DRG directory

Beta Shiny (R) app to interactively host -omics data.

Updated version hosted at: https://livedataoxford.shinyapps.io/drg-directory/

---

see omics-database for updated software and user guide.

---

This userguide is associated with the paper Barry et al . 2023 PAIN, biorxiv

Allison M Barry, Na Zhao, Xun Yang, David L Bennett, Georgios Baskozos
‘Deep RNA-seq of male and female murine sensory neuron subtypes after nerve injury’

Data is also availale from Barry (2022) (PhD Thesis), GSE216444, Supplemental Tables

Barry, A. M. (2022). Exploring the molecular underpinnings of neuropathic pain in primary afferent subtypes [PhD thesis]. University of Oxford. ORA, Supplemental Tables

---

A more comprehensive guide will follow.

USER GUIDE:

1. The data is in .Rdata format. To load user's own data, substitute the directory in data_dir variable at the top of server.R file.
2. The .Rdata should include three files:
   1. a count matrix, here bulkseq_mat
   2. a processed differential gene data file that contain the following columns: ('symbol', 'loadfold2change', 'padj') with gene ID as row names.
   3. a colData, (here bulkseq_colData), describing the Timepoint, Population, Sex, Condition of each sample (or associated factors).
3. If user's datafile have different names, use the find-replace method to change the name of variables in the file into user's own filenames
4. Variable names:
   • tcounts_med is a data matrix containing median expression of each gene
   • population_labels, subpopulation_labels and sex labels are full names of abbreviations used; can be substituted according to user's data
   • subpopulations is a list of csv file names storing processed datasets for each subpopulation. Can be adjusted based on user's data
5. There are server modules for each graph type
   1. df(data, type): a function for processing count data; converts count matrix into a transpose version i.e. row names = sample, and colnames = genes
   2. plotline_server(id, df, sex): a server module for plotting line graphs; used to show interaction between timepoints and injury. "df" is the count data for selected genes.
      • if user data does not contain 'Timepoint' variable, add a new column named Timepoint in user's colData with values 'na' for each sample. This will prevent errors when grouping samples by Sex Timepoint; Also, delete "linetype=Timepoint" in the ggplot object.
   3. plotdot_server(id, df, sex): a server module for plotting Gene expression dot plots showing
      • plotting gene expression for selected genes.
      • if user's data does not contain 'Sex' variable, add a new column named "Sex" in user's colData with values 'mixed' for each sample. This will prevent errors when grouping samples by Sex. Delete facet_wrap(~Sex, ncol=2, labeller = labeller(Sex = sexlabels)) in the ggplot
      • df is the count data for selected genes.
      • Dots are coloured by gene expression.
      • Dot sizes reflect differences in more highly expressed genes.
   4. plotsubtype_server(id, df, sex): Gene expression dot plots showing differences across population
      • if user's data does not contain 'Sex' variable, add a new column named "Sex" in user's colData with values 'mixed' for each sample. This will prevent errors when grouping samples by Sex. Delete facet_wrap(~Sex, ncol=2, labeller = labeller(Sex = sexlabels)) in the ggplot
      • if user's data does not contain 'Population' variable, this plot is useful to be included
   5. contrast_table_server(id, df): A module for displaying tables showing the significance of load2foldchange. df is the processed differential gene data file from DESeq2.
   6. goi_table_server(id, df): A module for displaying the individual count for selected genes of all samples. "df" is the count data for selected genes.
   7. deg_plot_server(id, df): A module for plotting the significance of load2foldchange for each gene. "df" is the count data for selected genes.
   8. volcano_plot(df, ils): A function for plotting volcano plots. This function is called in the volcano plot server. ils is the list of selected genes.
      • Volcano plots are generated per subtype, showing large fold changes in injury (or other conditions)-related genes.
   9. volcano_plot_server(id, df, ils): A module for plotting volcano plots
      • processed differential gene data files are loaded. A volcano plot was plotted for each subtype to show fold changes in injury (or other conditions)-related genes.
      • volcano plots are generated using the external function volcano_plot()

Additional Information:

1. If users would like to delete existing plots, they can directly delete server modules.
2. If users would like to add more plots, they can create new server modules.