PMIDs database, semi-automatized

pmidata.2.R

#########################################################
#### CAPTURE ARTICLE INFO, IMPACT FACTORS FROM PMIDs ####
#########################################################

# Script to retrieve and organise publications data from PubMed (http://www.ncbi.nlm.nih.gov/pubmed)
# Uses the function ReadPubMed from the package RefManageR. It reads the PubMed API similarly to the search engine in PubMed's page.

# First, automagically install needed libraries:
list.of.packages <- c("RCurl", "RefManageR", "devtools", "plyr", "XML", "data.table")
new.packages <- list.of.packages[!(list.of.packages %in% installed.packages()[,"Package"])]
if(length(new.packages)) install.packages(new.packages, repos='https://cran.rstudio.com/')

# 1. # Import PMIDs
library(RCurl)
pmids <- read.csv("http://atriumkm.idisantiago.es/bin/ICT/PMIDs?xpage=plain&outputSyntax=plain")
# head(pmids)
# urli <- getURL("https://gist.githubusercontent.com/aurora-mareviv/14e5837814a8d8d47c20/raw/90b198bae82154688dcd9a2596af798612e6619f/pmids.csv", ssl.verifypeer = FALSE)
# pmids <- read.csv(textConnection(urli))

# 2. Loop several queries to PubMed and return in a data.frame
index <- pmids$pmId[1:length(pmids$pmId)]  #vector
index10 <- pmids$pmId[1:10]  #vector

library(RefManageR)
# my modified ReadPubMed function; it removes both "language" and "DOI" fields, to make easier to fit results in a dataframe.
source_url("https://gist.github.com/aurora-mareviv/8d5ee4c34d27cf084165/raw/4e66503e14ab767075fd743589ea797c01becd51/ReadPubMed.R") # workaround to source_gist error/bug?
library(devtools)
library(plyr)
library(XML)
auth.pm <- ldply(index, function(x){
            tmp <- unlist(ReadPubMed(x, database = "PubMed", mindate = 1950))
            tmp <- lapply(tmp, function(z) if(is(z, "person")) paste0(z, collapse = ",") else z)
            data.frame(tmp, stringsAsFactors = FALSE)
           })


# 3. Change weird symbols to make possible journal correspondences:
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="ñ",replacement="n"))
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="í",replacement="i"))
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="Ä­",replacement="i"))
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="é",replacement="e"))
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="á",replacement="a"))
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="ã",replacement="a"))
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="ä",replacement="a"))
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="â",replacement="a"))
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="ó",replacement="o"))
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="ò",replacement="o"))
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="ú",replacement="u"))
auth.pm <- as.data.frame(sapply(auth.pm,gsub,pattern="ü",replacement="u"))


# 4. Capture Impact factor and Journal abbreviations
# Import WOS data -- SEE THE SCRIPT: ifactors2014.R
source_gist("https://gist.github.com/aurora-mareviv/459401dc782e5f054767") # ifactors database
# Import journal abbreviations --  SEE THE SCRIPT: journaltab.R
source_gist("https://gist.github.com/aurora-mareviv/7680180cbe0283f87854") # journaltab database

# Capture journal abbreviations
dum <- merge(auth.pm, journaltab,by="journal", all.x=TRUE,ignore.case=TRUE)
# Refill empty ISSN.Print cells with info in ISSN.Online
dum$ISSN.Print <- as.character(dum$ISSN.Print)
dum$ISSN.Online <- as.character(dum$ISSN.Online)

dum$ISSN.Print.Online <- ifelse(dum$ISSN.Print=="",dum$ISSN.Online,dum$ISSN.Print) # Only ISSN.Online if Print is empty
dum$ISSN.Online.Print <- ifelse(dum$ISSN.Online=="",dum$ISSN.Print,dum$ISSN.Online) # Only ISSN.Print if Online is empty
dum$ISSN <- dum$ISSN.Print.Online # our preference to search for IFs: match with ISSN.Print, and Online if Print is empty

# Merge with impact factors:
auth.if <- merge(dum, ifactors,by="ISSN", all.x=TRUE)

# Correct ISSN to capture IFs driven by ISSN.Online:
auth.if$ISSN <- ifelse(is.na(auth.if$journal.bad), auth.if$ISSN.Online.Print, auth.if$ISSN)

# Re-merge with ifactors. Now we will have some junk-variables... but we capture as many IFs as we can!
auth.pmif <- merge(auth.if, ifactors, by="ISSN", all.x=TRUE)  # The field we are searching for is "auth.pmif$2013.2014.y"


# 5. Split auth.pmif$author in the authors:
library(data.table)
source_gist("https://gist.github.com/mrdwab/11380733") # function cSplit
auth.pmiff <- cSplit(auth.pmif, "author", ",")


# 6. Display data:
View(auth.pmiff)
auth.pmiff$X2013.2014.y <- as.integer(auth.pmiff$X2013.2014.y)
mean(auth.pmiff$X2013.2014.y, na.rm = TRUE)
hist(auth.pmiff$X2013.2014.y, breaks=60)


# 7. Write output to .csv
#write.csv(auth.pm, file = "auth.pm.csv")
#write.csv(auth.pmif, file = "auth.pmif.csv")
#write.csv(auth.pmiff, file = "auth.pmiff.csv")

Para que este script funcione correctamente, es necesario que los 3 ficheros fuente (lista de PMIDs pmids, tabla de Abreviaturas journaltab y tabla de Factores de impacto ifactors) estén con los datos correctos y el formato adecuado.
Los links de los que se descargan estos datos (NCBI, CiteFactor) pueden cambiar.