Systems Biology Research Blog

It isn't a secret that one basic prerequisite in any research is to be up to date by reading the newest publications. To get knowledge of a new article, journals offering email alerts. The interdisciplinary character of Systems Biology causes a fragmentation of publications to a number of journals. Journals dedicated to biology, physics, mathematics, computer science or a combination of it, may publish articles related to Systems Biology. However, if I would subscribe to all of the offered email notification my mailbox will be flooded. Thats why I prefer RSS feeds. Since I updated my laptop to the new KDE4 desktop, I use a plasmoid to get notified for new articles (see screenshot).

The KDE Desktop includes a plasmoid (widget) to read RSS feeds. It allows to stay updated with the newest research issues.

Unfortunately it's not always easy to find a feed from the publisher website. That motivates me to investigate the provided web feeds of some major journals related to Systems Biology. The results varies from a valid feed containing the abstract to no feed at all. Additionally I'll offer an OPML file containing the RSS feed for an easy import to an appropriated feed reader.

Follow up:

The organisation of the huge mass of information provided by the Internet motivates to use software for an easy handling. In particular Web2.0 tools like blogs and wikis allows user created contend. This and also the increase of information given by more classical news sites makes it complicated to stay up to date. It is not useful to visit every site only to check if new posts or articles are published. An alternative is the use of web feeds, which can be read with an feed reader software. A clever employment of this technique can reduce the bandwidth of transmitted data and recruit desired user. This is also true for scientific website.

The chosen journals are taken from the list of the website from the SBI-group Rostock. It can be assumed, that the list is not complete and every scientist have particular needs, but the major journals should be included. The validation of the feeds have been analysed by using the W3C Feed Validator. To read the information of the particular feed the software aggregator of the KDE Personal Information Manager Kontact version 1.5.1 and Liferea 1.6.0 have been used. The later one were also used to create the OPML file.

The first journal is "IET Systems Biology", a journal specialised in Systems Biology provided by "The Institution of Engineering and Technology". The website is hosted on the Scitation platform, provided by AIP. This is the only publisher website, were I didn't found a web feed at all. This is astonishing, because the website of the Scitation platform itself provides this function (whether the validator fails). This is seriously not state of the art and should be implemented as fast as possible, especially if the name of the institution contains the buzzwords technology and engineering.

The next website is the "Proccedings of the National Academy of Sciences of the United States of America" (PNAS). The website provides different web feeds, f.i. Current Issue and Recent Issue. The validation check for the Current Issue feed shows, that it isn't valid. The second disadvantage is the mixed content of the issue. If you subscribe to the feed, you will get also information about Ecology or pure Chemistry. This makes the feed crowded. Beside the authors of the article, the very first lines of the abstract will be delivered.

The ScienceDirect website, maintained by the Elsevier B.V company, contains several journals related to Systems Biology including web feeds for new articles, e.g. Journal of Theoretical Biology, Journal of Mathematical Biosciences and FEBS Letters. Because they use the same backend, the web feeds are comparable and all are not valid. The feed includes the publication year, the name of the journal, status, the authors and mostly the abstract. Unfortunatly, all of the information are in the description tag! It is readable for humans, but totally useless for a computer program. The journals have also own websites on Elseviers mainpage. No web feeds where found, with one exception: The Biophysical Journal from Cell Press, an imprint of Elsevier, provide a syntactical valid web feed, but contains also the semantic misinterpretation of the description tag. Instead of an abstract, only the very first words of it are provided.

The Journal of Mathematical Biology published by the Springer Science Publisher, provides also a not valid web feed. Because the validator is case sensitive, it produces some errors for an existing attribute, but there are also missing elements which makes the feed invalid. The RSS2.0 feed is well organised and contains full abstract and also the DOI number, which is heavy useful. The authors are again in the description tag.

One additional journal is Molecular Bioscience. The validator produces several errors, including unknown name spaces. The web feed contains the first part of the abstract. Again, the authors are not equally set and are only written in the description tag.

Another invalid web feed is given by the Journal of The Royal Society Interface a journal with different topics and free articles. The date in the xml file and one name space are marked as an error. This seems only minor errors, nevertheless its not valid. An advantage is that the date and the authors of an article are written in date and creator tags. The full abstract is also included. If the developer of the website may correct the named errors, the feed will become a useful service.

The Molecular Systems Biology Journal from Nature provides an valid stream, nevertheless the validator have suggestions to improve the feed. The article information containing the authors, each author in an own creator tag. My feed reader can't show more then one, but it seems to be correct and only an issue for the aggregator software. Additionally the DOI number, some journal informations and a second list of the authors are included. Unfortunately the abstract is excluded. The artikel topic is also again in the content field, so that the feed contains several redundant information. The articles are published under an Open Access license (Creative Common).

The last web feed, that I use comes from the Open Access BMC Systems Biology Journal. This feed is also valid. The validator shows also suggestions and the feed contains again more tags then my aggregator can handle. The full abstract is included.

To summarise the results: it is disappointing how a well established technique is implemented in journal websites. It is remarkable, that web feeds from journals supporting Open Access or equal licenses seeming more comprehensive. Unexpectedly, my feed aggregators are not sufficient to handle feeds from scientific journals properly. But it makes also no sense to start developing a better one or improve the existing, because every publisher is providing an own "interpretation" of a well defined specification. In my opinion it is time for an agreement, what information are provided by web feeds. Preliminary a valid realisation of the specification is necessary!

Download OPML File including the web feeds from this post