add ChEMBL functions

NAMESPACE

@@ -21,6 +21,7 @@ export(getCellosaurusAPI)
 export(getCellosaurusDataFrame)
 export(getCelloxml)
 export(getChemblAllMechanisms)
+export(getChemblMechanism)
 export(getDrugTargets)
 export(getFDAOrangeBookProducts)
 export(getFailed)

R/getChEMBL.R

@@ -223,6 +223,60 @@ moleculeQuery <- function(field, filter_type, value) {
 }
 
 
+#' Query the ChEMBL mechanism API end-point for a given CHEMBL_ID or list of CHEMBL_IDs
+#' 
+#' @md 
+#' @title Query the ChEMBL mechanism API end-point for a given CHEMBL_ID or list of CHEMBL_IDs
+#' @description A specialised function for querying the mechanism resource from the ChEMBL API. 
+#' 
+#' @param chembl.ID `character(1)` or `character(n)` CHEMBL_ID(s) to query
+#' @param resources `character(1)` Resource to query. Default is "mechanism"
+#' @param field `character(1)` or `character(n)` Field(s) to query. Default is "molecule_chembl_id"
+#' @param filter_type `character(1)` Filter type. Default is "in"
+#' @param returnURL `logical(1)` Return the URL(s) instead of the data table
+#' 
+#' @return `data.table` A table containing all mechanism of action entries and the CHEMBL_IDs queried
+#' 
+#' @examples 
+#' getChemblMechanism("CHEMBL1413")
+#' 
+#' chembl_ids <- c("CHEMBL515", "CHEMBL235191", "CHEMBL1413")
+#' getChemblMechanism(chembl_ids)
+#' 
+#' @export 
+getChemblMechanism <- function(
+    chembl.ID, 
+    resources = "mechanism", 
+    field = "molecule_chembl_id", 
+    filter_type = "in",
+    returnURL = FALSE){
+
+
+    # constructChemblQuery(resource = "mechanism", field = "molecule_chembl_id", filter_type = "in", value = "CHEMBL1413")
+    urls <- constructChemblQuery(resource = resources, field = field, filter_type = filter_type, value = chembl.ID)
+    urls <- URLencode(urls)
+
+    if(returnURL) return(urls)
+
+    # [1] "action_type"               "binding_site_comment"     
+    # [3] "direct_interaction"        "disease_efficacy"         
+    # [5] "max_phase"                 "mec_id"                   
+    # [7] "mechanism_comment"         "mechanism_of_action"      
+    # [9] "mechanism_refs"            "molecular_mechanism"      
+    # [11] "molecule_chembl_id"        "parent_molecule_chembl_id"
+    # [13] "record_id"                 "selectivity_comment"      
+    # [15] "site_id"                   "target_chembl_id"         
+    # [17] "variant_sequence"         
+
+    responses <- data.table::rbindlist(lapply(urls, function(url){
+        response <- httr::GET(url)
+        response <- AnnotationGx::parseJSON(response)
+        mechanisms <- data.table::as.data.table(response$mechanisms)
+        result <- mechanisms[,.(molecule_chembl_id, action_type, mechanism_of_action, molecular_mechanism, mechanism_comment, parent_molecule_chembl_id, target_chembl_id)]
+    }))
+
+    return(responses)
+}
 
 
 # Write testing code here, this is only executed if the file is run as a script
@@ -233,4 +287,6 @@ if (sys.nframe() == 0) {
     library(httr)
     test1 <- constructChemblQuery("target", "pref_name", "contains", "kinase")
     test2 <- compoundQuery("field", "fil", "j")
-}
+}
+
+

R/getPubChem.R

@@ -766,15 +766,23 @@ getAllPubChemAnnotations <-
 #' @export
 getPubChemAnnotation <- function(
     compound,
-    header = 'ChEMBL ID',
+    annotationType = 'ChEMBL ID',
     url = 'https://pubchem.ncbi.nlm.nih.gov/rest/pug_view/data/compound',
     output = 'JSON', 
     timeout_s = 29,
     retries = 3,
     quiet = TRUE,
     throttleMessage = FALSE
     ){
-
+        header <- annotationType
+
+        validHeaders <-  c('ChEMBL ID', 'NSC Number', 'Drug Induced Liver Injury', 'DILI', 'CAS', 'ATC Code')
+
+        # make sure type is in the list of valid types
+        if(!(header %in% validHeaders)){
+            stop(paste0("header must be one of the following: ", paste0(validHeaders, collapse = ", ")))
+        }
+
         # TODO:: add a check to see if the compound is a valid CID or SID
         # TODO:: allow for variaitons of headers due to spelling errors
         # Temporary:
@@ -837,23 +845,23 @@ getPubChemAnnotation <- function(
         names(result) <- c("cid", header)
         return(result)
     }
-
+ 
 #' Function that parses the results of the PubChem PUG-VIEW API for the CHEMBL ID header
-.parseCHEMBLresponse <- function(response){
+.parseCHEMBLresponse <- function(result){
     result <- result$Record$Reference$SourceID
     result <- gsub("::Compound", "", result)
     return(result)
 }
 
 #' Function that parses the results of the PubChem PUG-VIEW API for the NSC Number header
-.parseNSCresponse <- function(response){
+.parseNSCresponse <- function(result){
     result <- result$Record$Reference$SourceID[1]
     result <- gsub(" ", "", result)
     return(result)
 }
 
 #' Function that parses the results of the PubChem PUG-VIEW API for the DILI header
-.parseDILIresponse <- function(response){
+.parseDILIresponse <- function(result){
     if(length(result$Record$Section) == 0){
                 result <- "NA"
     }else{
@@ -879,13 +887,13 @@ getPubChemAnnotation <- function(
 }
 
 #' Function that parses the results of the PubChem PUG-VIEW API for the CAS header
-.parseCASresponse <- function(response){
+.parseCASresponse <- function(result){
     result <- result$Record$Reference$SourceID[1]
     return(result)
 }
 
 #' Function that parses the results of the PubChem PUG-VIEW API for the ATC Code header
-.parseATCresponse <- function(response){
+.parseATCresponse <- function(result){
     if(length(result$Record$Section) == 0){
                 result <- "NA"