dwinter
commented
9 years ago Hi @Monty9, thanks for your interest in rentrez and your questions. Let's take them one at a time.
- I don't think there is a way to filter the results of
entrez_searchthat is equivalent to the interactive filtering online. Instead, you can either, (a) build new queries that narrow down you results or (b) fetch entrez_summary records for every match, and filter the results based on your criteria. So, for instance, you could make your search
big_search <- entrez_search(db="pubmed", term="BRAF AND LUNG")
big_searchEntrez search result with 545 hits (object contains 20 IDs and no web_history object)
Search term (as translated): BRAF[All Fields] AND ("lung"[MeSH Terms] OR "lung" ... That seems like too many, let's get only the recent ones (the fields denoted by [ are described in the new vignette and documented here.
recent_search <- entrez_search(db="pubmed", term="(BRAF) AND (LUNG) AND 2014:2015[PDAT]")
recent_searchEntrez search result with 250 hits (object contains 20 IDs and no web_history object)
Search term (as translated): BRAF[All Fields] AND ("lung"[MeSH Terms] OR "lung" ..If you'd rather get them all and filter you could instead fetch summariesfor each (I'm just using 20 records, use retmax= or use_history to work with all the hits). This below uses the development version on rentrez (installed with devtools::install_github('ropensci/rentrez')):
braf_summs <- entrez_summary(db="pubmed", id=recent_search$ids)
paper_types <- extract_from_esummary(braf_summs, "pubtype")This will give you a list of the "paper type" field. You could use this to find only the records that are reviews:
sapply(paper_types, function(x) "Review" %in% x)26288737
3 - Pubmed only knows about papers that are included in pubmed central, which is only those papers in (very broadly) biomedical fields and being open access. There are lots of closed papers and papers published in other journals that won't be included in the pmc citations but will get sucked up by Google Scholar.
- You get to use one of the new things I've been working on :smile:. In the development version
entrez_linkcan give you this information. Working on this made it clear that what we have at the moment is a bit messy, but here's a work around that should do the trick.prlinksgives us the primary link for each record,extract_urlis just a function to get them cleanly from the resulting list:
extract_url <- function(linkout){
if (length(linkout) == 0){
return(NA)
}
linkout[[1]][["Url"]]
}
links <- entrez_link(dbfrom="pubmed", id = recent_search$ids, cmd="prlinks"
urls <- sapply(links$linkouts, extract_url)
urls[3] ID_26288737
"http://www.immunotherapyofcancer.org/content/3/1/36Hope that's some help to you, and feel free to ask more questions if you have them
gadepallivs
Hi David Winter, Thank you for the amazhing entrez package. 1) How to input filters in entrez_search funtion ? For instance, When we search pubmed online for BRAF AND LUNG, it results in some summary hits. On the left hand side of the page, we can select article types, text availability year of publication, species and so on. Selecting one of these helps to filter the summary output further. How do I input these parameters in entrez_search function ?
2) When we extract citation information for particular article ( Eg: Cited by 48 PubMed Central articles). Why is the count different from google scholar citations ?
3) How do we extract the links for the articles we fetched from pubmed using rentrez ? For Eg. I am trying to make a dataframe for the feteched pubmed abstracts, I would like to inluce a hyperlink in one column so that, user can click on the hyperlink that opens the fulltext or the abstract.
4) Finally, is there a way impact factor of the journal be extracted from NCBI. Not sure, If I saw impact factor listed in NCBI in general. If you are aware of an other packages that help to do so, please let me know. I am trying to create a table with NCBI abstract info, impact factor of journal and the link to the journal full text.
Thank you very much