name: inverse layout: true class: center, middle, inverse
--- # Visualization of RNA-Seq results with CummeRbund --- ## Requirements Before diving into this slide deck, we recommend you to have a look at: - [Galaxy introduction](/training-material/topics/introduction/) - [Quality control](/training-material/topics/sequence-analysis/) --- ### <i class="fa fa-question-circle" aria-hidden="true"></i><span class="visually-hidden">question</span> Questions - How are RNA-Seq results stored? - Why are visualization techniques needed? - How to select our desired subjects for differential gene expression analysis? --- ### <i class="fa fa-bullseye" aria-hidden="true"></i><span class="visually-hidden">objectives</span> Objectives - Manage RNA-Seq results - Extract the desired subject for differential gene expression analysis - Visualize information --- # Why visualization? ??? Data alone does not bring any information: to carry information, data needs to be contextualized. Transcriptomic data is no exception, therefore to organize the growing body of knowledge pertaining RNA-Seq experiments and infer valuable insights, data needs to be organized, annotated, and ultimately visualized. --- ### Where is my data coming from? ![Wang et al, Nat Rev Genet, 2009](../../images/cummerbund-rna-seq-experiment.jpg) <small>[*Wang et al, Nat Rev Genet, 2009*](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2949280/)</small> ??? - RNAs are converted into cDNA fragments through RNA fragmentation - sequencing adaptors (in blue) are then added to each cDNA fragment - cDNA sequences are obtained via NGS sequencing - the obtained reads are subsequently aligned against a reference genome or transcriptome, and classified in-silico as exonic reads, junction reads, and poly-A reads - these three types are then used to outline an expression profile for each gene This is the process that will: - reveal new genes and splice variants - help quantifying cell-specific gene expression within the genome under study But once this pipeline is implemented, how are sequence data going to be analysed and managed? --- ### Bioinformatic tools for RNA-Seq analysis Once the RNA-Seq pipeline is implemented, we still need to handle and analyse all data that is generated. This requires: - computer science skills to be handled - mathematical knowledge to be interpreted --- ### Bioinformatic tools for RNA-Seq analysis ![Trapnell et al, Nat Protoc, 2012](../../images/cummerbund-rna-seq-experiment-tuxedo.jpg) <small>[*Trapnell et al, Nat Protoc, 2012*](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3334321/)</small> ??? - first, reads from each condition are mapped to the reference genome using TopHat - the resulting alignment files are given to Cufflinks, which generates a transcriptome assembly for each condition - the two assemblies are then merged together to provides a uniform basis for calculating of gene and transcript expression in each condition - both reads and merged assemblies are fed to CuffDiff, to calculate expression levels via statistical significance test for the observed changes --- ### Bioinformatic tools for RNA-Seq analysis The last step in our RNA-Seq analysis is CuffDiff. Its output comprises multiple files containing the results of the differential expression analysis. - Gene expression levels are reported as <i>tab-separated</i> values: a simple tabular output that can be viewed with any spreadsheet application. Such files contain statistics, gene-related, and transcript-related attributes .image-75[![CuffDiff output](../../images/cummerbund-cuffdiff-output.png)] - Another way to collect all these data is to organize it within a dedicated database for later consultation. CuffDiff can be instructed to do so .image-50[![CuffDiff output SQLite](../../images/cummerbund-cuffdiff-set-sqlite.png)] ??? - CuffDiff provides analyses of differential expression and regulation at the gene and transcript level - its results are reported in a tab separated format - the overall collection of data is difficult to read to obtain a bird's-eye view of the change of expression - the data can be organized in a SQLite database --- ### Bioinformatic tools for RNA-Seq analysis Whatever storage strategy you opted for, i.e. multiple tab-separated-value files or a SQLite database, all data is still retained within text format. We need to have a bird's-eye view of that data, and <i>make sense</i> of it --- # Visualization --- ### CummeRbund [CummeRbund](http://compbio.mit.edu/cummeRbund/) is an R package for visualizing the results of a CuffDiff output. - Manages, integrates, and visualizes all data produced by CuffDiff - Simplifies data exploration - Provides a bird's-eye view of the expression analysis - Helps creating publication-ready plots --- ### CummeRbund CummeRbund needs to be instructed on which data to be visualized: - <i>Extract CuffDiff</i>'s "Transcript differential expression testing" table - <i>Filter</i> the table on the column storing the significance of a differentially expressed gene - <i>Sort</i> all entries on the basis of most significant differentially expressed gene - Identify the most significant differentially expressed gene --- ### CummeRbund Once the most significant differentially expressed gene has been identified, CummeRbund can generate publication-ready plots to highlight... .image-50[![CummeRbund Expression plot](../../images/cummerbund-expression-plot.png)] The expression of all isoforms of the single gene with replicate FPKMs --- ### CummeRbund Once the most significant differentially expressed gene has been identified, CummeRbund can generate publication-ready plots to highlight... .image-50[![CummeRbund Expression plot](../../images/cummerbund-expression-bar-plot.png)] The expression bar-plot of all isoforms of a gene with replicate FPKMs --- ### CummeRbund ...and many more .image-50[![Other CummeRbund plots](../../images/cummerbund-other-plots.png)] Have a look at [CummerBund's tutorial](https://bioconductor.org/packages/2.11/bioc/vignettes/cummeRbund/inst/doc/cummeRbund-manual.pdf) to overview all possibilities! --- ### <i class="fa fa-key" aria-hidden="true"></i><span class="visually-hidden">keypoints</span> Key points - Extract informations from a SQLite CuffDiff database - Filter and sort results to highlight differential expressed genes of interest - Generate publication-ready visualizations for RNA-Seq analysis results --- ## Thank you! This material is the result of a collaborative work. Thanks the [Galaxy Training Network](https://wiki.galaxyproject.org/Teach/GTN) and all the contributors (Andrea Bagnacani) !