---
title: "`SingleCellSignalR` : </br> Inference of ligand-receptor 
interactions from single-cell data"

author:
  - name: Jacques Colinge
    affiliation:
    - Institut de Recherche en Cancérologie de Montpellier,
     Inserm, Montpellier, France 
    email: jacques.colinge@inserm.fr
  - name: Jean-Philippe Villemin
    affiliation:
    - Institut de Recherche en Cancérologie de Montpellier,
     Inserm, Montpellier, France 
    email: jean-philippe.villemin@inserm.fr
date: "`r format(Sys.Date(), '%m/%d/%Y')`"
abstract: > 
  <p> `SingleCellSignalR` is used to infer ligand-receptor (L-R) interactions
   from single-cell transcriptomics or proteomics data. Potential L-R
   interactions are taken from the  LR*db* database, which is accessed
   through our `BulkSignalR` library.

   By default, inferences rely on a statistical model linking potential
   L-R interactions with biological pathways from Reactome or biological
   processes from GO. This model is provided by `BulkSignalR`, which
   we used as a basic layer to implement `SingleCellSignalR` Version 2.
   Alternatively, the purely expression-based score
   in Version 1 of `SingleCellSignalR`, the so-called LR-score, remains
   available.
   
  <br/><br/>
output: 
  rmarkdown::html_vignette:
      self_contained: true
      toc: true
      toc_depth: 4
      highlight: pygments
      fig_height: 3
      fig_width: 3
      fig_caption: no
      code_folding: show
vignette: >
  %\VignetteIndexEntry{SingleCellSignalR-Main}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include = FALSE}
knitr::knit_hooks$set(optipng = knitr::hook_optipng)
options(rmarkdown.html_vignette.check_title = FALSE)
```

```{r load-libs, message = FALSE,  warning = FALSE, results = FALSE}
library(BulkSignalR)
library(SingleCellSignalR)
```

SingleCellSignalR package version: `r packageVersion("SingleCellSignalR")`  

&nbsp;

## What is it for? 

`SingleCellSignalR` is a tool that enables the inference of L-R
interactions from single-cell data.  


See also `BulkSignalR` vignette for a more complete description
of all functionalities.

&nbsp;

## Main worfklow

The following code snippet give an example of 
the main commands to use in order to process
your dataset with `SingleCellSignalR` package.  


```{r code1 , eval=TRUE,cache=FALSE}

data(example_dataset,package='SingleCellSignalR')
mat <- log1p(data.matrix(example_dataset[,-1]))/log(2)
rownames(mat) <- example_dataset[[1]]
rme <- rowMeans(mat)
mmat <- mat[rme>0.05,]
d <- dist(t(mmat))
h <- hclust(d, method="ward.D")
clusters <- paste0("pop_", cutree(h, 5))

# SCSRNoNet -> LRscore based approach

scsrnn <- SCSRNoNet(mat,normalize=FALSE,method="log-only",
    min.count=1,prop=0.001,
    log.transformed=TRUE,populations=clusters)

scsrnn <- performInferences(scsrnn,verbose=TRUE,
    min.logFC=1e-10,max.pval=1,min.LR.score=0.5)

# SCSRNet -> DifferentialMode based approach

scsrcn <- SCSRNet(mat,normalize=FALSE,method="log-only",
    min.count=1,prop=0.001,
    log.transformed=TRUE,populations=clusters)

if(FALSE){
scsrcn <- performInferences(scsrcn,
    selected.populations = c("pop_1","pop_2","pop_3"),
    verbose=TRUE,rank.p=0.8,
    min.logFC=log2(1.01),max.pval=0.05)

print("getAutocrines")
inter1 <- getAutocrines(scsrcn, "pop_1")
head(inter1)

print("getParacrines")
inter2 <- getParacrines(scsrcn, "pop_1","pop_2")
head(inter2)

# Visualisation

cellNetBubblePlot(scsrcn)
}

``` 

&nbsp;

## Acknowledgements

We thank Morgan Maillard for his help with the LRdb database and
Pierre Giroux for the work with proteomics.

&nbsp;

Thank you for reading this guide and for using `SingleCellSignalR`.

&nbsp;
    
## Session Information

```{r session-info}
sessionInfo()
```