Six alternatives to PubMed for searching scientific content
June 24th, 2007In my opinion, great coding skills, a thorough knowledge of statistics, and Shakespearian writing ability do not make a great bioinformatician. They help, but the most important things are a relevant scientific question and a good understanding of the literature. If you’re like me, the path to scientific enlightenment begins with typing keywords into PubMed until you get the results you were after - the same way you use Google. However the are other options besides PubMed, here are six other options you might not have heard of, worth a look perhaps?
Probably the most used scientific literature search engine after PubMed, Google Scholar has three great things going for it. The first is that results are sorted by relevance rather than by date order. This is useful if you’re looking for the most important articles in a field you’re not familiar with, not so much if you’re looking for the latest publications. The second useful feature is the option, for a given paper, to find related articles. For each result in a GS search there is the option to find articles that Google believes are related. For me this has turned up papers that I hadn’t found through keyword searching. Lastly, GS has the option to find all citing articles. Again this is different way to identify relevant papers, other than just by keyword searching.
This literature search engine is based on the data that PubMed has collected, but adds functionality. Like Google Scholar, HubMed results can be ordered by relevance, and related articles can be viewed. HubMed also has the option to visualise related articles using a graph view - nodes closer to the central article are more related. Another feature of HubMed clusters search results, the articles are categorised based on keywords. Finally, a feature that I use regularly is the option to save searches as an RSS feed. This means that I don’t have to check regularly for new articles - I get updated via feeds.
A Nature Group project, the aim is for scientists to highlight, organise and discuss literature. Users add articles that they’re interested in, and tag the research by adding words that best describe it. Connotea gives the option to add your own comments on each article, just for yourself or for everyone to read. Because papers are all tagged, it’s possible to search Connotea using these tags. The results of a search contain the papers users have thought interesting enough to highlight, with the possibility people may have added their own thoughts. A similar functionality is also provided by the Faculty of 1000.
As the name implies a BLAST like search for the literature. Take the abstract of your work, or of a paper you’re interested, paste it into the box then press go. After a few minutes you’ll get a list of articles ordered on their related-ness to the text you entered. Furthermore, continuing the resemblance with blast functionality, you can add select the results of the search that you consider relevant, and repeat including these. Hopefully presenting a more specific to-what-you’re-after set of results. ET Blast is also able to suggest journals to submit manuscripts to, and possible referees, based on the text you entered.
Often it can be useful to get early access to pre-publicised data or manuscripts, usually via collaborations or attending conferences. The Nature preprint server, Preceedings, now offers an additional avenue whereby authors upload unpublished manuscripts for early scientific consideration. Of course, these manuscripts aren’t peer reviewed and it’s up to you to use your discretion on interpreting the results and conclusions. There isn’t much on Proceedings at the moment, but hopefully it will be used by the our scientific community more and more in the future.
Not all scientific opinion appears only in journal type articles, scientists are beginning to publish more of their opinions on personal websites and blogs. Since you’re reading this on a blog you probably know that though. Scintilla and Post Genomic collect and aggregate these opinions for you perusal. Since these both follow scientific blogging opinion the subject matter is usually quite broad. You’re likely to find a fair amount of commentary on stem cells, but typing “amino acid synthetic cost on protein evolution” is likely to less successful.
Summary
The aim of this post was to highlight some of the other options for searching on content your research. PubMed is still going to be your first stop for keyword based literature searching. But I think it’s worth considering the options based on related articles and citing articles for the things you might have missed the first time around.
June 26th, 2007 at 6:38 pm
I try Google scholar occasionally, and always find it to be missing one crucial thing to make it useful - journal specific search operators.
Scientific searches are of two equally important kinds: keyword searches and author name searches.
Whereas Pubmed will return articles published by Bob Smith if you search for B Smith and, conversely, will return articles published by B Smith if you search for Bob Smith, Google Scholar can’t do that. Since different journals have different author naming conventions, this makes Google Scholar almost worthless for finding papers by name, and therefore a poor search engine for scientific papers.
Additionally, Pubmed now has RSS feeds for search queries, and has had “related items” for queries for quite some time. They’ve specifically disabled related items by RSS, though.
Another item of note is that sciencedirect.com now has “Cited By” rss feeds for any article in their database.
June 28th, 2007 at 11:17 am
Thanks for another useful set of feedback Mr. Gunn.
Google scholar does allow journal specific searches, this can be done in the advanced search section. But you’re right there’s not operator such as “journal:nature”, that you can use in the search box.
I wasn’t aware that PubMed has RSS options, I had a look before writing this post but I couldn’t find it. I like HubMed’s RSS because it’s really easy to set up, and as you say it’s possible to RSS feeds for related articles which is handy.
Thanks for the information about ScienceDirect citation feeds, this is another useful feature for keeping up to date on your favourite papers.
June 29th, 2007 at 5:30 pm
[...] Six alternatives to PubMed for searching scientific content (Bioinformatics Zen) However the are other options besides PubMed, here are six other options you might not have heard of, worth a look perhaps? [...]
June 29th, 2007 at 8:30 pm
scopus.com isn’t free - your institute must pay for a license, but I like the interface a lot.
July 1st, 2007 at 5:31 pm
I wasn’t aware that PubMed has RSS options, I had a look before writing this post but I couldn’t find it.
Run your search, then in the pull-down “Send to” menu beside it select “RSS feed”.
I believe you need to have a “My NCBI” account to take advantage of this.
July 2nd, 2007 at 1:27 am
[...] d’autres outils existent, permettant d’approfondir la recherche scientifique : Bioinformatics Zen donne plusieurs alternatives, dont voici quelques fonctionnalités traduites [...]
July 2nd, 2007 at 9:39 am
dans la liste des PUBMED alternatives, base sur pubmed:
* http://www.gopubmed.org/
Search PubMed and get results sorted by GO and MeSH
July 2nd, 2007 at 3:07 pm
@ Madelaine and Psychedelys, Thanks for your recommendations for other search tools.
@Ian, Thanks for clarifying how to get RSS from PubMed
July 3rd, 2007 at 4:12 am
[...] 2nd, 2007 by vidhyaj Michael Barton has a wonderful article published on six different alternatives to pubmed. My favourite is the [...]
July 5th, 2007 at 8:35 am
Hi Mike
I must admit I don’t use PubMed much, maybe because I’m a chemist. From your list I use Google Scholar and have looked at Scintilla, Connotea and Preceedings (my how the Natureplex grows) and they look very interesting. I think these resources will become far more useful as the number of users grows.
You didn’t mention the Web of Knowledge or Scirus. Any thoughts on these? Like all search engines they both have advantages and disadvantages.
For the WoK the biggest drawback is that you have to pay. It also, in my opinion, has less of a biological and a more physical slant than PubMed. But I think it has the best search options of any of the search engines I have used.
Scirus is more like Google scholar than PubMed and can be useful for finding MSc and PhD thesis’s that are not covered by the other engines.
PS your RSS feed from feedburner doesn’t seem to work.
July 9th, 2007 at 9:50 pm
Thanks for your comments Mike.
I’ve never used Scirus, I briefly used Web of Science during my undergraduate degree. I was a bit put off by the number of steps required before you can perform a search. All the one’s I’ve listed you can enter your simply enter your query into the box, and as you mention they’re free as well. One thing I remember about WoK, that was good, were the useful features for searching for related articles, based on citations.
I’ve been having problems with my RSS feed for a bit. I’m not really sure what’s happening, I regularly get messages from FeedBurner saying that there’s a problem, then later it’s been fixed. I think it might be that my server is returning the feed quite slowly.