Kmerator Suite: design of specific k-mer signatures and automatic metadata discovery in large RNA-seq datasets.

The huge body of publicly available RNA-sequencing (RNA-seq) libraries is a treasure of functional information allowing to quantify the expression of known or novel transcripts in tissues. However, transcript quantification commonly relies on alignment methods requiring a lot of computational resources and processing time, which does not scale easily to large datasets. K-mer decomposition constitutes a new way to process RNA-seq data for the identification of transcriptional signatures, as k-mers can be used to quantify accurately gene expression in a less resource-consuming way. We present the Kmerator Suite, a set of three tools designed to extract specific k-mer signatures, quantify these k-mers into RNA-seq datasets and quickly visualize large dataset characteristics. The core tool, Kmerator, produces specific k-mers for 97% of human genes, enabling the measure of gene expression with high accuracy in simulated datasets. KmerExploR, a direct application of Kmerator, uses a set of predictor gene-specific k-mers to infer metadata including library protocol, sample features or contaminations from RNA-seq datasets. KmerExploR results are visualized through a user-friendly interface. Moreover, we demonstrate that the Kmerator Suite can be used for advanced queries targeting known or new biomarkers such as mutations, gene fusions or long non-coding RNAs for human health applications.

Long non-coding RNA exploration for mesenchymal stem cell characterisation.

BACKGROUND: The development of RNA sequencing (RNAseq) and the corresponding emergence of public datasets have created new avenues of transcriptional marker search. The long non-coding RNAs (lncRNAs) constitute an emerging class of transcripts with a potential for high tissue specificity and function. Therefore, we tested the biomarker potential of lncRNAs on Mesenchymal Stem Cells (MSCs), a complex type of adult multipotent stem cells of diverse tissue origins, that is frequently used in clinics but which is lacking extensive characterization. RESULTS: We developed a dedicated bioinformatics pipeline for the purpose of building a cell-specific catalogue of unannotated lncRNAs. The pipeline performs ab initio transcript identification, pseudoalignment and uses new methodologies such as a specific k-mer approach for naive quantification of expression in numerous RNAseq data. We next applied it on MSCs, and our pipeline was able to highlight novel lncRNAs with high cell specificity. Furthermore, with original and efficient approaches for functional prediction, we demonstrated that each candidate represents one specific state of MSCs biology. CONCLUSIONS: We showed that our approach can be employed to harness lncRNAs as cell markers. More specifically, our results suggest different candidates as potential actors in MSCs biology and propose promising directions for future experimental investigations.

Recurrent Genetic Abnormalities in Human Pluripotent Stem Cells: Definition and Routine Detection in Culture Supernatant by Targeted Droplet Digital PCR.

Genomic integrity of human pluripotent stem cells (hPSCs) is essential for research and clinical applications. However, genetic abnormalities can accumulate during hPSC generation and routine culture and following gene editing. Their occurrence should be regularly monitored, but the current assays to assess hPSC genomic integrity are not fully suitable for such regular screening. To address this issue, we first carried out a large meta-analysis of all hPSC genetic abnormalities reported in more than 100 publications and identified 738 recurrent genetic abnormalities (i.e., overlapping abnormalities found in at least five distinct scientific publications). We then developed a test based on the droplet digital PCR technology that can potentially detect more than 90% of these hPSC recurrent genetic abnormalities in DNA extracted from culture supernatant samples. This test can be used to routinely screen genomic integrity in hPSCs.

New chimeric RNAs in acute myeloid leukemia.

Background: High-throughput next generation sequencing (NGS) technologies enable the detection of biomarkers used for tumor classification, disease monitoring and cancer therapy. Whole-transcriptome analysis using RNA-seq is important, not only as a means of understanding the mechanisms responsible for complex diseases but also to efficiently identify novel genes/exons, splice isoforms, RNA editing, allele-specific mutations, differential gene expression and fusion-transcripts or chimeric RNA (chRNA). Methods: We used Crac, a tool that uses genomic locations and local coverage to classify biological events and directly infer splice and chimeric junctions within a single read. Crac's algorithm extracts transcriptional chimeric events irrespective of annotation with a high sensitivity, and CracTools was used to aggregate, annotate and filter the chRNA reads. The selected chRNA candidates were validated by real time PCR and sequencing.  In order to check the tumor specific expression of chRNA, we analyzed a publicly available dataset using a new tag search approach. Results:  We present data related to acute myeloid leukemia (AML) RNA-seq analysis. We highlight novel biological cases of chRNA, in addition to previously well characterized leukemia chRNA. We have identified and validated 17 chRNAs among 3 AML patients: 10 from an AML patient with a translocation between chromosomes 15 and 17 (AML-t(15;17), 4  from patient with normal karyotype (AML-NK) 3 from a patient with chromosomal 16 inversion (AML-inv16). The new fusion transcripts can be classified into four groups according to the exon organization. Conclusions:  All groups suggest complex but distinct synthesis mechanisms involving either collinear exons of different genes, non-collinear exons, or exons of different chromosomes. Finally, we check tumor-specific expression in a larger RNA-seq AML cohort and identify new AML biomarkers that could improve diagnosis and prognosis of AML.

On the evaluation of the fidelity of supervised classifiers in the prediction of chimeric RNAs.

BACKGROUND: High-throughput sequencing technology and bioinformatics have identified chimeric RNAs (chRNAs), raising the possibility of chRNAs expressing particularly in diseases can be used as potential biomarkers in both diagnosis and prognosis. RESULTS: The task of discriminating true chRNAs from the false ones poses an interesting Machine Learning (ML) challenge. First of all, the sequencing data may contain false reads due to technical artifacts and during the analysis process, bioinformatics tools may generate false positives due to methodological biases. Moreover, if we succeed to have a proper set of observations (enough sequencing data) about true chRNAs, chances are that the devised model can not be able to generalize beyond it. Like any other machine learning problem, the first big issue is finding the good data to build models. As far as we were concerned, there is no common benchmark data available for chRNAs detection. The definition of a classification baseline is lacking in the related literature too. In this work we are moving towards benchmark data and an evaluation of the fidelity of supervised classifiers in the prediction of chRNAs. CONCLUSIONS: We proposed a modelization strategy that can be used to increase the tools performances in context of chRNA classification based on a simulated data generator, that permit to continuously integrate new complex chimeric events. The pipeline incorporated a genome mutation process and simulated RNA-seq data. The reads within distinct depth were aligned and analysed by CRAC that integrates genomic location and local coverage, allowing biological predictions at the read scale. Additionally, these reads were functionally annotated and aggregated to form chRNAs events, making it possible to evaluate ML methods (classifiers) performance in both levels of reads and events. Ensemble learning strategies demonstrated to be more robust to this classification problem, providing an average AUC performance of 95 % (ACC=94 %, Kappa=0.87 %). The resulting classification models were also tested on real RNA-seq data from a set of twenty-seven patients with acute myeloid leukemia (AML).

Long non-coding RNAs in human early embryonic development and their potential in ART.

BACKGROUND: Human long non-coding RNAs (lncRNAs) are an emerging category of transcripts with increasingly documented functional roles during development. LncRNAs and roles during human early embryo development have recently begun to be unravelled. OBJECTIVE AND RATIONALE: This review summarizes the most recent knowledge on lncRNAs and focuses on their expression patterns and role during early human embryo development and in pluripotent stem cells (PSCs). Public mRNA sequencing (mRNA-seq) data were used to illustrate these expression signatures. SEARCH METHODS: The PubMed and EMBASE databases were first interrogated using specific terms, such as 'lncRNAs', to get an extensive overview on lncRNAs up to February 2016, and then using 'human lncRNAs' and 'embryo', 'development', or 'PSCs' to focus on lncRNAs involved in human embryo development or in PSC.Recently published RNA-seq data from human oocytes and pre-implantation embryos (including single-cell data), PSC and a panel of normal and malignant adult tissues were used to describe the specific expression patterns of some lncRNAs in early human embryos. OUTCOMES: The existence and the crucial role of lncRNAs in many important biological phenomena in each branch of the life tree are now well documented. The number of identified lncRNAs is rapidly increasing and has already outnumbered that of protein-coding genes. Unlike small non-coding RNAs, a variety of mechanisms of action have been proposed for lncRNAs. The functional role of lncRNAs has been demonstrated in many biological and developmental processes, including cell pluripotency induction, X-inactivation or gene imprinting. Analysis of RNA-seq data highlights that lncRNA abundance changes significantly during human early embryonic development. This suggests that lncRNAs could represent candidate biomarkers for developing non-invasive tests for oocyte or embryo quality. Finally, some of these lncRNAs are also expressed in human cancer tissues, suggesting that reactivation of an embryonic lncRNA program may contribute to human malignancies. WIDER IMPLICATIONS: LncRNAs are emerging potential key players in gene expression regulation. Analysis of RNA-seq data from human pre-implantation embryos identified lncRNA signatures that are specific to this critical step. We anticipate that further studies will show that these new transcripts are major regulators of embryo development. These findings might also be used to develop new tests/treatments for improving the pregnancy success rate in IVF procedures or for regenerative medicine applications involving PSC.

Combining DGE and RNA-sequencing data to identify new polyA+ non-coding transcripts in the human genome.

Recent sequencing technologies that allow massive parallel production of short reads are the method of choice for transcriptome analysis. Particularly, digital gene expression (DGE) technologies produce a large dynamic range of expression data by generating short tag signatures for each cell transcript. These tags can be mapped back to a reference genome to identify new transcribed regions that can be further covered by RNA-sequencing (RNA-Seq) reads. Here, we applied an integrated bioinformatics approach that combines DGE tags, RNA-Seq, tiling array expression data and species-comparison to explore new transcriptional regions and their specific biological features, particularly tissue expression or conservation. We analysed tags from a large DGE data set (designated as 'TranscriRef'). We then annotated 750,000 tags that were uniquely mapped to the human genome according to Ensembl. We retained transcripts originating from both DNA strands and categorized tags corresponding to protein-coding genes, antisense, intronic- or intergenic-transcribed regions and computed their overlap with annotated non-coding transcripts. Using this bioinformatics approach, we identified ∼34,000 novel transcribed regions located outside the boundaries of known protein-coding genes. As demonstrated using sequencing data from human pluripotent stem cells for biological validation, the method could be easily applied for the selection of tissue-specific candidate transcripts. DigitagCT is available at http://cractools.gforge.inria.fr/softwares/digitagct.

New prognostic markers, determined using gene expression analyses, reveal two distinct subtypes of chronic myelomonocytic leukaemia patients.

Chronic myelomonocytic leukaemia (CMML) is a heterogeneous haematopoietic disorder characterized by myeloproliferative or myelodysplastic features. At present, the pathogenesis of this malignancy is not completely understood. In this study, we sought to analyse gene expression profiles of CMML in order to characterize new molecular outcome predictors. A learning set of 32 untreated CMML patients at diagnosis was available for TaqMan low-density array gene expression analysis. From 93 selected genes related to cancer and cell cycle, we built a five-gene prognostic index after multiplicity correction. Using this index, we characterized two categories of patients with distinct overall survival (94% vs. 19% for good and poor overall survival, respectively; P = 0·007) and we successfully validated its strength on an independent cohort of 21 CMML patients with Affymetrix gene expression data. We found no specific patterns of association with traditional prognostic stratification parameters in the learning cohort. However, the poor survival group strongly correlated with high-risk treated patients and transformation to acute myeloid leukaemia. We report here a new multigene prognostic index for CMML, independent of the gene expression measurement method, which could be used as a powerful tool to predict clinical outcome and help physicians to evaluate criteria for treatments.

Maternal mRNA deadenylation and decay by the piRNA pathway in the early Drosophila embryo.

Piwi-associated RNAs (piRNAs), a specific class of 24- to 30-nucleotide-long RNAs produced by the Piwi-type of Argonaute proteins, have a specific germline function in repressing transposable elements. This repression is thought to involve heterochromatin formation and transcriptional and post-transcriptional silencing. The piRNA pathway has other essential functions in germline stem cell maintenance and in maintaining germline DNA integrity. Here we uncover an unexpected function of the piRNA pathway in the decay of maternal messenger RNAs and in translational repression in the early embryo. A subset of maternal mRNAs is degraded in the embryo at the maternal-to-zygotic transition. In Drosophila, maternal mRNA degradation depends on the RNA-binding protein Smaug and the deadenylase CCR4, as well as the zygotic expression of a microRNA cluster. Using mRNA encoding the embryonic posterior morphogen Nanos (Nos) as a paradigm to study maternal mRNA decay, we found that CCR4-mediated deadenylation of nos depends on components of the piRNA pathway including piRNAs complementary to a specific region in the nos 3' untranslated region. Reduced deadenylation when piRNA-induced regulation is impaired correlates with nos mRNA stabilization and translational derepression in the embryo, resulting in head development defects. Aubergine, one of the Argonaute proteins in the piRNA pathway, is present in a complex with Smaug, CCR4, nos mRNA and piRNAs that target the nos 3' untranslated region, in the bulk of the embryo. We propose that piRNAs and their associated proteins act together with Smaug to recruit the CCR4 deadenylation complex to specific mRNAs, thus promoting their decay. Because the piRNAs involved in this regulation are produced from transposable elements, this identifies a direct developmental function for transposable elements in the regulation of gene expression.

Using reads to annotate the genome: influence of length, background distribution, and sequence errors on prediction capacity.

Ultra high-throughput sequencing is used to analyse the transcriptome or interactome at unprecedented depth on a genome-wide scale. These techniques yield short sequence reads that are then mapped on a genome sequence to predict putatively transcribed or protein-interacting regions. We argue that factors such as background distribution, sequence errors, and read length impact on the prediction capacity of sequence census experiments. Here we suggest a computational approach to measure these factors and analyse their influence on both transcriptomic and epigenomic assays. This investigation provides new clues on both methodological and biological issues. For instance, by analysing chromatin immunoprecipitation read sets, we estimate that 4.6% of reads are affected by SNPs. We show that, although the nucleotide error probability is low, it significantly increases with the position in the sequence. Choosing a read length above 19 bp practically eliminates the risk of finding irrelevant positions, while above 20 bp the number of uniquely mapped reads decreases. With our procedure, we obtain 0.6% false positives among genomic locations. Hence, even rare signatures should identify biologically relevant regions, if they are mapped on the genome. This indicates that digital transcriptomics may help to characterize the wealth of yet undiscovered, low-abundance transcripts.

Transcriptome annotation using tandem SAGE tags.

Analysis of several million expressed gene signatures (tags) revealed an increasing number of different sequences, largely exceeding that of annotated genes in mammalian genomes. Serial analysis of gene expression (SAGE) can reveal new Poly(A) RNAs transcribed from previously unrecognized chromosomal regions. However, conventional SAGE tags are too short to identify unambiguously unique sites in large genomes. Here, we design a novel strategy with tags anchored on two different restrictions sites of cDNAs. New transcripts are then tentatively defined by the two SAGE tags in tandem and by the spanning sequence read on the genome between these tagged sites. Having developed a new algorithm to locate these tag-delimited genomic sequences (TDGS), we first validated its capacity to recognize known genes and its ability to reveal new transcripts with two SAGE libraries built in parallel from a single RNA sample. Our algorithm proves fast enough to experiment this strategy at a large scale. We then collected and processed the complete sets of human SAGE tags to predict yet unknown transcripts. A cross-validation with tiling arrays data shows that 47% of these TDGS overlap transcriptional active regions. Our method provides a new and complementary approach for complex transcriptome annotation.

Variability and expression of ankyrin domain genes in Wolbachia variants infecting the mosquito Culex pipiens.

Wolbachia strains are maternally inherited endosymbiotic bacteria that infect many arthropod species and have evolved several different ways of manipulating their hosts, the most frequent way being cytoplasmic incompatibility (CI). CI leads to embryo death in crosses between infected males and uninfected females as well as in crosses between individuals infected by incompatible Wolbachia strains. The mosquito Culex pipiens exhibits the highest crossing type variability reported so far. Our crossing data support the notion that CI might be driven by at least two distinct genetic units that control the CI functions independently in males and females. Although the molecular basis of CI remains unknown, proteins with ankyrin (ANK) domains represent promising candidates since they might interact with a wide range of host proteins. Here we searched for sequence variability in the 58 ANK genes carried in the genomes of Wolbachia variants infecting Culex pipiens. Only five ANK genes were polymorphic in the genomes of incompatible Wolbachia variants, and none correlated with the CI pattern obtained with 15 mosquito strains (representing 14 Wolbachia variants). Further analysis of ANK gene expression evidenced host- and sex-dependent variations, which did not improve the correlation. Taken together, these data do not support the direct implication of ANK genes in CI determinism.

Evolution of the Rho family of ras-like GTPases in eukaryotes.

GTPases of the Rho family are molecular switches that play important roles in converting and amplifying external signals into cellular effects. Originally demonstrated to control the dynamics of the F-actin cytoskeleton, Rho GTPases have been implicated in many basic cellular processes that influence cell proliferation, differentiation, motility, adhesion, survival, or secretion. To elucidate the evolutionary history of the Rho family, we have analyzed over 20 species covering major eukaryotic clades from unicellular organisms to mammals, including platypus and opossum, and have reconstructed the ontogeny and the chronology of emergence of the different subfamilies. Our data establish that the 20 mammalian Rho members are structured into 8 subfamilies, among which Rac is the founder of the whole family. Rho, Cdc42, RhoUV, and RhoBTB subfamilies appeared before Coelomates and RhoJQ, Cdc42 isoforms, RhoDF, and Rnd emerged in chordates. In vertebrates, gene duplications and retrotranspositions increased the size of each chordate Rho subfamily, whereas RhoH, the last subfamily, arose probably by horizontal gene transfer. Rac1b, a Rac1 isoform generated by alternative splicing, emerged in amniotes, and RhoD, only in therians. Analysis of Rho mRNA expression patterns in mouse tissues shows that recent subfamilies have tissue-specific and low-level expression that supports their implication only in narrow time windows or in differentiated metabolic functions. These findings give a comprehensive view of the evolutionary canvas of the Rho family and provide guides for future structure and evolution studies of other components of Rho signaling pathways, in particular regulators of the RhoGEF family.

Two distinct pools of Src family tyrosine kinases regulate PDGF-induced DNA synthesis and actin dorsal ruffles.

The mechanism by which the Src family of protein-tyrosine kinases (SFKs) regulate mitogenesis and morphological changes induced by platelet-derived growth factor (PDGF) is not well known. The cholesterol-enriched membrane microdomains, caveolae, regulate PDGF receptor signalling in fibroblasts and we examined their role in SFK functions. Here we show that caveolae disruption by membrane cholesterol depletion or expression of the dominant-negative caveolin-3 DGV mutant impaired Src mitogenic signalling including kinase activation, Myc gene induction and DNA synthesis. The impact of caveolae on SFK function was underscored by the capacity of Myc to overcome mitogenic inhibition as a result of caveolae disruption. Using biochemical fractionation we show that caveolae-enriched subcellular membranes regulate the formation of PDGF-receptor-SFK complexes. An additional pool of PDGF-activated SFKs that was insensitive to membrane cholesterol depletion was characterised in non-caveolae fractions. SFK activation outside caveolae was linked to the capacity of PDGF to induce F-actin rearrangements leading to dorsal ruffle formation. Inhibition of phospholipase C gamma (PLCgamma), sphingosine kinase and heterotrimeric Gi proteins implicates a PLC gamma-sphingosine-1-phosphate-Gi pathway for PDGF-induced SFK activation outside caveolae and actin assembly. In addition, the cytoplasmic tyrosine kinase Abl was identified as an important effector of this signalling cascade. We conclude that PDGF may stimulate two spatially distinct pools of SFKs leading to two different biological outcomes: DNA synthesis and dorsal ruffle formation.

Abl tyrosine kinase regulates a Rac/JNK and a Rac/Nox pathway for DNA synthesis and Myc expression induced by growth factors.

The cytoplasmic tyrosine kinase Abl is a Src substrate required for platelet-derived growth factor (PDGF) receptor signaling leading to Myc expression and DNA synthesis. Abl targets are, however, ill defined. Here we report that the small GTPase Rac is an important effector of its mitogenic function. PDGF-induced Rac activation was impaired in cells with inactive Abl and active Rac overcame the mitogenic defects found in these cells. Rac function required both a Jun N-terminal kinase (JNK) and a NADPH oxidase (Nox) pathway. Furthermore, co-activation of JNK and Nox were sufficient to mimic the Rac mitogenic rescue. Abl also regulated PDGF-induced JNK and Nox activation. Finally, we found that Myc is an important target of this signaling cascade: Myc induction was sensitive to small inhibitors of JNK and Nox activities and forced expression of Myc overcame the G1 block induced by dominant interfering mutants of mitogen-activated protein kinase kinase 4 (MKK4) and Nox2 activating subunit. We concluded that cytoplasmic Abl operates on a Rac/JNK and a Rac/Nox pathway for PDGF-induced Myc induction and DNA synthesis.