Merge pull request #16 from tidytranscriptomics-workshops/details_stefano

update vignette and presentation

Author: mblue9

.gitignore

@@ -10,3 +10,5 @@ bioc2022tidytranscriptomics*.tgz
 dev
 /doc/
 /Meta/
+.DS_Store
+._.DS_Store

data-raw/seurat_obj.R

@@ -44,7 +44,10 @@ sce_obj =
 	rename(cell_type = curated_cell_type) |> 
 
 	# filtering because of to few samples per cell types
-	filter(cell_type !="CD8+_Tem")
+	filter(cell_type !="CD8+_Tem") |> 
+	
+	# Replace file path
+	mutate(file = file |> str_replace("bhupinder_10X_260819", "single_cell"))
 
 # job::job({
 	save(sce_obj , file="data/sce_obj.rda", compress = "xz")

data/sce_obj.rda

@@ -0,0 +1,3 @@
+._.smbdeleteAAA240c2803
+._bioc2022_tidytranscriptomics.pdf
+.smbdeleteAAA240c2803

inst/.gitignore

@@ -453,16 +453,16 @@ We use tidyverse `nest` to group the data. The command below will create a tibbl
 
 ```{r}
 pseudo_bulk |>
-  nest(data = -cell_type)
+  nest(grouped_summarized_experiment = -cell_type)
 ```
 
 To explore the grouping, we can use tidyverse `slice` to choose a row (cell_type) and `pull` to extract the values from a column. If we pull the data column we can view the SummarizedExperiment object. 
 
 ```{r}
 pseudo_bulk |>
-  nest(data = -cell_type) |>
+  nest(grouped_summarized_experiment = -cell_type) |>
   slice(1) |>
-  pull(data)
+  pull(grouped_summarized_experiment)
 ```
 
 We can then identify differentially expressed genes for each cell type for our condition of interest, treated versus untreated patients. We use tidyverse `map` to apply differential expression functions to each cell type group in the nested data. 
@@ -473,11 +473,11 @@ pseudo_bulk <-
     
   pseudo_bulk |>
     
-  nest(data = -cell_type) |>
+  nest(grouped_summarized_experiment = -cell_type) |>
     
   # map accepts a data column (.x) and applies functions to each element
-  mutate(data = map(
-    data,
+  mutate(grouped_summarized_experiment = map(
+    grouped_summarized_experiment,
     ~ .x |>
       # Now using tidybulk on SummarizedExperiment
       identify_abundant(factor_of_interest = treatment) |>
@@ -497,7 +497,7 @@ If we pull out the SummarizedExperiment object for the first cell type, as befor
 ```{r}
 pseudo_bulk |>
   slice(1) |>
-  pull(data)
+  pull(grouped_summarized_experiment)
 ```
 
 Now we can create plots for significant genes for each cell type, visualising their transcriptional abundance, also without needing to create multiple objects. 
@@ -509,16 +509,19 @@ pseudo_bulk <-
     
   # Filter out significant
   # using a high FDR value as this is toy data
-  mutate(data = map(data, ~ filter(.x, FDR < 0.5))) |>
+  mutate(grouped_summarized_experiment = map(
+  	grouped_summarized_experiment, 
+  	~ filter(.x, FDR < 0.5)
+  )) |>
     
   # Filter cell types with no differential abundant gene-transcripts
   # map_int is map that returns integer instead of list
-  filter(map_int(data, ~ nrow(.x)) > 0) |>
+  filter(map_int(grouped_summarized_experiment, ~ nrow(.x)) > 0) |>
 
   # Plot
   # map2 is map that accepts 2 input columns (.x, .y)
   mutate(plot = map2(
-    data, cell_type,
+    grouped_summarized_experiment, cell_type,
     ~ .x |>
       ggplot(aes(treatment, counts_scaled + 1)) +
       geom_boxplot(aes(fill = treatment)) +