Extract Gene frequency from Pubmed XML files

gistfile1.txt

#################################################
# Input:
input = "./pubmed_result.xml"

#################################################
# Output:
output = "./abstract.xls"


# install.packages("pubmed.mineR")
library(pubmed.mineR)

# Input
dat0 <- xmlreadabs(input)

dat1 <- gene_atomization(dat0)
symbol <- dat1[, 1]
symbol.blank <- paste("\\b", symbol, "\\b", sep = "")
name <- dat1[, 2]

EntrezID <- c()
Number <- c()
Symbol <- c()
Name <- c()
Location <- c()
PMID <- c()
for (i in 1:length(symbol.blank)){
  sb <- symbol.blank[i]
  b <- grepl(sb, dat0@Abstract)
  if (sum(b) == 0){
    Number <- append(Number, sum(b))
    PMID <- append(PMID, "")
  } else {
    Number <- append(Number, sum(b))
    #         Symbol <- append(Symbol, symbol[i])
    PMID <- append(PMID, paste(dat0@PMID[b], collapse = "//"))
    print(sum(b))
    print(symbol[i])
  }
}

source("https://bioconductor.org/biocLite.R")
biocLite("org.Hs.eg.db")
library(org.Hs.eg.db)
entrez <- mget(symbol, org.Hs.egALIAS2EG)
chrom <- c()
for (i in 1:length(entrez)){
  print(i)
  chrom1 <- unlist(mget(entrez[[i]], org.Hs.egMAP))
  chrom <- append(chrom, paste(chrom1, collapse = "//"))
}

dat2 <- data.frame(Symbol = symbol, Name = name, Number, Location = chrom, PMID)
length(dat2$PMID == "")
write.table(dat2[order(Number, decreasing = T), ], output, sep="\t")