In sigven/oncoEnrichR: Cancer-dedicated gene set interpretation

Tissue and cell type enrichment

Using data from the Human Protein Atlas (HPA), we are here interrogating classification of all protein coding genes in the query set with respect to elevated expression in normal/healthy tissues and cell types
Gene expression per tissue is based on consensus data from HPA and Genotype-Tissue Expression (GTEx) Project, containing data from 50 distinct tissue types. Gene expression measure unit: normalized transcripts pr million (nTPM)
RNA single-cell data comes from HPA, in which transcript expression levels are summarized per gene in 81 cell types from 31 datasets. Gene expression measure unit: normalized transcripts pr million (nTPM)

Target set - tissue specificity {.tabset}

Genes have been classified, based on mean expression (across samples) per tissue in GTex, into distinct specificity categories (algorithm developed within HPA):
Not detected: Genes with a mean expression level less than 1 (nTPM < 1) across all the tissues.
Tissue enriched: Genes with a mean expression level greater than or equal to 1 (nTPM >= 1) that also have at least four-fold higher expression levels in a particular tissue compared to all other tissues.
Group enriched: Genes with a mean expression level greater than or equal to 1 (nTPM >= 1) that also have at least four-fold higher expression levels in a group of 2-5 tissues compared to all other tissues, and that are not considered Tissue enriched.
Tissue enhanced: Genes with a mean expression level greater than or equal to 1 (nTPM >= 1) that also have at least four-fold higher expression levels in a particular tissue compared to the average levels in all other tissues, and that are not considered Tissue enriched or Group enriched.
Low tissue specificity: Genes with an expression level greater than or equal to 1 (nTPM >= 1) across all of the tissues that are not in any of the above 4 groups.
Mixed: Genes that are not assigned to any of the above 5 groups.
Enrichment of specific tissues in the query set (with respect to tissue-specific gene expression) is performed with TissueEnrich
Only tissues that are enriched with an adjusted (Benjamini-Hochberg) p-value < 0.05 are listed

oe_tissue_category_df <- 
  onc_enrich_report[['data']][['cell_tissue']][['tissue_overview']][['category_df']]
oe_tissue_category_plot <-
    ggplot2::ggplot(oe_tissue_category_df, ggplot2::aes(x = category, y = pct)) +
    ggplot2::geom_col(ggplot2::aes(color = group, fill = group),
                      position = ggplot2::position_dodge(0.8), width = 0.7) +
    ggsci::scale_color_locuszoom() +
    ggsci::scale_fill_locuszoom() +
    ggplot2::ylab("Percent") +
    ggplot2::xlab("") +
    ggplot2::ylim(0,100) +
    ggplot2::theme_classic() +
    ggplot2::theme(
      legend.position = "top",
      legend.title = ggplot2::element_blank(),
      axis.text.x = ggplot2::element_text(angle = 30, size = 11, vjust = 0.5),
      legend.text = ggplot2::element_text(face="bold", family = "Helvetica", size = 12),
      axis.text.y = ggplot2::element_text(family = "Helvetica", size = 12),
      axis.title.x = ggplot2::element_text(family = "Helvetica", size = 12),
      axis.title.y = ggplot2::element_text(family = "Helvetica", size = 12)


    )

rm(oe_tissue_category_df)
plotly::ggplotly(oe_tissue_category_plot, height = 500, width = 900) |>  
  plotly::layout(legend = list(orientation = "h", x = 0.2, y = -0.45))
rm(oe_tissue_category_plot)

Tissue specificities per target gene

tissue_specs_per_gene <- 
  crosstalk::SharedData$new(
    onc_enrich_report[['data']][['cell_tissue']][['tissue_enrichment']][['per_gene']] |> dplyr::select(-cancer_max_rank))

crosstalk::bscols(
  list(
    crosstalk::filter_select("category", "Category", tissue_specs_per_gene, ~category)
  ),
  list(
    crosstalk::filter_select("tissue", "Tissues", tissue_specs_per_gene, ~tissue)
  )
)

htmltools::br()

DT::datatable(
  tissue_specs_per_gene, 
  escape = F, 
  extensions=c("Buttons","Responsive"), 
  width = "100%",
  style = 'bootstrap',
  rownames = F,
  options=list(buttons = c('csv','excel'),
               pageLength = 10,
               bPaginate = T,
               dom = 'Bfrtip')
  ) |>
  DT::formatStyle(
    'category', 
    color = "white",
    backgroundColor = DT::styleEqual(
      onc_enrich_report[["config"]][["cell_tissue"]][["tissue_enrichment_levels"]],
      onc_enrich_report[["config"]][["cell_tissue"]][["enrichment_colors"]]
    )
  )

Tissue enrichment - query set

num_significant_tissues <- 
  onc_enrich_report[['data']][['cell_tissue']][['tissue_enrichment']]$per_type |>
  dplyr::filter(log10_pvalue >= 1.3) |>
  nrow()

htmltools::br()

cat("<ul><li>")
cat('<i><font style="font-size: 100%">Considering the tissue specificities of members of the query set, <b>NO TISSUES</b> are enriched (adjusted p-value < 0.05) compared to the background set.</i></font>', sep='\n')
cat('</ul></li>')

htmltools::br()
htmltools::br()
htmltools::br()
htmltools::br()

enriched_tissues_df <- 
  onc_enrich_report[['data']][['cell_tissue']][['tissue_enrichment']]$per_type |>
  dplyr::filter(log10_pvalue >= 1.3)

DT::datatable(
  enriched_tissues_df, 
  escape = F, 
  extensions=c("Buttons","Responsive"), 
  width = "100%",
  style = 'bootstrap',
  rownames = F,
  options=list(buttons = c('csv','excel'),
               pageLength = 10,
               bPaginate = T,
               dom = 'Bfrtip')
  )
htmltools::br()

Target set - cell type specificity {.tabset}

Genes have been classified, based on mean expression (across samples) per cell type, into distinct specificity categories (algorithm developed within HPA):
Not detected: Genes with a mean expression level less than 1 (nTPM < 1) across all the cell types.
Cell type enriched: Genes with a mean expression level greater than or equal to 1 (nTPM >= 1) that also have at least four-fold higher expression levels in a particular cell type compared to all other cell types.
Group enriched: Genes with a mean expression level greater than or equal to 1 (nTPM >= 1) that also have at least four-fold higher expression levels in a group of 2-10 cell types compared to all other cell types, and that are not considered Cell type enriched.
Cell type enhanced: Genes with a mean expression level greater than or equal to 1 (nTPM >= 1) that also have at least four-fold higher expression levels in a particular cell type compared to the average levels in all other cell types, and that are not considered Cell type enriched or Group enriched.
Low cell type specificity: Genes with an expression level greater than or equal to 1 (nTPM >= 1) across all of the cell types that are not in any of the above 4 groups.
Mixed: Genes that are not assigned to any of the above 5 groups.
Enrichment of specific cell types in the query set (with respect to cell type-specific gene expression) is performed with TissueEnrich
Only cell types that are enriched with an adjusted (Benjamini-Hochberg) p-value < 0.05 are listed

oe_ctype_category_df <- 
  onc_enrich_report[['data']][['cell_tissue']][['scRNA_overview']][['category_df']]
oe_ctype_category_plot <-
    ggplot2::ggplot(oe_ctype_category_df, ggplot2::aes(x = category, y = pct)) +
    ggplot2::geom_col(ggplot2::aes(color = group, fill = group),
                      position = ggplot2::position_dodge(0.8), width = 0.7) +
    ggsci::scale_color_locuszoom() +
    ggsci::scale_fill_locuszoom() +
    ggplot2::ylab("Percent") +
    ggplot2::xlab("") +
    ggplot2::ylim(0,100) +
    ggplot2::theme_classic() +
    ggplot2::theme(
      legend.position = "top",
      legend.title = ggplot2::element_blank(),
      axis.text.x = ggplot2::element_text(angle = 30, size = 11, vjust = 0.5),
      legend.text = ggplot2::element_text(face="bold", family = "Helvetica", size = 12),
      axis.text.y = ggplot2::element_text(family = "Helvetica", size = 12),
      axis.title.x = ggplot2::element_text(family = "Helvetica", size = 12),
      axis.title.y = ggplot2::element_text(family = "Helvetica", size = 12)
    )

rm(oe_ctype_category_df)
plotly::ggplotly(oe_ctype_category_plot, height = 500, width = 900) |>  
  plotly::layout(legend = list(orientation = "h", x = 0.2, y = -0.45))
rm(oe_ctype_category_plot)

Cell type specififies per target gene

celltype_specs_per_gene <- 
  crosstalk::SharedData$new(
    onc_enrich_report[['data']][['cell_tissue']][['scRNA_enrichment']][['per_gene']] |> dplyr::select(-cancer_max_rank))

crosstalk::bscols(
  list(
    crosstalk::filter_select("category", "Category", celltype_specs_per_gene, ~category)
  ),
  list(
    crosstalk::filter_select("cell_type", "Cell type", celltype_specs_per_gene, ~cell_type)
  )
)

htmltools::br()
DT::datatable(
  celltype_specs_per_gene, 
  escape = F, 
  extensions=c("Buttons","Responsive"), 
  width = "100%",
  style = 'bootstrap',
  rownames = F,
  options=list(buttons = c('csv','excel'),
               pageLength = 10,
               bPaginate = T,
               dom = 'Bfrtip')
  ) |>
  DT::formatStyle(
    'category', 
    color = "white",
    backgroundColor = DT::styleEqual(
      onc_enrich_report[["config"]][["cell_tissue"]][["ctype_enrichment_levels"]],
      onc_enrich_report[["config"]][["cell_tissue"]][["enrichment_colors"]]
    )
  )

Cell type enrichment - query set

num_significant_celltypes <- 
  onc_enrich_report[['data']][['cell_tissue']][['scRNA_enrichment']]$per_type |>
  dplyr::filter(log10_pvalue >= 1.3) |>
  nrow()

htmltools::br()

cat("<ul><li>")
cat('<i><font style="font-size: 100%">Considering the cell-type specificities of members of the query set, <b>NO CELL TYPES</b> are enriched (adjusted p-value < 0.05) compared to the background set.</i></font>', sep='\n')
cat('</ul></li>')

htmltools::br()
htmltools::br()
htmltools::br()
htmltools::br()

enriched_celltypes_df <- 
  onc_enrich_report[['data']][['cell_tissue']][['scRNA_enrichment']]$per_type |>
  dplyr::filter(log10_pvalue >= 1.3)

DT::datatable(
  enriched_celltypes_df, 
  escape = F, 
  extensions=c("Buttons","Responsive"), 
  width = "100%",
  style = 'bootstrap',
  rownames = F,
  options=list(buttons = c('csv','excel'),
               pageLength = 10,
               bPaginate = T,
               dom = 'Bfrtip')
  )

htmltools::br()

# p <- ggplot2::ggplot(expression_dist, ggplot2::aes(cell_type, symbol)) +
#   ggplot2::geom_tile(aes(fill = exp),
#             colour = "white") +
#   ggplot2::scale_fill_gradient(low = "white",
#                       high = "steelblue") +
#   ggplot2::labs(x='', y = '') +
#   ggplot2::theme_bw()+
#   ggplot2::guides(fill = ggplot2::guide_legend(title = "Log2(TPM)"))+
#   #theme(legend.position="none")+
#   ggplot2::theme(plot.title = ggplot2::element_text(
#     hjust = 0.5,size = 20),
#     axis.title = ggplot2::element_text(size=15)) +
#   ggplot2::theme(axis.text.x = ggplot2::element_text(angle = 45,
#                                    vjust = 1, hjust = 1),
#         panel.grid.major= ggplot2::element_blank(),
#         panel.grid.minor = ggplot2::element_blank())

{.unlisted .unnumbered .toc-ignore}

Citation Note : If you use the output of the Tissue and cell type enrichment module of oncoEnrichR in your research, please cite the following resources and tools: