Laura L. Elo

A survey of best practices for RNA-seq data analysis

RNA-sequencing (RNA-seq) has a wide variety of applications, but no single analysis pipeline can be used in all cases. We review all of the major steps in RNA-seq data analysis, including experimental design, quality control, read alignment, quantification of gene and transcript levels, visualization, differential gene expression, alternative splicing, functional analysis, gene fusion detection and eQTL mapping. We highlight the challenges associated with each step. We discuss the analysis of small RNAs and the integration of RNA-seq with other functional genomics techniques. Finally, we discuss the outlook for novel technologies that are changing the state of the art in transcriptomics.

Comparison of software packages for detecting differential expression in RNA-seq studies

RNA-sequencing (RNA-seq) has rapidly become a popular tool to characterize transcriptomes. A fundamental research problem in many RNA-seq studies is the identification of reliable molecular markers that show differential expression between distinct sample groups. Together with the growing popularity of RNA-seq, a number of data analysis methods and pipelines have already been developed for this task. Currently, however, there is no clear consensus about the best practices yet, which makes the choice of an appropriate method a daunting task especially for a basic user without a strong statistical or computational background. To assist the choice, we perform here a systematic comparison of eight widely used software packages and pipelines for detecting differential expression between sample groups in a practical research setting and provide general guidelines for choosing a robust pipeline. In general, our results demonstrate how the data analysis tool utilized can markedly affect the outcome of the data analysis, highlighting the importance of this choice.

A practical comparison of methods for detecting transcription factor binding sites in ChIP-seq experiments

BackgroundChromatin immunoprecipitation coupled with massively parallel sequencing (ChIP-seq) is increasingly being applied to study transcriptional regulation on a genome-wide scale. While numerous algorithms have recently been proposed for analysing the large ChIP-seq datasets, their relative merits and potential limitations remain unclear in practical applications.ResultsThe present study compares the state-of-the-art algorithms for detecting transcription factor binding sites in four diverse ChIP-seq datasets under a variety of practical research settings. First, we demonstrate how the biological conclusions may change dramatically when the different algorithms are applied. The reproducibility across biological replicates is then investigated as an internal validation of the detections. Finally, the predicted binding sites with each method are compared to high-scoring binding motifs as well as binding regions confirmed in independent qPCR experiments.ConclusionsIn general, our results indicate that the optimal choice of the computational approach depends heavily on the dataset under analysis. In addition to revealing valuable information to the users of this technology about the characteristics of the binding site detection approaches, the systematic evaluation framework provides also a useful reference to the developers of improved algorithms for ChIP-seq data.

Improving missing value estimation in microarray data with gene ontology

MOTIVATION Gene expression microarray experiments produce datasets with frequent missing expression values. Accurate estimation of missing values is an important prerequisite for efficient data analysis as many statistical and machine learning techniques either require a complete dataset or their results are significantly dependent on the quality of such estimates. A limitation of the existing estimation methods for microarray data is that they use no external information but the estimation is based solely on the expression data. We hypothesized that utilizing a priori information on functional similarities available from public databases facilitates the missing value estimation. RESULTS We investigated whether semantic similarity originating from gene ontology (GO) annotations could improve the selection of relevant genes for missing value estimation. The relative contribution of each information source was automatically estimated from the data using an adaptive weight selection procedure. Our experimental results in yeast cDNA microarray datasets indicated that by considering GO information in the k-nearest neighbor algorithm we can enhance its performance considerably, especially when the number of experimental conditions is small and the percentage of missing values is high. The increase of performance was less evident with a more sophisticated estimation method. We conclude that even a small proportion of annotated genes can provide improvements in data quality significant for the eventual interpretation of the microarray experiments. AVAILABILITY Java and Matlab codes are available on request from the authors. SUPPLEMENTARY MATERIAL Available online at http://users.utu.fi/jotatu/GOImpute.html.

Genome-wide Profiling of Interleukin-4 and STAT6 Transcription Factor Regulation of Human Th2 Cell Programming

Dissecting the molecular mechanisms by which T helper (Th) cells differentiate to effector Th2 cells is important for understanding the pathogenesis of immune-mediated diseases, such as asthma and allergy. Because the STAT6 transcription factor is an upstream mediator required for interleukin-4 (IL-4)-induced Th2 cell differentiation, its targets include genes important for this process. Using primary human CD4(+) T cells, and by blocking STAT6 with RNAi, we identified a number of direct and indirect targets of STAT6 with ChIP sequencing. The integration of these data sets with detailed kinetics of IL-4-driven transcriptional changes showed that STAT6 was predominantly needed for the activation of transcription leading to the Th2 cell phenotype. This integrated genome-wide data on IL-4- and STAT6-mediated transcription provide a unique resource for studies on Th cell differentiation and, in particular, for designing interventions of human Th2 cell responses.

Systematic construction of gene coexpression networks with applications to human T helper cell differentiation process

MOTIVATION Coexpression networks have recently emerged as a novel holistic approach to microarray data analysis and interpretation. Choosing an appropriate cutoff threshold, above which a gene-gene interaction is considered as relevant, is a critical task in most network-centric applications, especially when two or more networks are being compared. RESULTS We demonstrate that the performance of traditional approaches, which are based on a pre-defined cutoff or significance level, can vary drastically depending on the type of data and application. Therefore, we introduce a systematic procedure for estimating a cutoff threshold of coexpression networks directly from their topological properties. Both synthetic and real datasets show clear benefits of our data-driven approach under various practical circumstances. In particular, the procedure provides a robust estimate of individual degree distributions, even from multiple microarray studies performed with different array platforms or experimental designs, which can be used to discriminate the corresponding phenotypes. Application to human T helper cell differentiation process provides useful insights into the components and interactions controlling this process, many of which would have remained unidentified on the basis of expression change alone. Moreover, several human-mouse orthologs showed conserved topological changes in both systems, suggesting their potential importance in the differentiation process. SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.

Innate immune activity is detected prior to seroconversion in children with HLA-conferred type 1 diabetes susceptibility.

The insult leading to autoantibody development in children who will progress to develop type 1 diabetes (T1D) has remained elusive. To investigate the genes and molecular pathways in the pathogenesis of this disease, we performed genome-wide transcriptomics analysis on a unique series of prospective whole-blood RNA samples from at-risk children collected in the Finnish Type 1 Diabetes Prediction and Prevention study. We studied 28 autoantibody-positive children, out of which 22 progressed to clinical disease. Collectively, the samples covered the time span from before the development of autoantibodies (seroconversion) through the diagnosis of diabetes. Healthy control subjects matched for date and place of birth, sex, and HLA-DQB1 susceptibility were selected for each case. Additionally, we genotyped the study subjects with Immunochip to identify potential genetic variants associated with the observed transcriptional signatures. Genes and pathways related to innate immunity functions, such as the type 1 interferon (IFN) response, were active, and IFN response factors were identified as central mediators of the IFN-related transcriptional changes. Importantly, this signature was detected already before the T1D-associated autoantibodies were detected. Together, these data provide a unique resource for new hypotheses explaining T1D biology.

Th1 Response and Cytotoxicity Genes Are Down-Regulated in Cutaneous T-Cell Lymphoma

Purpose: Increased production of Th2 cytokines characterizes Sezary syndrome, the leukemic form of cutaneous T-cell lymphomas (CTCL). To identify the molecular background and to study whether shared by the most common CTCL subtype, mycosis fungoides, we analyzed the gene expression profiles in both subtypes. Experimental Design: Freshly isolated cells from 30 samples, representing skin, blood, and enriched CD4+ cell populations of mycosis fungoides and Sezary syndrome, were analyzed with Affymetrix (Santa Clara, CA) oligonucleotide microarrays, quantitative PCR, or immunohistochemistry. The gene expression profiles were combined with findings of comparative genomic hybridization of the same samples to identify chromosomal changes affecting the aberrant gene expression. Results: We identified a set of Th1-specific genes [e.g., TBX21 (T-bet), NKG7, and SCYA5 (RANTES)] to be down-regulated in Sezary syndrome as well as in a proportion of mycosis fungoides samples. In both Sezary syndrome and mycosis fungoides blood samples, the S100P and LIR9 gene expression was up-regulated. In lesional skin, IL7R and CD52 were up-regulated. Integration of comparative genomic hybridization and transcriptomic data identified chromosome arms 1q, 3p, 3q, 4q, 12q, 16p, and 16q as likely targets for new CTCL-associated gene aberrations. Conclusions: Our findings revealed several new genes involved in CTCL pathogenesis and potential therapeutic targets. Down-regulation of a set of genes involved in Th1 polarization, including the major Th1-polarizing factor, TBX21, was for the first time associated with CTCL. In addition, a plausible explanation for the proliferative response of CTCL cells to locally produced interleukin-7 was revealed.

SATB1 dictates expression of multiple genes including IL-5 involved in human T helper cell differentiation

Special AT-rich binding protein 1 (SATB1) is a global chromatin organizer and a transcription factor regulated by interleukin-4 (IL-4) during the early T helper 2 (Th2) cell differentiation. Here we show that SATB1 controls multiple IL-4 target genes involved in human Th cell polarization or function. Among the genes regulated by SATB1 is that encoding the cytokine IL-5, which is predominantly produced by Th2 cells and plays a key role in the development of eosinophilia in asthma. We demonstrate that, during the early Th2 cell differentiation, IL-5 expression is repressed through direct binding of SATB1 to the IL-5 promoter. Furthermore, SATB1 knockdown-induced up-regulation of IL-5 is partly counteracted by down-regulating GATA3 expression using RNAi in polarizing Th2 cells. Our results suggest that a competitive mechanism involving SATB1 and GATA3 regulates IL-5 transcription, and provide new mechanistic insights into the stringent regulation of IL-5 expression during human Th2 cell differentiation.

Reproducibility-Optimized Test Statistic for Ranking Genes in Microarray Studies

A principal goal of microarray studies is to identify the genes showing differential expression under distinct conditions. In such studies, the selection of an optimal test statistic is a crucial challenge, which depends on the type and amount of data under analysis. Although previous studies on simulated or spike-in data sets do not provide practical guidance on how to choose the best method for a given real data set, we introduce an enhanced reproducibility-optimization procedure, which enables the selection of a suitable gene-ranking statistic directly from the data. In comparison with existing ranking methods, the reproducibility-optimized statistic shows good performance consistently under various simulated conditions and on Affymetrix spike-in data set. Further, the feasibility of the novel statistic is confirmed in a practical research setting using data from an in-house cDNA microarray study of asthma-related gene expression changes. These results suggest that the procedure facilitates the selection of an appropriate test statistic for a given data set without relying on a priori assumptions, which may bias the findings and their interpretation. Moreover, the general reproducibility-optimization procedure is not limited to detecting differential expression only but could be extended to a wide range of other applications as well.