Merge pull request #118 from ARTbio/update_IOC_singlecell

Update IOC single cell
ARTbio · Apr 24, 2024 · 7f7f36f · 7f7f36f
2 parents c9e0386 + 023f75f
commit 7f7f36f
Show file tree

Hide file tree

Showing 27 changed files with 393 additions and 114 deletions.
diff --git a/docs/scRNAseq_basics/00_IOCsc_week0.md b/docs/scRNAseq_basics/00_IOCsc_week0.md
@@ -0,0 +1,57 @@
+## First Steps with Seurat
+
+Please go read the following pages : 
+
+- [Introduction to Single-Cell RNAseq analysis](introduction.md)  
+- [Initialization of R analysis](intro_seurat.md)  
+- [Import data and intialization of Seurat object](import.md)  
+
+---
+
+![](../R-IOC/images/toolbox-do-it-yourself.png){: style="width:75px"} **Do it yourself!**
+
+Now it's your time to shine ! We are going to put into pratice what we 
+have just seen. By using a more complicated use case, you are going to 
+reproduce the whole scRNAseq analysis with Seurat.  
+
+# Dataset test
+
+The dataset for this analysis will be single cell RNAseq from zebrafish embryos 
+from [Metikala *et al*](https://doi.org/10.1371/journal.pone.0254024). You can download 
+the dataset at the GEO accession [GSE152982](https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE152982).
+
+??? question "Do you need help to find the data ?"
+    ??? tip "First tip : "
+        Look to supplementary file....
+        ??? tip "Second tip : "
+            To see what's inside the tar archive you can click on `(custom)`
+
+Once you download the data, all you have to do is import it onto the Rstudio server.
+
+## Biomart is your friend
+
+Don't forget to import biomaRt in order to help you annotate your genes.
+Make sure to tweak parameters to fit this new dataset. 
+
+!!! question "Trouble to use biomaRt ?"
+    Here is some links to help you with biomaRt :
+
+    - [Vignette of the R package](https://bioconductor.org/packages/release/bioc/vignettes/biomaRt/inst/doc/accessing_ensembl.html)
+    - [Short presentation](https://docs.google.com/presentation/d/1ck41d_0a6bMEreTfeeES67RExEc2pp_OXQvqbJ3ZdhU/edit?usp=sharing) of the use of the R package, comparing it with the Ensembl interface
+
+
+# Render your RMD/QMD
+
+To complete this week you'll need to :
+
+- [x] 1. Retrieve the zebrafish dataset
+- [x] 2. Import data in Rstudio
+- [x] 3. Import data in your global environment
+- [x] 4. Create a Seurat Object
+- [x] 5. Create an annotation table of zebrafish genes using `biomaRt`. 
+
+Add your RMD/QMD in your Trello card.
+
+**Thank you for your attention and see you next week :clap: :clap: :clap:**
+
+----
diff --git a/docs/scRNAseq_basics/01_IOCsc_week1.md b/docs/scRNAseq_basics/01_IOCsc_week1.md
@@ -0,0 +1,46 @@
+## Data Preprocessing
+
+The preprocessing is the most important part of a single cell analysis because you
+can skew your result if you filter too much **or too little** and you must really
+understand what's going on these steps.
+
+The preprocesing is composed of: 
+
+- Filtering of low quality barcodes
+- Barcode normalization
+- Selection of most variable features
+
+Please go read the [preprocessing](preprocessing.md) pages to learn more about it.
+
+---
+
+![](../R-IOC/images/toolbox-do-it-yourself.png){: style="width:75px"} **Do it yourself!**
+
+Now it's your time to shine ! We are going to put into pratice what we 
+have just seen. By using a more complicated use case, you are going to 
+reproduce the whole scRNAseq analysis with Seurat.  
+
+# Dataset test
+
+The dataset for this analysis will be single cell RNAseq from zebrafish embryos 
+from [Metikala *et al*](https://doi.org/10.1371/journal.pone.0254024). You need
+to continue your RMD/QMD from last week. 
+
+# Render your RMD/QMD
+
+To complete this week you'll need to :
+
+- [x] 1. Filtering the low quality barcodes **and explain each cutoff**
+- [x] 2. Normalize the data
+- [x] 3. Identify the most variable genes
+
+!!! warning "IMPORTANT"
+    Please note you **must** be explicative in your cutoff choices and detailled
+    each step of your thoughts. 
+    In general, try to explain in your own words, each step of your analysis !
+
+Add your RMD/QMD in your Trello card.
+
+**Thank you for your attention and see you next week :clap: :clap: :clap:**
+
+----
diff --git a/docs/scRNAseq_basics/02_IOCsc_week2.md b/docs/scRNAseq_basics/02_IOCsc_week2.md
@@ -0,0 +1,13 @@
+
+
+
+---
+
+![](../R-IOC/images/toolbox-do-it-yourself.png){: style="width:75px"} **Do it yourself!**
+
+
+
+
+**Thank you for your attention and see you next week :clap: :clap: :clap:**
+
+----
diff --git a/docs/scRNAseq_basics/03_IOCsc_week3.md b/docs/scRNAseq_basics/03_IOCsc_week3.md
@@ -0,0 +1,13 @@
+
+
+
+---
+
+![](../R-IOC/images/toolbox-do-it-yourself.png){: style="width:75px"} **Do it yourself!**
+
+
+
+
+**Thank you for your attention and see you next week :clap: :clap: :clap:**
+
+----
diff --git a/docs/scRNAseq_basics/04_IOCsc_week4.md b/docs/scRNAseq_basics/04_IOCsc_week4.md
@@ -0,0 +1,13 @@
+
+
+
+---
+
+![](../R-IOC/images/toolbox-do-it-yourself.png){: style="width:75px"} **Do it yourself!**
+
+
+
+
+**Thank you for your attention and see you next week :clap: :clap: :clap:**
+
+----
diff --git a/docs/scRNAseq_basics/05_IOCsc_week5.md b/docs/scRNAseq_basics/05_IOCsc_week5.md
@@ -0,0 +1,13 @@
+
+
+
+---
+
+![](../R-IOC/images/toolbox-do-it-yourself.png){: style="width:75px"} **Do it yourself!**
+
+
+
+
+**Thank you for your attention and see you next week :clap: :clap: :clap:**
+
+----
diff --git a/docs/scRNAseq_basics/06_IOCsc_week6.md b/docs/scRNAseq_basics/06_IOCsc_week6.md
@@ -0,0 +1,13 @@
+
+
+
+---
+
+![](../R-IOC/images/toolbox-do-it-yourself.png){: style="width:75px"} **Do it yourself!**
+
+
+
+
+**Thank you for your attention and see you next week :clap: :clap: :clap:**
+
+----
diff --git a/docs/scRNAseq_basics/07_IOCsc_week7.md b/docs/scRNAseq_basics/07_IOCsc_week7.md
@@ -0,0 +1,13 @@
+
+
+
+---
+
+![](../R-IOC/images/toolbox-do-it-yourself.png){: style="width:75px"} **Do it yourself!**
+
+
+
+
+**Thank you for your attention and see you next week :clap: :clap: :clap:**
+
+----
diff --git a/docs/scRNAseq_basics/clustering.md b/docs/scRNAseq_basics/clustering.md
@@ -23,7 +23,7 @@ represents our cell populations.
 pbmc_small <- FindNeighbors(pbmc_small,          #SeuratObject
                             reduction = "pca",   #Reduction to used
                             k.param = 20,
-                            dims = 1:10)         #Number of PCs to keep (previously determined)
+                            dims = 1:pc_to_keep) #Number of PCs to keep (previously determined)
 
 pbmc_small <- FindClusters(pbmc_small,                                        #SeuratObject
                            resolution = seq(from = 0.2, to = 1.2, by = 0.2),  #Compute clustering with several resolutions (from 0.2 to 1.2 : values usually used)

diff --git a/docs/scRNAseq_basics/images/Clustree-1.png b/docs/scRNAseq_basics/images/Clustree-1.png
diff --git a/docs/scRNAseq_basics/images/ElbowPlot-1.png b/docs/scRNAseq_basics/images/ElbowPlot-1.png
diff --git a/docs/scRNAseq_basics/images/MitoGenes-1.png b/docs/scRNAseq_basics/images/MitoGenes-1.png
diff --git a/docs/scRNAseq_basics/images/MitoGenes-2.png b/docs/scRNAseq_basics/images/MitoGenes-2.png
diff --git a/docs/scRNAseq_basics/images/PCA-1.png b/docs/scRNAseq_basics/images/PCA-1.png
diff --git a/docs/scRNAseq_basics/images/QCFilter-1.png b/docs/scRNAseq_basics/images/QCFilter-1.png
diff --git a/docs/scRNAseq_basics/images/SetIdents-1.png b/docs/scRNAseq_basics/images/SetIdents-1.png
diff --git a/docs/scRNAseq_basics/images/UMAP-1.png b/docs/scRNAseq_basics/images/UMAP-1.png
diff --git a/docs/scRNAseq_basics/images/VariableFeature-1.png b/docs/scRNAseq_basics/images/VariableFeature-1.png
diff --git a/docs/scRNAseq_basics/images/plotJackStraw-1.png b/docs/scRNAseq_basics/images/plotJackStraw-1.png
diff --git a/docs/scRNAseq_basics/images/visualMarkers-1.png b/docs/scRNAseq_basics/images/visualMarkers-1.png
diff --git a/docs/scRNAseq_basics/images/visualMarkers-2.png b/docs/scRNAseq_basics/images/visualMarkers-2.png
diff --git a/docs/scRNAseq_basics/import.md b/docs/scRNAseq_basics/import.md
@@ -232,31 +232,41 @@ pbmc_small <- CreateSeuratObject(tenX_matrix,                    #Expression mat
 
         ## Formal class 'Seurat' [package "SeuratObject"] with 13 slots
         ##   ..@ assays      :List of 1
-        ##   .. ..$ RNA:Formal class 'Assay' [package "SeuratObject"] with 8 slots
-        ##   .. .. .. ..@ counts       :Formal class 'dgCMatrix' [package "Matrix"] with 6 slots
-        ##   .. .. .. .. .. ..@ i       : int [1:2286884] 70 166 178 326 363 410 412 492 494 495 ...
-        ##   .. .. .. .. .. ..@ p       : int [1:2701] 0 781 2133 3264 4224 4746 5528 6311 7101 7634 ...
-        ##   .. .. .. .. .. ..@ Dim     : int [1:2] 32738 2700
-        ##   .. .. .. .. .. ..@ Dimnames:List of 2
-        ##   .. .. .. .. .. .. ..$ : chr [1:32738] "ENSG00000243485" "ENSG00000237613" "ENSG00000186092" "ENSG00000238009" ...
+        ##   .. ..$ RNA:Formal class 'Assay5' [package "SeuratObject"] with 8 slots
+        ##   .. .. .. ..@ layers    :List of 1
+        ##   .. .. .. .. ..$ counts:Formal class 'dgCMatrix' [package "Matrix"] with 6 slots
+        ##   .. .. .. .. .. .. ..@ i       : int [1:2286884] 70 166 178 326 363 410 412 492 494 495 ...
+        ##   .. .. .. .. .. .. ..@ p       : int [1:2701] 0 781 2133 3264 4224 4746 5528 6311 7101 7634 ...
+        ##   .. .. .. .. .. .. ..@ Dim     : int [1:2] 32738 2700
+        ##   .. .. .. .. .. .. ..@ Dimnames:List of 2
+        ##   .. .. .. .. .. .. .. ..$ : NULL
+        ##   .. .. .. .. .. .. .. ..$ : NULL
+        ##   .. .. .. .. .. .. ..@ x       : num [1:2286884] 1 1 2 1 1 1 1 41 1 1 ...
+        ##   .. .. .. .. .. .. ..@ factors : list()
+        ##   .. .. .. ..@ cells     :Formal class 'LogMap' [package "SeuratObject"] with 1 slot
+        ##   .. .. .. .. .. ..@ .Data: logi [1:2700, 1] TRUE TRUE TRUE TRUE TRUE TRUE ...
+        ##   .. .. .. .. .. .. ..- attr(*, "dimnames")=List of 2
+        ##   .. .. .. .. .. .. .. ..$ : chr [1:2700] "AAACATACAACCAC-1" "AAACATTGAGCTAC-1" "AAACATTGATCAGC-1" "AAACCGTGCTTCCG-1" ...
+        ##   .. .. .. .. .. .. .. ..$ : chr "counts"
+        ##   .. .. .. .. .. ..$ dim     : int [1:2] 2700 1
+        ##   .. .. .. .. .. ..$ dimnames:List of 2
         ##   .. .. .. .. .. .. ..$ : chr [1:2700] "AAACATACAACCAC-1" "AAACATTGAGCTAC-1" "AAACATTGATCAGC-1" "AAACCGTGCTTCCG-1" ...
-        ##   .. .. .. .. .. ..@ x       : num [1:2286884] 1 1 2 1 1 1 1 41 1 1 ...
-        ##   .. .. .. .. .. ..@ factors : list()
-        ##   .. .. .. ..@ data         :Formal class 'dgCMatrix' [package "Matrix"] with 6 slots
-        ##   .. .. .. .. .. ..@ i       : int [1:2286884] 70 166 178 326 363 410 412 492 494 495 ...
-        ##   .. .. .. .. .. ..@ p       : int [1:2701] 0 781 2133 3264 4224 4746 5528 6311 7101 7634 ...
-        ##   .. .. .. .. .. ..@ Dim     : int [1:2] 32738 2700
-        ##   .. .. .. .. .. ..@ Dimnames:List of 2
+        ##   .. .. .. .. .. .. ..$ : chr "counts"
+        ##   .. .. .. ..@ features  :Formal class 'LogMap' [package "SeuratObject"] with 1 slot
+        ##   .. .. .. .. .. ..@ .Data: logi [1:32738, 1] TRUE TRUE TRUE TRUE TRUE TRUE ...
+        ##   .. .. .. .. .. .. ..- attr(*, "dimnames")=List of 2
+        ##   .. .. .. .. .. .. .. ..$ : chr [1:32738] "ENSG00000243485" "ENSG00000237613" "ENSG00000186092" "ENSG00000238009" ...
+        ##   .. .. .. .. .. .. .. ..$ : chr "counts"
+        ##   .. .. .. .. .. ..$ dim     : int [1:2] 32738 1
+        ##   .. .. .. .. .. ..$ dimnames:List of 2
         ##   .. .. .. .. .. .. ..$ : chr [1:32738] "ENSG00000243485" "ENSG00000237613" "ENSG00000186092" "ENSG00000238009" ...
-        ##   .. .. .. .. .. .. ..$ : chr [1:2700] "AAACATACAACCAC-1" "AAACATTGAGCTAC-1" "AAACATTGATCAGC-1" "AAACCGTGCTTCCG-1" ...
-        ##   .. .. .. .. .. ..@ x       : num [1:2286884] 1 1 2 1 1 1 1 41 1 1 ...
-        ##   .. .. .. .. .. ..@ factors : list()
-        ##   .. .. .. ..@ scale.data   : num[0 , 0 ]
-        ##   .. .. .. ..@ key          : chr "rna_"
-        ##   .. .. .. ..@ assay.orig   : NULL
-        ##   .. .. .. ..@ var.features : logi(0)
-        ##   .. .. .. ..@ meta.features:'data.frame':   32738 obs. of  0 variables
-        ##   .. .. .. ..@ misc         : list()
+        ##   .. .. .. .. .. .. ..$ : chr "counts"
+        ##   .. .. .. ..@ default   : int 1
+        ##   .. .. .. ..@ assay.orig: chr(0) 
+        ##   .. .. .. ..@ meta.data :'data.frame':  32738 obs. of  0 variables
+        ##   .. .. .. ..@ misc      :List of 1
+        ##   .. .. .. .. ..$ calcN: logi TRUE
+        ##   .. .. .. ..@ key       : chr "rna_"
         ##   ..@ meta.data   :'data.frame': 2700 obs. of  3 variables:
         ##   .. ..$ orig.ident  : Factor w/ 1 level "PBMC analysis": 1 1 1 1 1 1 1 1 1 1 ...
         ##   .. ..$ nCount_RNA  : num [1:2700] 2421 4903 3149 2639 981 ...
@@ -271,20 +281,14 @@ pbmc_small <- CreateSeuratObject(tenX_matrix,                    #Expression mat
         ##   ..@ project.name: chr "PBMC analysis"
         ##   ..@ misc        : list()
         ##   ..@ version     :Classes 'package_version', 'numeric_version'  hidden list of 1
-        ##   .. ..$ : int [1:3] 4 1 0
+        ##   .. ..$ : int [1:3] 5 0 1
         ##   ..@ commands    : list()
         ##   ..@ tools       : list()
 
 === "Dimension of expression matrices"
 
     ``` r
-    dim(pbmc_small@assays$RNA@counts)
-    ```
-
-        ## [1] 32738  2700
-
-    ``` r
-    dim(pbmc_small@assays$RNA@data)
+    dim(pbmc_small[["RNA"]])
     ```
 
         ## [1] 32738  2700
@@ -306,34 +310,35 @@ pbmc_small <- CreateSeuratObject(tenX_matrix,                    #Expression mat
 
 We can observe several slots via the `str` command:  
 
-- `assays`: general slot that will include the different information of each
-  study. They are composed of several things:
-    - the starting expression matrix (`@counts`), usually raw counts or raw UMI
-    - the one that will "undergo" all the modifications (filters,
-      normalization, etc) (`@data`)
-    - dataframe that will be created when scaling the data (`@scale_data`)
-    - prefix used for each calculation that will use this assay (study)
-      (`@key`)
-    - vector of gene names that will be determined to have a variable
-      expression (`@var.features`)
-    - dataframe associated with genes with different metadata
-      (`@meta.features`)
-
-- `meta.data` : gathers all the information about the cells. At the beginning
-  Seurat will calculate the size of the library (nCount_RNA, or the total
-  number of UMI) and the number of detected genes (nFeatures_RNA) for each
-  cell. If a dataframe is given in the `meta.data` parameter of the
-  `CreateSeuratObject` function, its columns will be added after those
-  calculated by Seurat.  
-
-- `active.assay` : study used by default
-- `active.ident` : default cell identity, here the name of the given project,
-  also stored under the column `orig.ident` in the metadata
+-   `assays`: general slot that will include the different information
+    of each study. It’s a list where each element is linked to a
+    expression data type. For instance in the element `RNA` you retrieve
+    all information link to you scRNAseq. But for example, if you have a
+    more complicated study you can also have `ADT` or `peaks` if you
+    perform a CITEseq, scATACseq or Multiome. Here we have a simple
+    scRNAseq, so by default `assays` will be composed of only the
+    element `RNA`. It’s class is `Assay5` you can have a great
+    description in the documentation of the `SeuratObject` package. In
+    you console `?Assay5` to read it.
+-   `meta.data` : gathers all the information about the cells. At the
+    beginning Seurat will calculate the size of the library (nCount_RNA,
+    or the total number of UMI) and the number of detected genes
+    (nFeatures_RNA) for each cell. If a dataframe is given in the
+    `meta.data` parameter of the `CreateSeuratObject` function, its
+    columns will be added after those calculated by Seurat.
+-   `active.assay` : study used by default
+-   `active.ident` : default cell identity, here the name of the given
+    project, also stored under the column `orig.ident` in the metadata
 
 !!! note
     Navigation in the different slots is done via `@` or `$`. Each main slot is
     accessible via the `@`, *i.e.* `object@main slot` to go further in the
     slots tree, most often complex objects are accessible with a `@` (dgCMatrix,
     dataframe) and lists, vectors are accessible via `$`. If in doubt, you can
     refer to the result of the `str` command and use the character in front of
-    each slot name.
+    each slot name. If you want to look to the expression matrix you can go :
+
+    ```r
+    # Retrieve data in an expression matrix RNA counts matrix
+    pbmc_small[["RNA"]]$counts[1:10,1:5]
+    ```
diff --git a/docs/scRNAseq_basics/intro_markers.md b/docs/scRNAseq_basics/intro_markers.md
@@ -79,12 +79,12 @@ kable(head(pbmc_markers))
 
 |                 | p_val | avg_log2FC | pct.1 | pct.2 | p_val_adj | cluster | gene            |
 |:----------------|------:|-----------:|------:|------:|----------:|:--------|:----------------|
-| ENSG00000137154 |     0 |  0.6674692 | 0.998 | 0.997 |         0 | 0       | ENSG00000137154 |
-| ENSG00000144713 |     0 |  0.6134636 | 0.998 | 0.997 |         0 | 0       | ENSG00000144713 |
-| ENSG00000112306 |     0 |  0.7002797 | 1.000 | 0.994 |         0 | 0       | ENSG00000112306 |
-| ENSG00000177954 |     0 |  0.7041791 | 0.998 | 0.994 |         0 | 0       | ENSG00000177954 |
-| ENSG00000164587 |     0 |  0.5988615 | 1.000 | 0.997 |         0 | 0       | ENSG00000164587 |
-| ENSG00000118181 |     0 |  0.7149172 | 1.000 | 0.978 |         0 | 0       | ENSG00000118181 |
+| ENSG00000137154 |     0 |  0.6952948 | 0.998 | 0.997 |         0 | 0       | ENSG00000137154 |
+| ENSG00000144713 |     0 |  0.6444963 | 0.998 | 0.997 |         0 | 0       | ENSG00000144713 |
+| ENSG00000112306 |     0 |  0.7383383 | 1.000 | 0.994 |         0 | 0       | ENSG00000112306 |
+| ENSG00000177954 |     0 |  0.7437027 | 0.998 | 0.994 |         0 | 0       | ENSG00000177954 |
+| ENSG00000164587 |     0 |  0.6352699 | 1.000 | 0.997 |         0 | 0       | ENSG00000164587 |
+| ENSG00000118181 |     0 |  0.7973422 | 1.000 | 0.978 |         0 | 0       | ENSG00000118181 |
 
 The result of this function is a dataframe with several columns: