Academic literature on the topic 'RNA-Seq expression levels'

Estimating cell type compositions for complex diseases is an important step to investigate the cellular heterogeneity for understanding disease etiology and potentially facilitate early disease diagnosis and prevention. Here, we developed a computationally statistical method, referring to Multi-Omics Matrix Factorization (MOMF), to estimate the cell-type compositions of bulk RNA sequencing (RNA-seq) data by leveraging cell type-specific gene expression levels from single-cell RNA sequencing (scRNA-seq) data. MOMF not only directly models the count nature of gene expression data, but also effectively accounts for the uncertainty of cell type-specific mean gene expression levels. We demonstrate the benefits of MOMF through three real data applications, i.e., Glioblastomas (GBM), colorectal cancer (CRC) and type II diabetes (T2D) studies. MOMF is able to accurately estimate disease-related cell type proportions, i.e., oligodendrocyte progenitor cells and macrophage cells, which are strongly associated with the survival of GBM and CRC, respectively.

Antisense transcription is known to have a range of impacts on sense gene expression, including (but not limited to) impeding transcription initiation, disrupting post-transcriptional processes, and enhancing, slowing, or even preventing transcription of the sense gene. Strand-specific RNA-Seq protocols preserve the strand information of the original RNA in the data, and so can be used to identify where antisense transcription may be implicated in regulating gene expression. However, our analysis of 199 strand-specific RNA-Seq experiments reveals that spurious antisense reads are often present in these datasets at levels greater than 1% of sense gene expression levels. Furthermore, these levels can vary substantially even between replicates in the same experiment, potentially disrupting any downstream analysis, if the incorrectly assigned antisense counts dominate the set of genes with high antisense transcription levels. Currently, no tools exist to detect or correct for this spurious antisense signal. Our tool, RoSA (Removal of Spurious Antisense), detects the presence of high levels of spurious antisense read alignments in strand-specific RNA-Seq datasets. It uses incorrectly spliced reads on the antisense strand and/or ERCC spikeins (if present in the data) to calculate both global and gene-specific antisense correction factors. We demonstrate the utility of our tool to filter out spurious antisense transcript counts in an Arabidopsis thaliana RNA-Seq experiment. Availability: RoSA is open source software available under the GPL licence via the Barton Group GitHub page https://github.com/bartongroup.

Xenopus is an important model organism for the study of genome duplication in vertebrates. With the full genome sequence of diploid Xenopus tropicalis available, and that of allotetraploid X. laevis close to being finished, we will be able to expand our understanding of how duplicated genes have evolved. One of the key features in the study of the functional consequence of gene duplication is how their expression patterns vary across different conditions, and RNA-seq seems to have enough resolution to discriminate the expression of highly similar duplicated genes. However, most of the current RNA-seq analysis methods were not designed to study samples with duplicate genes such as in X. laevis. Here, various computational methods to quantify gene expression in RNA-seq data were evaluated, using 2 independent X. laevis egg RNA-seq datasets and 2 reference databases for duplicated genes. The fact that RNA-seq can measure expression levels of similar duplicated genes was confirmed, but long paired-end reads are more informative than short single-end reads to discriminate duplicated genes. Also, it was found that bowtie, one of the most popular mappers in RNA-seq analysis, reports significantly smaller numbers of unique hits according to a mapping quality score compared to other mappers tested (BWA, GSNAP, STAR). Calculated from unique hits based on a mapping quality score, both expression levels and the expression ratio of duplicated genes can be estimated consistently among biological replicates, demonstrating that this method can successfully discriminate the expression of each copy of a duplicated gene pair. This comprehensive evaluation will be a useful guideline for studying gene expression of organisms with genome duplication using RNA-seq in the future.

RNA sequencing (RNA-seq) is a powerful approach for measuring gene expression levels in cells and tissues, but it relies on high-quality RNA. We demonstrate here that statistical adjustment using existing quality measures largely fails to remove the effects of RNA degradation when RNA quality associates with the outcome of interest. Using RNA-seq data from molecular degradation experiments of human primary tissues, we introduce a method—quality surrogate variable analysis (qSVA)—as a framework for estimating and removing the confounding effect of RNA quality in differential expression analysis. We show that this approach results in greatly improved replication rates (>3×) across two large independent postmortem human brain studies of schizophrenia and also removes potential RNA quality biases in earlier published work that compared expression levels of different brain regions and other diagnostic groups. Our approach can therefore improve the interpretation of differential expression analysis of transcriptomic data from human tissue.

Abstract Motivation In the analysis of RNA-Seq data, detecting differentially expressed (DE) genes has been a hot research area in recent years and many methods have been proposed. DE genes show different average expression levels in different sample groups, and thus can be important biological markers. While generally very successful, these methods need to be further tailored and improved for cancerous data, which often features quite diverse expression in the samples from the cancer group, and this diversity is much larger than that in the control group. Results We propose a statistical method that can detect not only genes that show different average expressions, but also genes that show different diversities of expressions in different groups. These ‘differentially dispersed’ genes can be important clinical markers. Our method uses a redescending penalty on the quasi-likelihood function, and thus has superior robustness against outliers and other noise. Simulations and real data analysis demonstrate that DiPhiSeq outperforms existing methods in the presence of outliers, and identifies unique sets of genes. Availability and implementation DiPhiSeq is publicly available as an R package on CRAN: https://cran.r-project.org/package=DiPhiSeq. Supplementary information Supplementary data are available at Bioinformatics online.

3566 Background: Clinical biomarker studies are often hindered by the availability of tissue specimens of sufficient quality and quantity. While RNA-Seq is often considered the gold standard for measuring mRNA expression levels in cancer tissue, it typically requires multiple formalin-fixed paraffin-embedded (FFPE) tissue sections to extract a sufficient amount of quality RNA for subsequent gene expression profiling analysis. The HTG EdgeSeq technology is a gene expression profiling platform that combines quantitative nuclease protection assay technology with next-generation sequencing detection. Unlike RNA-Seq, the HTG EdgeSeq technology does not require RNA extraction, and can use small amounts of tissue material, typically several mm2, to generate reproducible gene expression profiles. Methods: This study compares the performance of RNA-Seq and HTG's profiling panel, the HTG EdgeSeq Precision Immuno-Oncology Panel (PIP), which is designed to measure expression levels of 1,392 genes focused on tumor/immune interaction. Approximately 1,200 samples from three tumor indications (gastric cancer, colorectal cancer and ovarian cancer) were tested using both technologies. Results: Up to four FFPE slides were used for RNA extraction to support RNA-Seq testing; out of the 1,202 samples processed, 1,099 generated extracted RNA of sufficient quality and quantity (as measured by RNA concentration, RIN score and %DV200) to proceed to sequencing, which resulted in a pass rate of 91.4% for RNA-Seq. The HTG EdgeSeq PIP panel resulted in a pass rate of 97.3% (samples passing QC metrics) when the same 1,200 samples were tested, and required only a single FFPE section owing to the small sample requirement. The t-SNE (a non-linear dimensionality reduction method) analysis of the common 1,358 genes revealed similar clustering of the three cancer indications between the two methods. Correlations across individual genes by sample resulted in the mean Spearman correlation coefficient of 0.73 (95% confidence interval of 0.61 - 0.80). Additionally, gene-wise comparisons across all samples were also evaluated. Conclusions: These data demonstrate that HTG EdgeSeq gene expression panels can be used as a competitive alternative to RNA-Seq, generating equivalent gene expression results, while offering the added benefits of a small sample size requirement, lack of RNA extraction bias, and fully automated data analysis pipeline.

RNA sequencing (RNA-seq) is a powerful technology that allows one to assess the RNA levels in a sample. Analysis of these levels can help in identifying novel transcripts (coding, non-coding and splice variants), understanding transcript structures, and estimating gene/allele expression. Biologists face specific challenges while designing RNA-seq experiments. The nature of these challenges lies in determining the total number of sequenced reads and technical replicates required for detecting marginally differentially expressed transcripts. Despite previous attempts to address these challenges, easily-accessible and biologist-friendly mobile applications do not exist. Thus, we developed RNAtor, a mobile application for Android platforms, to aid biologists in correctly designing their RNA-seq experiments. The recommendations from RNAtor are based on simulations and real data.

Genomic variations are associated with gene expression levels, which are called expression quantitative trait loci (eQTL). Most eQTL may affect the total gene expression levels by regulating transcriptional activities of a specific promoter. However, the direct exploration of genomic loci associated with promoter activities using RNA-seq data has been challenging because eQTL analyses treat the total expression levels estimated by summing those of all isoforms transcribed from distinct promoters. Here we propose a new method for identifying genomic loci associated with promoter activities, called promoter usage quantitative trait loci (puQTL), using conventional RNA-seq data. By leveraging public RNA-seq datasets from the lymphoblastoid cell lines of 438 individuals from the GEUVADIS project, we obtained promoter activity estimates and mapped 2,592 puQTL at the 10% FDR level. The results of puQTL mapping enabled us to interpret the manner in which genomic variations regulate gene expression. We found that 310 puQTL genes (16.1%) were not detected by eQTL analysis, suggesting that our pipeline can identify novel variant–gene associations. Furthermore, we identified genomic loci associated with the activity of “hidden” promoters, which the standard eQTL studies have ignored. We found that most puQTL signals were concordant with at least one genome-wide association study (GWAS) signal, enabling novel interpretations of the molecular mechanisms of complex traits. Our results emphasize the importance of the re-analysis of public RNA-seq datasets to obtain novel insights into gene regulation by genomic variations and their contributions to complex traits.

Mestrado em Biomedicina Molecular
The non-standard decoding of the CUG codon in Candida cylindracea raises a number of questions about the evolutionary process of this organism and other species Candida clade for which the codon is ambiguous. In order to find some answers we studied the transcriptome of C. cylindracea, comparing its behavior with that of Saccharomyces cerevisiae (standard decoder) and Candida albicans (ambiguous decoder). The transcriptome characterization was performed using RNA-seq. This approach has several advantages over microarrays and its application is booming. TopHat and Cufflinks were the software used to build the protocol that allowed for gene quantification. About 95% of the reads were mapped on the genome. 3693 genes were analyzed, of which 1338 had a non-standard start codon (TTG/CTG) and the percentage of expressed genes was 99.4%. Most genes have intermediate levels of expression, some have little or no expression and a minority is highly expressed. The distribution profile of the CUG between the three species is different, but it can be significantly associated to gene expression levels: genes with fewer CUGs are the most highly expressed. However, CUG content is not related to the conservation level: more and less conserved genes have, on average, an equal number of CUGs. The most conserved genes are the most expressed. The lipase genes corroborate the results obtained for most genes of C. cylindracea since they are very rich in CUGs and nothing conserved. The reduced amount of CUG codons that was observed in highly expressed genes may be due, possibly, to an insufficient number of tRNA genes to cope with more CUGs without compromising translational efficiency. From the enrichment analysis, it was confirmed that the most conserved genes are associated with basic functions such as translation, pathogenesis and metabolism. From this set, genes with more or less CUGs seem to have different functions. The key issues on the evolutionary phenomenon remain unclear. However, the results are consistent with previous observations and shows a variety of conclusions that in future analyzes should be taken into consideration, since it was the first time that such a study was conducted.
A descodificação não-standard do codão CUG na Candida cylindracea levanta uma série de questões sobre o processo evolutivo deste organismo e de outras espécies do subtipo Candida para as quais o codão é ambíguo. No sentido de encontrar algumas respostas procedeu-se ao estudo do transcriptoma de C. cylindracea, comparando o seu comportamento com o de Saccharomyces cerevisiae (descodificador standard) e de Candida albicans (descodificador ambíguo). A caracterização do transcriptoma foi realizada a partir de RNA-seq. Esta metodologia apresenta várias vantagens em relação aos microarrays e a sua aplicação encontra-se em franca expansão. TopHat e Cufflinks foram os softwares utilizados na construção do protocolo que permitiu efectuar a quantificação génica. Cerca de 95% das reads alinharam contra o genoma. Foram analisados 3693 genes, 1338 dos quais com codão start não-standard (TTG/CTG) e a percentagem de genoma expresso foi de 99,4%. Maioritarimente, os genes têm níveis de expressão intermédios, alguns apresentam pouca ou nenhuma expressão e uma minoria é altamente expressa. O perfil de distribuição do codão CUG entre as três espécies é muito diferente, mas pode associar-se significativamente aos níveis de expressão: os genes com menos CUGs são os mais altamente expressos. Porém, o conteúdo em CUG não se relaciona com o nível de conservação: genes mais e menos conservados têm, em média, igual número de CUGs. Os genes mais conservados são os mais expressos. Os genes de lipases corroboram os resultados obtidos para os genes de C. cylindracea em geral, sendo muito ricos em CUGs e nada conservados. A quantidade reduzida de codões CUG que se observa em genes altamente expressos pode dever-se, eventualmente, a um número insuficiente de genes de tRNA para fazer face a mais CUGs sem comprometer a eficiência da tradução. A partir da análise de enriquecimento foi possível confirmar que os genes mais conservados estão associados a funções básicas como tradução, patogénese e metabolismo. Dentro destes, os genes com mais e menos CUGs parecem ter funções diferentes. As questões-chave sobre o fenómeno evolutivo permanecem por esclarecer. No entanto, os resultados são compatíveis com as observações anteriores e são apresentadas várias conclusões que em futuras análises devem ser tidas em consideração, já que foi a primeira vez que um estudo deste tipo foi realizado.

Massively parallel whole transcriptome sequencing and its ability to generate full transcriptome data at the single transcript level provides a powerful tool with multiple interrelated applications, including transcriptome reconstruction, gene/isoform expression estimation, also known as transcriptome quantification. As a result, whole transcriptome sequencing has become the technology of choice for performing transcriptome analysis, rapidly replacing array-based technologies. The most commonly used transcriptome sequencing protocol, referred to as RNA-Seq, generates short (single or paired) sequencing tags from the ends of randomly generated cDNA fragments. RNA-Seq protocol reduces the sequencing cost and significantly increases data throughput, but is computationally challenging to reconstruct full-length transcripts and accurately estimate their abundances across all cell types. We focus on two main problems in transcriptome data analysis, namely, transcriptome reconstruction and quantification. Transcriptome reconstruction, also referred to as novel isoform discovery, is the problem of reconstructing the transcript sequences from the sequencing data. Reconstruction can be done de novo or it can be assisted by existing genome and transcriptome annotations. Transcriptome quantification refers to the problem of estimating the expression level of each transcript. We present a genome-guided and annotation-guided transcriptome reconstruction methods as well as methods for transcript and gene expression level estimation. Empirical results on both synthetic and real RNA-seq datasets show that the proposed methods improve transcriptome quantification and reconstruction accuracy compared to previous methods.

DNA and RNA play an essential role in the life of each living organism. The two molecules have different characteristics and properties but their functions are strictly related. DNA encodes all the genetic instructions needed by the main cell activities in the so-called genome. DNA is related to RNA through the gene expression process, which transcribes the information encoded by DNA into RNAs. Opposite to the static information provided by DNA, the set of transcribed RNAs at a specific instant represents the current state of each cell and, at the end, it provides a dynamic characterization of its activity. For this reason, transcriptome analysis represents a powerful tool to identify the dynamic behavior of an organism, such as the response to environmental stimuli or the pathological mechanisms involved in diseases. In recent years, transcriptomic analyses were revolutionized by the advent of RNA sequencing (RNA-Seq), a new methodology that applies current Next Generation Sequencing (NGS) techniques to RNA molecules. RNA-Seq enables to investigate at high resolution all the RNA species present in a sample, characterizing their sequences and quantifying their abundances at the same time. In practice, millions of short transcript sub-sequences, called reads, are sequenced from random positions of the input RNAs using the same NGS platforms employed in DNA sequencing. Unfortunately, no information is provided about which transcripts have generated the reads or from which part of the transcripts they come from. For this reason, reads represent at the same time the output of the sequencing process and the input of complex RNA-Seq data analysis pipelines. The first task in many RNA-Seq data analysis pipelines consists in identifying the relation between the sequencing output (i.e. reads) and the sequenced transcripts. The most common approach to this problem consists in aligning the reads against a reference genome. Once the reads are positioned in the genome, it is possible to infer which transcripts have generated them analyzing the read locations. The information coming from the positions and the number of reads could be employed in a wide range of downstream analyses. For example, counting the number of reads aligned to a gene could give a measure of its expression level, whereas studying which reads are located across exon junction could identify different isoforms. At first glance, these tasks may seem very simple, but the implementation of both the single steps and the whole analysis workflow are in fact complex and still not well defined. Among all the analysis steps in the pipeline, this thesis is focused on the read alignment problem. Read alignment is identified as one of the most critical steps, both for its almost ubiquitous presence in the different RNA-Seq analysis workflows and for its complexity. The study of this pivotal task was carried out through several steps. First, a complete characterization of the problem was performed, analyzing the alignment challenges both from a methodological and a computational point of view. In addition, the algorithms and data structures employed in the alignment process were analyzed together with different ways of modeling the read alignment problem. Then, state of the art methods for RNA-Seq read alignment were identified performing a thorough literature search about RNA-Seq, which revealed the presence of many available methods. At the same time, the literature search highlighted that the identification of a suitable alignment method for a specific application is challenging, mainly due to the lack of accurate comparative analyses. Thus, a comprehensive benchmark analysis of fourteen splice aware alignment methods and four splice unaware tools was designed and performed. The simulation of several datasets describing real scenarios and the definition of a comprehensive set of accuracy and efficiency metrics were performed in order to assess the different alignment methods. The assessment revealed considerable differences between methods’ performance, highlighting often a poor correlation between accuracy and popularity. Finally, the effect of the alignment accuracy on the reliability of an expression level quantification study was assessed for a subset of alignment methods. Overall, this thesis considers the RNA-Seq read alignment problem and presents a thorough characterization of its characteristics and challenges. In a fast evolving research field such as RNA-Seq, the information resulting from the assessment of state of the art methods provides some valuable guidelines for the definition of robust and accurate analysis pipelines.
DNA e RNA giocano un ruolo essenziale nelle vita di ogni organismo. Le due molecole hanno differenti caratteristiche e proprietà ma le loro funzioni sono strettamente legate. Il DNA codifica nel genoma tutte le informazioni genetiche necessarie alle principali attività delle cellula. Il DNA è legato all’RNA tramite il processo della espressione genica, processo che trascrive le informazioni codificate dal DNA nel RNA. Diversamente dalle informazioni statiche fornite dal DNA, l’insieme degli RNA trascritti in un certo istante temporale rappresenta lo stato attuale di ogni cellula e fornisce una caratterizzazione dinamica della sua attività. Per questa ragione, l’analisi del trascrittoma rappresenta un potente strumento per identificare il comportamento dinamico di un organismo, come la risposta a stimoli ambientali o i meccanismi patologici alla base di diverse malattie. Negli ultimi anni, le analisi del trascrittoma sono state rivoluzionate dall’avvento dell’RNA sequencing (RNA-Seq), una nuova metodologia che applica le attuali tecnologie di sequenziamento di nuova generazione (NGS) a molecole di RNA. L’RNA-Seq consente di studiare tutte le specie di RNA presenti nel campione in esame, caratterizzando allo stesso tempo a loro sequenza nucleotidica e la loro quantità. In pratica, milioni di sotto sequenze dei trascritti, chiamate read, vengono sequenziate a partire da posizioni casuali dei trascritti presenti nel campione, utilizzando le medesime piattaforme NGS impiegate nel sequenziamento di DNA. Sfortunatamente le tecnologie NGS producono in output le sono read e nessuna informazione viene quindi fornita riguardo a quali trascritti abbiano generato le read o da quale porzione dei trascritti esse provengano. Per questo motivo le read rappresentano allo stesso tempo l’output del processo di sequenziamento e l’input di complesse pipeline di analisi dati RNA-Seq. Il primo passo in molte pipeline consiste proprio nella identificazione della relazione tra l’output del sequenziamento (le read) e i trascritti che sono stati sequenziati. L’approccio più comune alla risoluzione di questo problema è l’allineamento delle read su un genoma di riferimento. Infatti, identificando la posizione di ogni read nel genoma è possibile inferire quale trascritto la abbia originata analizzando la sua posizione all’interno dei geni. L’informazione derivante dalla posizione e dal numero di read può essere poi utilizzata in un ampio spettro di analisi. Ad esempio, il conteggio del numero di read allineate presso un gene può essere utilizzato come misura del suo livello di espressione, mentre lo studio di quali read si trovino a cavallo di una giunzione può permettere l’identificazione di diverse isoforme. A prima vista queste analisi possono sembrare semplici, ma l’implementazione sia della intera pipeline di analisi sia delle singole fasi che la compongono è invece complessa ed ancora non ben definita. Tra tutte le fasi che compongono la pipeline di analisi dati RNA-Seq, questa tesi si focalizza sulla fase di allineamento delle read. L’allineamento delle read costituisce uno dei passi più critici nella intera analisi di dati RNA-Seq, sia per la sua complessità che per la sua diffusione e presenza nella maggior parte delle pipeline di analisi utilizzate. Lo studio di questa fondamentale operazione è stato effettuato attraverso varie fasi. In primo luogo è stata effettuata una completa caratterizzazione del problema dell’allineamento, analizzando gli aspetti critici e i problemi aperti sia dal punto di vista metodologico che computazionale. In secondo luogo, gli algoritmi e le strutture dati utilizzate nel processo di allineamento sono state analizzate insieme alle diverse strategie di modellazione del problema. Successivamente, i metodi stato dell’arte per l’allineamento di read RNA-Seq sono stati individuati attraverso una approfondita analisi della letteratura, la quale ha evidenziato la presenza di molteplici metodi per la risoluzione di questo problema. Contemporaneamente, l’analisi della letteratura ha evidenziato la difficoltà nella scelta del metodo più accurato per il particolare scenario da analizzare. La difficoltà nella individuazione del corretto metodo è dovuta principalmente per la carenza in letteratura di accurate analisi comparative. Per questa ragione, il passo successivo è stato la progettazione ed esecuzione di una approfondita analisi comparativa di 14 metodi per l’allineamento splice aware e di 4 metodi per l’allineamento splice unaware. A questo scopo, è stata effettua la simulazione di diversi dati a descrizione di molteplici scenari reali. In aggiunta, sono state sviluppate diverse metriche per la valutazione della accuratezza ed efficienza dei singoli metodi analizzati. I risultati di questa analisi hanno rivelato considerevoli differenze tra le prestazioni dei singoli metodi, sottolineando spesso uno scarso legame tra popolarità e accuratezza. L’ultimo passo dello studio è stato l’analisi degli effetti delle diverse accuratezze raggiunge in fase di allineamento sulla precisione e affidabilità delle fasi successive nella pipeline di analisi. Nello specifico, sono state studiate le conseguenze dell’uso di un sottoinsieme dei metodi di allineamento sulla accuratezza della quantificazione del livello di espressione. In conclusione, questa tesi analizza il problema dell’allineamento di read RNA-Seq e presenta una approfondita descrizione delle caratteristiche e delle criticità di questa complessa fase della pipeline. In un campo di ricerca dalla veloce evoluzione come l’RNA-Seq, le informazioni risultanti dalla valutazione comparativa dei metodi stato dell’arte fornisce preziose linee guida per l’aggiornamento e la definizione di accurate e affidabili pipeline di analisi.

RNA-Seq is an increasingly popular approach to transcriptome profiling that uses the capabilities of next generation sequencing technologies and provides better measurement of levels of transcripts and their isoforms. In this thesis, we apply RNA-Seq protocol and transcriptome quantification to estimate gene expression and pathway activity levels. We present a novel method, called IsoDE, for differential gene expression analysis based on bootstrapping. In the first version of IsoDE, we compared the tool against four existing methods: Fisher's exact test, GFOLD, edgeR and Cuffdiff on RNA-Seq datasets generated using three different sequencing technologies, both with and without replicates. We also introduce the second version of IsoDE which runs 10 times faster than the first implementation due to some in-memory processing applied to the underlying gene expression frequencies estimation tool and we also perform more optimization on the analysis. The second part of this thesis presents a set of tools to differentially analyze metabolic pathways from RNA-Seq data. Metabolic pathways are series of chemical reactions occurring within a cell. We focus on two main problems in metabolic pathways differential analysis, namely, differential analysis of their inferred activity level and of their estimated abundance. We validate our approaches through differential expression analysis at the transcripts and genes levels and also through real-time quantitative PCR experiments. In part Four, we present the different packages created or updated in the course of this study. We conclude with our future work plans for further improving IsoDE 2.0.

Latest developments in high-throughput cDNA sequencing (RNA-seq) have revolutionized gene expression profiling. This analysis aims to compare the expression levels of multiple genes between two or more samples, under specific circumstances or in a specific cell to give a global picture of cellular function. Thanks to these advances, gene expression data are being generated in large throughput. One of the primary data analysis tasks for gene expression studies involves data-mining techniques such as clustering and classification. Clustering, which is an unsupervised learning technique, has been widely used as a computational tool to facilitate our understanding of gene functions and regulations involved in a biological process. Cluster analysis aims to group the large number of genes present in a sample of gene expression profile data, such that similar or related genes are in same clusters, and different or unrelated genes are in distinct ones. Classification on the other hand can be used for grouping samples based on their expression profile. There are many clustering and classification algorithms that can be applied in gene expression experiments, the most widely used are hierarchical clustering, k-means clustering and model-based clustering that depend on a model to sort out the number of clusters. Depending on the data structure, a fitting clustering method must be used. In this chapter, we present a state of art of clustering algorithms and statistical approaches for grouping similar gene expression profiles that can be applied to RNA-seq data analysis and software tools dedicated to these methods. In addition, we discuss challenges in cluster analysis, and compare the performance of height commonly used clustering methods on four different public datasets from recount2.

Neuroblastoma, the most common extra-cranial pediatric solid tumor, is responsible for 9–15% of all pediatric cancer deaths. Its intrinsic heterogeneity makes it difficult to successfully treat, resulting in overall survival of 50% for half of the patients. Here we analyze the role in neuroblastoma of the adaptor protein p140Cap, encoded by the SRCIN1 gene. RNA-Seq profiles of a large cohort of neuroblastoma patients show that SRCIN1 mRNA levels are an independent risk factor inversely correlated to disease aggressiveness. In high-risk patients, SRCIN1 was frequently altered by hemizygous deletion, copy-neutral loss of heterozygosity, or disruption. Functional assays demonstrated that p140Cap is causal in dampening both Src and Jak2 kinase activation and STAT3 phosphorylation. Moreover, p140Cap expression decreases in vitro migration and anchorage-independent cell growth, and impairs in vivo tumor progression, in terms of tumor volume and number of spontaneous lung metastasis. p140Cap also contributes to an increased sensitivity of neuroblastoma cells to chemotherapy drugs and to the combined usage of doxorubicin and etoposide with Src inhibitors. Overall, we provide the first evidence that SRCIN1/p140Cap is a new independent prognostic marker for patient outcome and treatment, with a causal role in curbing the aggressiveness of neuroblastoma. We highlight the potential clinical impact of SRCIN1/p140Cap expression in neuroblastoma tumors, in terms of reducing cytotoxic effects of chemotherapy, one of the main issues for pediatric tumor treatment.

We propose an algorithm for RNA-Seq data analysis useful for revealing the “master regulators” of gene expression in experimental condition, as well as of cis-elements regulating transcript level of genes from certain groups.