2 differential splicing by low salt, which induces a shift from NKCC2-A to NKCC2-B primarily in the cortical thick ascending limb and macula densa cells, significantly enhances salt reabsorption in the thick limb and reduces Na(+) and Cl(-) delivery to the macula densa by 3.7 and 12.5%, respectively. Simulation results also predict that the NKCC2 isoform shift hyperpolarizes the macula densa basolateral cell membrane, which, taken in isolation, may inhibit the release of the TGF signal. However, excessive early distal salt delivery and renal salt loss during a low-salt diet may be prevented by an asymmetric TGF response, which may be more sensitive to flow increases.

Negative regulation of STAT92E by an N-terminally truncated STAT protein derived from an alternative promoter site

Previously unrecognized mRNAs originating from a dual promoter at the stat92E locus are described. One of these encodes a truncated protein, DeltaNSTAT92E, that lacks the N-terminal 133 amino acids. Antibodies detect both the full-length and truncated molecules early in embryogenesis (1-5 h), and mRNA detection by specific RT-PCR reactions accords with the protein distribution. Given that the N termini of mammalian STATs are known to have positive functions in transcriptional activation, we explored the role of DeltaNSTAT92E early in embryogenesis. By increasing the DeltaNSTAT92E-to-STAT92E ratio in overexpression and RNAi experiments, we observe phenotypes compatible with suppression of wild-type STAT92E activity. We therefore conclude that the short form of STAT92E is a naturally occurring dominant-negative product that can be added to the growing list of negative regulators of STAT activity.

The 121 amino acid isoform of vascular endothelial growth factor is more strongly tumorigenic than other splice variants in vivo

Vascular endothelial growth factor (VEGF) is known to occur as at least six differentially spliced variants, giving rise to mature isoforms containing 121, 145, 165, 183, 189 and 206 amino acids. However, little is yet known concerning the in vivo activities of this differential splicing. Stably transfected MCF-7 breast carcinoma cells were constructed that secreted comparable amounts of the 121, 165 or 189 isoforms. Rabbit corneal angiogenesis assays showed the VEGF121 transfectant to have much greater angiogenic activity than the 165 or 189 expressing MCF-7 cells. While the VEGF121-expressing MCF-7 cells were reproducibly more tumorigenic than the control transfectants, this was not the case with the VEGF165- or VEGF189-expressing cells. More surprising was the observation that VEGF189 located to the nucleus, consistent with the presence of a highly conserved nuclear localization sequence in exon 6a that is expressed in VEGF189 but not 121 or 165. It was concluded that the VEGF121 isoform is both more angiogenic and tumorigenic than are the 165 and 189 isoforms. This is probably due to the ability of the 121 isoform, unlike the 165 and 189 isoforms, to freely diffuse from the cells producing it.