Reproducible Pitch

Nicole Gruenheit
12/02/2017

RNA-seq analysis app

The app was designed to assist in the downstream analysis of ~~RNA-seq data~~. Once the user has obtained the read counts for each gene and has done the differential gene expression analysis via different means, they can then filter the data based on different criteria:

  • Strength of expression of the gene (read counts),
  • p-value, which indicates whether a gene is differentially expressed between two conditions
  • log2 Foldchange, which indicates how different the average expression level is between the genes

Results table

Example data for the app will be loaded automatically at startup and the table displayed on the first tab. This table can be searched for specific genes and sorted based on each column. Filtering the data using the checkboxes and textboxes will change the table reactively.

The second tab displays 100 genes with the smallest p-value as a heatmap using all three replicates of both conditions. The number of genes displayed can be changed using the textbox.

The third tab can be used to explore single genes. e.g. if you know a gene that should be different between the conditions you can look at the expression level in all replicates.

Results of filtering the table

plot of chunk unnamed-chunk-1

There are 13408 genes in the table. 11592 are actually expressed, while 9197 are differentially expressed using a p-value of 0.05. 7337 genes also have an average Foldchange of more than twofold and 3435 are up-regulated in one condition.

Function reponsible for loading file

    filedata <- reactive({
        data <- input$datafile
        if (input$readsC){r = input$reads} else{r = 0}
        if (input$pvalueC){p = input$pvalue} else{p = 1}
        if (input$fcC){f = input$fc} else{f = 0}
        data1 = filter(data,sum > r, padj < p, abs(log2FoldChange) > f)
        data1
    })