Tuesday, December 23, 2008
Yale Image Finder
It searches the text within the images, but you can also search on image captions or on the article text. I tried it using the search string streptococcus pneumoniae. Searching the default of "image text (high recall)" found many genome and phylogenetic tree images, and a screenshot of the ISABEL diagnostic program. Searching "image text (high precision)" found the same (I assume) 18 images. Searching captions instead found more, including an image of S. pneumoniae infection. You can search within, say, captions and full text.
There is more detail on how this works in a paper in Bioinformatics.
BioText, from Berkeley, searches image captions and article full text.
Tidying my desk I found a paper from Bioinformatics from 2004 describing FigSearch, a text mining system for figures from full text biological papers, developed at the University of Oslo. But the URL is no longer found, and a search of the UiO site finds nothing. A Google search for FigSearch finds lots of things about it, but not it itself. If you know where it is, please leave me a comment.
Tuesday, December 02, 2008
Caesarian increases asthma risk
The study is: well, I can't actually find it in Thorax, or anywhere else, so perhaps the BBC (and all the other media sources that appear in a search of G**gle for the author's name and the word "asthma") have got an advance copy.
I will try to come back to this another day and fill in the gaps. Meanwhile if you can find the Thorax paper, leave me a comment. Especially if I have missed it.
Monday, December 01, 2008
Any of the associations are links so that you can see their own associations.
FACTA is described in a open access paper in Bioinformatics:
Yoshimasa Tsuruoka, Jun'ichi Tsujii, and Sophia Ananiadou. 2008. FACTA: a text search engine for finding associated biomedical concepts, Bioinformatics, Vol. 24, No. 21, pp. 2559-2560 (there is a link on the FACTA site to this).
The Personal Genome Project wants to recruit people willing to share their own genome sequence in the interest of furthering research. 10 people have enrolled so far and the website tells you about them, and about who runs and funds the project.
Gene Gateway (Exploring Genes and Genetic Disorders) is a site from the Oak Ridge National Laboratory. There is a poster that outlines disorders that have been mapped to particular chromosomes, and the website offers a workbook and materials that accompany that. The very useful looking workbook uses hereditary h(a)emochromatosis as an example to show how to use OMIM, NCBI Viewer, Swiss-Prot, and other things. ORNL, in Tennessee, was founded during World War 2 to carry out plutonium production for the Manhattan Project.
The Mathematical Association of America's Minutemath website is rather fun, especially because I can actually do some of the problems. There is a daily problem, with an interactive solution which can offer hints before you see the answer, and which gives you an idea of the difficulty (which could be quite depressing, I guess). The MAA also have a Number a Day site, which chooses a different number each day and gives you a fact about it or a property of it.
Google Scholar and material in repositories
Google Scholar will now give a link to freely available versions of an item, with this link appearing next to the title of the item. It will still tell you how many versions of an item it has found, and if you click that information, give links to all those versions, whether free or not, but next to the title, in your main results list, there will be a link to any free version.
Wouter's blogpost has a screenshot showing an item with a link to the repository of the University of Utrecht. I tried this with the search calcium signalling bacteria, and this finds a copy of the paper by Norris et al in the Journal of Bacteriology in 1996 in Leicester Research Archive.
This, I think, is a very interesting development. There are of course search engines that search only open access material, and I have started slipping OAIster into some classes that I do, but this development to a search engine that we know a lot of people already use is very interesting. Will it push up traffic from Google Scholar to LRA - we shall see.
Labels: open access;