In bhklab/AnnotationGx: AnnotationGx: A package for building, updating and querying an annotation database for pharmaco-genomic data

THIS VIGNETTE IS A WORK IN PROGRESS

This is an example Pipeline to annotate treatments from the CTRP dataset.

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(AnnotationGx)
library(data.table)
options(mc.cores = parallel::detectCores())

data(CCLE_treatmentMetadata)
data(gCSI_treatmentMetadata)
data(CTRP_treatmentMetadata)
data(GDSC_treatmentMetadata)

annotations_drugs_with_ids <- data.table::fread("/home/bioinf/bhklab/jermiah/Bioconductor/AnnotationGx/inst/extdata/drugs_with_ids.csv")

``` {r mapCCLE} (compounds_to_cids <- CCLE_treatmentMetadata[, AnnotationGx::mapCompound2CID( names = cleanCharacterStrings(CCLE.treatmentid), first = TRUE ) ] ) data.table::setnames(compounds_to_cids, "name", "CCLE.treatmentid") properties <- c("Title", "MolecularFormula", "InChIKey", "MolecularWeight")

cids_to_properties <- compounds_to_cids[, { AnnotationGx::mapCID2Properties( ids = cids, properties = properties ) } ]

CCLE_treatmentMetadata_annotated <- merge( compounds_to_cids, cids_to_properties, by.x = "cids", by.y = "CID", all.x = TRUE ) old_names <- c("cids", "MolecularFormula", "MolecularWeight", "InChIKey", "Title")

data.table::setnames(CCLE_treatmentMetadata_annotated, old_names, paste0("PubChem.", old_names))

CCLE_treatmentMetadata_annotated

``` {r mapGCSI}
(compounds_to_cids <- 
  gCSI_treatmentMetadata[, 
    AnnotationGx::mapCompound2CID(
        names = cleanCharacterStrings(gCSI.treatmentid),
        first = TRUE
        )
      ]
)

data.table::setnames(compounds_to_cids, "name", "gCSI.treatmentid")
failed <- attributes(compounds_to_cids)$failed
failed_drugs_gCSI <- names(failed)
message("Failed to map the following drugs: ", paste(failed_drugs_gCSI, collapse = ", "))


properties <- c("Title", "MolecularFormula", "InChIKey", "MolecularWeight")

cids_to_properties <- compounds_to_cids[
  !is.na(cids), 
  {
    AnnotationGx::mapCID2Properties(
      ids = cids,
      properties = properties
      )
  }
]

gCSI_treatmentMetadata_annotated <- merge(
  compounds_to_cids,
  cids_to_properties,
  by.x = "cids",
  by.y = "CID",
  all.x = TRUE
)

old_names <- c("cids", "MolecularFormula", "MolecularWeight", "InChIKey", "Title")

data.table::setnames(gCSI_treatmentMetadata_annotated, old_names, paste0("PubChem.", old_names))

gCSI_treatmentMetadata_annotated

``` {r mapCTRP} (CTRP_compounds_to_cids <- CTRP_treatmentMetadata[, AnnotationGx::mapCompound2CID( names = CTRP.treatmentid, first = TRUE ) ] ) failed <- attributes(CTRP_compounds_to_cids)$failed |> names() failed

properties <- c("Title", "MolecularFormula", "InChIKey", "MolecularWeight")

cids_to_properties <- CTRP_compounds_to_cids[ !is.na(cids), { AnnotationGx::mapCID2Properties( ids = cids, properties = properties ) } ]

CTRP_treatmentMetadata_annotated <- merge( CTRP_compounds_to_cids, cids_to_properties, by.x = "cids", by.y = "CID", all.x = TRUE )

old_names <- c("cids", "MolecularFormula", "MolecularWeight", "InChIKey", "Title")

# GDSC
``` {r mapGDSC}

(compounds_to_cids <- 
  GDSC_treatmentMetadata[, 
    AnnotationGx::mapCompound2CID(
        names =  GDSC.treatmentid,
        first = TRUE
        )
      ]
)
failed <- 
  attributes(compounds_to_cids)$failed |> 
    names()
failed

properties <- c("Title", "MolecularFormula", "InChIKey", "MolecularWeight")

cids_to_properties <- compounds_to_cids[
  !is.na(cids), 
  {
    AnnotationGx::mapCID2Properties(
      ids = cids,
      properties = properties
      )
  }
]

GDSC_treatmentMetadata_annotated <- merge(
  compounds_to_cids,
  cids_to_properties,
  by.x = "cids",
  by.y = "CID",
  all.x = TRUE
)

old_names <- c("cids", "MolecularFormula", "MolecularWeight", "InChIKey", "Title")

data.table::setnames(GDSC_treatmentMetadata_annotated, old_names, paste0("PubChem.", old_names))

GDSC_treatmentMetadata_annotated

``` {r using cancerxgene API} url <- "https://www.cancerrxgene.org/api/compounds?list=all&export=tsv"

responses <- AnnotationGx:::.build_request(url) |> AnnotationGx:::.perform_request()

GDSC_API_treatmentMetadata <- AnnotationGx:::.parse_resp_tsv(responses) |> data.table::as.data.table()

GDSC_API_treatmentMetadata[, unique(PubCHEM), by = "Name"][!is.na(V1) & !V1 %in% c("none","None"),]

``` {r compare the two}

gdsc_merged <- merge(
  GDSC_API_treatmentMetadata[, unique(PubCHEM), by = "Name"][!is.na(V1) & !V1 %in% c("none","None"),],
  compounds_to_cids[!is.na(cids)][order(name)],
  by.x = "Name", 
  by.y = "name",
  all = T)


non_na_gdsc_merged <- gdsc_merged[!is.na(V1) & !is.na(cids),]

# subset where the V1 value is not the same as the cids value
non_na_gdsc_merged[V1 != cids,]

bhklab/AnnotationGx documentation built on April 3, 2025, 4:27 p.m.