useCase1b.Rmd

---
title: "iLINCS API R Notebook"
output:
  pdf_document: default
  html_document:
    df_print: paged
---
<style type="text/css">
.main-container {
  max-width: 1200px;
  margin-left: auto;
  margin-right: auto;
}
</style>

<!-- ## Prerequisites  -->
#Loading packages
```{r include = FALSE}
library(knitr)
library(tinytex)
library(httr)
library(jsonlite)
library(htmltools)
library(ggplot2)
library(enrichR)
```
#Setting default width option
```{r set-options, echo=FALSE, cache=FALSE}
options(width = "200")
```

# Display Signature Libraries
```{r list signature libraries}
apiUrl <- "http://www.ilincs.org/api/SignatureLibraries"
req <- GET(apiUrl)
json <- httr::content(req, as = "text")
ilincs_libraries <- fromJSON(json)
ilincs_libraries[,c("libraryID","libraryName")]
```

# Searching for Everolimus chemical perturbage (CP) perturbation signatures in the MCF7 cell line

```{r search signatures}
term <- "Everolimus"
ilincs_libId<-"LIB_5"
apiUrl <- paste("http://www.ilincs.org/api/SignatureMeta/findTermWithSynonyms?term=",term,"&library=",ilincs_libId,sep="")
req <- GET(apiUrl)

everolimusCps<-fromJSON(httr::content(req,type="text"))$data
everolimusCpsMcf7<-everolimusCps[intersect(grep("Everolimus",everolimusCps$compound),grep("MCF7",everolimusCps$cellline)),]
everolimusCpsMcf7[,c("compound","time","concentration","signatureid","cellline")]
```
# Enrichr analysis of everolimus signature at 0.04uM concentration (LINCSCP_137891) (Fig 2C)
```{r}
ilincs_signatureId <- "LINCSCP_137891"

req <- POST("http://www.ilincs.org/api/ilincsR/downloadSignature", body = list(sigID = ilincs_signatureId, display = FALSE), encode = "json")
ilincs_sessionId<-unlist(httr::content(req))
ilincs_sessionId

signatureFileUrl=paste("http://www.ilincs.org/tmp/",ilincs_sessionId,".xls",sep="")
everolimusSignatureData<-read.table(signatureFileUrl,sep="\t",header=T,stringsAsFactors = F)
sigGenes100<-everolimusSignatureData$Name_GeneSymbol[order(everolimusSignatureData$Significance_pvalue)[1:100]]

setEnrichrSite("Enrichr") 
enriched <- enrichr(genes=sigGenes100, databases="KEGG_2019_Human")

enriched[["KEGG_2019_Human"]][1:5,]
```

# Finding and summarizing connected CGSes with the Everolimus signature at 0.04uM concentration (LINCSCP_137891)
## Finding connected CGSes
```{r connected CGSes}
ilincs_libId<-"LIB_6"

apiUrl <- paste("http://www.ilincs.org/api/SignatureMeta/findConcordantSignatures?sigID=",ilincs_signatureId,"&lib=",ilincs_libId,sep="")
req <- GET(apiUrl)

connectedCgs<-fromJSON(httr::content(req,type="text"))
head(connectedCgs[,c("signatureid","similarity","pValue","cellline","treatment")])
```

## Summary boxplot, Figure 2D in the manuscript
```{r connected CGSes Boxplot}
geneFreq <- sort(table(connectedCgs$treatment[1:100]),decreasing=TRUE)[5:1]
geneFreqDf<-data.frame(gene=factor(names(geneFreq),levels=names(geneFreq),ordered=T),geneFreq=as.vector(geneFreq))

tp100cgs<-ggplot(data=geneFreqDf,aes(x=gene,y=geneFreq)) + 
  geom_bar(stat="identity",fill="steelblue") + 
  coord_flip() + theme_bw() +
  theme(axis.ticks=element_line(color="grey80"),text=element_text(size=20),legend.key.size=unit(1,"cm"),axis.line.x=element_line(colour="black"),panel.grid.major=element_blank(),panel.grid.minor=element_blank(),panel.border=element_blank(),panel.background=element_blank()) +
  labs(x="",y="Number of Signatures")
tp100cgs
```

## Enrichr analysis of connected CGSes  (Fig 2D)
```{r}
setEnrichrSite("Enrichr") 
enrichedCgsTargets <- enrichr(genes=unique(connectedCgs$treatment), databases="KEGG_2019_Human")
enrichedCgsTargets[["KEGG_2019_Human"]][1:5,]
```

# Finding and summarizing connected chemical perturbage (CP) signatures with the Everolimus signature at 0.04uM concentration  (LINCSCP_137891)
## Finding connected CPs
```{r connected CPs}
ilincs_libId<-"LIB_5"

apiUrl <- paste("http://www.ilincs.org/api/SignatureMeta/findConcordantSignatures?sigID=",ilincs_signatureId,"&lib=",ilincs_libId,sep="")
req <- GET(apiUrl)

connectedCps<-fromJSON(httr::content(req,type="text"))
head(connectedCps)
```

## Summary boxplot, Figure 2E in the manusript
```{r connected CPs Boxplot}
top100GeneTargets<-unlist(strsplit(connectedCps$GeneTargets[1:100],split="\\|"))
geneFreq <- sort(table(top100GeneTargets),decreasing=TRUE)[5:1]
geneFreqDf<-data.frame(gene=factor(names(geneFreq),levels=names(geneFreq),ordered=T),geneFreq=as.vector(geneFreq))

tp100cp<-ggplot(data=geneFreqDf,aes(x=gene,y=geneFreq)) + 
  geom_bar(stat="identity",fill="steelblue") + 
  coord_flip() + theme_bw() +
  theme(axis.ticks=element_line(color="grey80"),text=element_text(size=20),legend.key.size=unit(1,"cm"),axis.line.x=element_line(colour="black"),panel.grid.major=element_blank(),panel.grid.minor=element_blank(),panel.border=element_blank(),panel.background=element_blank()) +
  labs(x="",y="Number of Signatures")
tp100cp
```

## Enrichr analysis of connected CGSes  (Fig 2E)
```{r}
setEnrichrSite("Enrichr") 
enrichedCgsTargets <- enrichr(genes=unique(top100GeneTargets), databases="KEGG_2019_Human")
enrichedCgsTargets[["KEGG_2019_Human"]][1:5,]
```