Skip to content

Commit

Permalink
Merge pull request #10 from cantinilab/dev_SeuratV5
Browse files Browse the repository at this point in the history
Dev seurat v5
  • Loading branch information
r-trimbour authored Jul 16, 2024
2 parents 1937238 + ac4da17 commit 121bc63
Show file tree
Hide file tree
Showing 2 changed files with 19 additions and 8 deletions.
11 changes: 11 additions & 0 deletions .github/workflows/pkgdown.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -40,6 +40,17 @@ jobs:
reticulate::py_install("hummuspy", envname = "r-reticulate", pip=TRUE)
shell: Rscript {0}

- name: ubuntu setup for Monocle3
run: sudo apt-get install libgdal-dev libgeos-dev libproj-dev

- name: Install Monocle3
run: devtools::install_github('cole-trapnell-lab/monocle3')
shell: Rscript {0}

- name: Install Cicero
run: devtools::install_github("cole-trapnell-lab/cicero-release", ref = "monocle3")
shell: Rscript {0}

- name: Build site
run: pkgdown::build_site_github_pages(new_process = FALSE, install = FALSE)
shell: Rscript {0}
Expand Down
16 changes: 8 additions & 8 deletions vignettes/chen_vignette.Rmd
Original file line number Diff line number Diff line change
Expand Up @@ -85,7 +85,7 @@ You can also request any "EnsDB" object adapted to your data
(e.g. EnsDb.Hsapiens.v86::EnsDb.Hsapiens.v86 for human genome annotations)
or use your own genome annotations in the same format.

```{r genome_annotations, eval=FALSE, warning=FALSE}
```{r genome_annotations, eval=TRUE, warning=FALSE}
# get human genome annotation from EndDb data
# wrapper of Signac::GetGRangesFromEnsDb, adapting output to UCSC format
genome_annotations <- get_genome_annotations(
Expand All @@ -95,7 +95,7 @@ genome_annotations <- get_genome_annotations(
```

Add genome annotations to hummus/seurat object
```{r add_genome_annotations, eval=FALSE}
```{r add_genome_annotations, eval=TRUE}
Signac::Annotation(hummus@assays$peaks) <- genome_annotations
rm(genome_annotations)
```
Expand All @@ -112,7 +112,7 @@ correspondance table between TFs and motifs. tfs is a named vector of the TFs.
You can also use your own motifs_db object, as long as it contains the same
slots.

```{r get_tf2motifs, eval=FALSE}
```{r get_tf2motifs, eval=TRUE}
# Load TF motifs from JASPAR2020 and chromVARmotifs in hummus object
hummus@motifs_db <- get_tf2motifs() # by default human motifs
```
Expand Down Expand Up @@ -146,7 +146,7 @@ BSGenome object is used to identify location of motifs and intersect them with
peak
<br>You can also specify the name of the bipartite that will be added to the
hummus object. By default, it will be named "tf_peak".
```{r bipartite_tf_peak, eval=FALSE}
```{r bipartite_tf_peak, eval=TRUE}
hummus <- bipartite_tfs2peaks(
hummus_object = hummus,
tf_expr_assay = "RNA", # use to filter TF on only expressed TFs,
Expand All @@ -159,7 +159,7 @@ hummus <- bipartite_tfs2peaks(

### 2.2. Genes - peaks bipartite reconstruction
Peaks - genes bipartite is computed
```{r bipartite_peaks2genes, eval=FALSE}
```{r bipartite_peaks2genes, eval=TRUE}
hummus <- bipartite_peaks2genes(
hummus_object = hummus,
gene_assay = "RNA",
Expand All @@ -184,7 +184,7 @@ interactions involving at least one TF present in the dataset will be kept.
network that will be added to the hummus object.
The added network will be undirected and unweighted since PPI and OmniPath
database are not directional nor return any weight here.
```{r tf_network, eval=FALSE}
```{r tf_network, eval=TRUE}
hummus <- compute_tf_network(hummus,
gene_assay = "RNA", # default = None ;
# If a assay is provided,
Expand Down Expand Up @@ -212,7 +212,7 @@ The returned network will be considered undirected and weighted. While GENIE3
returns a directed network, we symmetrize it for the random walk with restart
exploration of the genes proximity.

```{r gene_network, eval=FALSE}
```{r gene_network, eval=TRUE}
hummus <- compute_gene_network(
hummus,
gene_assay = "RNA",
Expand All @@ -236,7 +236,7 @@ the network, you will need to specify the output file name (`output_file`).
The returned network will be considered undirected and weighted, since cis-regulatory
interaction and Cicero outputs are not directional.

```{r peak_network, eval=FALSE}
```{r peak_network, eval=TRUE}
hummus <- compute_atac_peak_network(hummus,
atac_assay = "peaks",
verbose = 1,
Expand Down

0 comments on commit 121bc63

Please sign in to comment.