phenoscanner regionqueries in batches

R/phenoscanner only allows for certain number of items supplied. This simple function return a large number of calls in batches as well as generating SNPIDs.

Usage

regionqueries(
  regionlist,
  catalogue = "pQTL",
  proxies = "EUR",
  p = 5e-08,
  r2 = 0.8,
  build = 37,
  wait = TRUE
)

Arguments

regionlist

a list of SNPs

catalogue

"None","eQTL","mQTL","methQTL","pQTL","GWAS".

proxies

"None", "AFR","AMR","EAS","EUR","SAS".

p

p value threshold.

r2

r2 for LD.

build

37, 38.

wait

a flag to wait for 1hr for every 50 regions.

Value

The returned value is a list containing tiles, regions and results.

Details

Batches are generated and queries are combined into one.

Note

adapted from custom codings

See also

phenoscanner

Examples

if (FALSE) {
# single region
regionqueries("chr17:26691290-26700110")

# SCALLOP -- SomaLogic lookup from PhenoScanner
INF <- Sys.getenv("INF")
INF1_merge <- merge(inf1,
                  read.delim(file.path(INF,"work","INF1.merge-rsid"),as.is=TRUE),
                  by="prot")
INF1_merge_uniprot <- with(INF1_merge,unique(uniprot))
SomaLogic_INF1_merge <- subset(SomaLogic160410,UniProt %in% INF1_merge_uniprot)
regions <- subset(INF1_merge,uniprot %in% with(SomaLogic_INF1_merge,UniProt))
singletons <- with(regions, Start-End<=2)
flank <- 5e+2
regions[singletons,"Start"] <- regions[singletons,"Start"] - flank
regions[singletons,"End"] <- regions[singletons,"End"] + flank
reset <- with(regions,Start < 0)
regions[reset,"Start"] <- 0
r <- regionqueries(with(regions,paste0(Chrom,":",Start,"-",End)))
save(r,file="INF1_merge.rda",compress='xz')
r2 <- with(r,
{
 region_ext <- cbind(tiles,regions)
 results_ext <- merge(region_ext,results,by="region")
 ord <- with(results_ext,order(group))
 results_ext[ord,]
})
results <- subset(r2,pmid=="29875488")
grp <- names(table(with(results,group)))
sink("INF1_merge.txt")
options(width=250)
for(g in as.numeric(grp))
{
  uniprot <- regions[g,"uniprot"]
  SNP <- regions[g,"SNP"]
  print(regions[g,])
  s <- subset(results,group==g&rsid==SNP)
  vars <- c("region","group","rsid","hg19_coordinates","hgnc","beta","se","p","snpid") 
  if(nrow(s)>1) print(s[vars])
}
sink()
}