Pubmed Trend via R

plot_bar.r

plot_bar <- function(x=sex.pub, linecol="royalblue", cols, addArg=TRUE) {
  bp <- barplot(x, col=cols, add=addArg)
  fit <- stats::lowess(x, f=1/3)
  lines(x=bp, fit$y, col=linecol, lwd=3)  
}

pubmed_trend.pl

#! /usr/bin/perl -w
#
# pubmed_trend.pl
#
# Created by David Ruau on 2011-02-17.
# Department of Pediatrics/Div. System Medicine Stanford University.
#
##################### USAGE #########################
#
# Query PubMed with Eutils tools
#
#####################################################
use Bio::TGen::EUtils;

use strict;

my $queryString = $ARGV[0];

## query info
my $eu = Bio::TGen::EUtils->new( 'tool' => 'pubmed_trend.pl',
                                 'email' => '[email protected]' );

## EFetch
my $query = $eu->esearch( db => 'pubmed',
                          term => $queryString,
                          usehistory => 'n' );

$query->write_raw( file => 'tempfile.xml' );

if (-z 'tempfile.xml') {
  # one more time
  my $query = $eu->esearch( db => 'pubmed',
                            term => $queryString,
                            usehistory => 'n' );

  $query->write_raw( file => 'tempfile.xml' );
  if (-z 'tempfile.xml') {
    open (FILE, '>', 'tempfile.xml') or die 'Could not open file, $!';
    
    print FILE "<begin>hello world</begin>";
    close (FILE);
  }
}

pubmed_trend.r

pubmed_trend <- function(search.str = 'Family+Nursing[mh] AND intervention', year.span=1990:2011) {
  require(XML)
  require(RCurl)
  
  results <- NULL
  tmpf <- "./tempfile.xml"
  ## clean before
  system(paste("rm", tmpf))

  for(i in year.span){
    queryString <- paste(search.str, ' AND ', i, '[dp]', sep="")
    print(paste('queryString:', queryString))
    sysString <- paste('./pubmed_trend.pl "', queryString,'"', sep="")
    system(sysString)

    xml <- xmlTreeParse(tmpf, useInternalNodes=TRUE)
    pubTerm <- as.numeric(xmlValue(getNodeSet(xml, "//Count")[[1]]))
    print(paste("#______num pub for",i,":",pubTerm))
    rm(xml)
    results <- append(results, pubTerm)
    ## avoid being kicked out!
    Sys.sleep(1)
  }
  names(results) <- year.span
  ## clean after
  system(paste("rm", tmpf))
  
  return(results)
}

Search-with-one-string.r

setwd('/home/produnis/Dokumente/Programmierung/R/PubMedTrend')# edit to fit your system
source('pubmed_trend.r')
source('plot_bar.r')
library("RColorBrewer")
#-------------------------------------------------------------
search.for1 <- "schnepp w[Author]"
#----------
search.pub1 <- pubmed_trend(search.str = search.for1, year.span=1990:2012)



#--------------------------------------------------------------
pdf(file='output-one-string.pdf', height=8, width=8)
par(las=1)
colorfunction = colorRampPalette(brewer.pal(9, "Blues"))
mycolors = colorfunction(length(search.pub1))
plot_bar(x=search.pub1, linecol="#525252", cols=mycolors, addArg=FALSE)


title('Number of publication per year')
legend('topleft',
legend=c(search.for1),
  fill=c("blue"),
  bty="n",
  cex=1.1
)
dev.off()


#END OF FILE

Search-with-two-strings.r

setwd('/home/produnis/Dokumente/Programmierung/R/PubMedTrend') # edit to fit your system
source('pubmed_trend.r')
source('plot_bar.r')
library("RColorBrewer")
#-------------------------------------------------------------
search.for1 <- "Midwifery[mh]"
search.for2 <- "Family+Nursing[mh]"
#----------
search.pub1 <- pubmed_trend(search.str = search.for1, year.span=1960:2012)
search.pub2 <- pubmed_trend(search.str = search.for2, year.span=1960:2012)



#--------------------------------------------------------------
pdf(file='output-two-string.pdf', height=8, width=8)
par(las=1)
colorfunction = colorRampPalette(brewer.pal(9, "Reds"))
mycolors = colorfunction(length(search.pub1))
plot_bar(x=search.pub1, linecol="#525252", cols=mycolors, addArg=FALSE)

colorfunction = colorRampPalette(brewer.pal(9, "Blues"))
mycolors = colorfunction(length(search.pub2))
plot_bar(x=search.pub2, linecol='black', cols=mycolors, addArg=TRUE)
title('Number of publication per year')
legend('topleft',
legend=c(search.for1, search.for2),
  fill=c("red", "blue"),
  bty="n",
  cex=1.1
)
dev.off()


#END OF FILE

On Ubuntu, install XML and Curl support.
Otherwise, installation of R-Packages will fail
On OSX, skip this step

sudo apt-get install libxml2-dev curl libcurl4-openssl-dev

the perl script needs the TGEN-EUtils
download package TGen-EUtils-0.xxx.tar.gz from http://bioinformatics.tgen.org/brunit/downloads/tgen-eutils/
extract it to some place you like, e.g. ~/Downloads/
in terminal go to the extracted folder
type in:

perl Makefile.PL

and then

make

and then

make test

and finally

sudo make install

Start R and install additionally packages 'XML', 'RCurl', 'RColorBrewer'

install.packages(c('XML', 'RCurl', 'RColorBrewer'),dependencies=T)

don't forget to make the perl-script executable
open terminal
go to the folder where you saved these pubmed-scripts

cd /path/to/pubmedscripts/

chmod +x pubmed_trend.pl

enjoy

thx to bobthecat who originally postet this hack at http://brainchronicle.blogspot.com/2012/05/using-r-to-graph-subject-trend-in.html