Friday, July 21, 2006

C&E News Open Source Science Article

An article on Open Source Science came out today in Chemical and Engineering News.

The UsefulChem Project got a mention:
Last summer the Drexel University chemist began two simultaneous experiments in his lab. The first: to synthesize potential inhibitors for the malarial enzyme enoyl reductase, which the malaria parasite uses to produce essential fatty acids. The second: to post online all raw data that his five students produce (usefulchem.wikispaces.com).

He's also recently joined Synaptic Leap. "There's not a lot of people doing this now, but I think open source is going to become an important way to do science," Bradley says.

He says people posting to his blog have helped improve experimental conditions. "We had one posting that suggested a different concentration, improving the reaction."

When it comes to the issue of publishing, Bradley argues that open-source discussions are similar to conferences, where people openly discuss unpublished research and do not fear being denied the right to publish in a journal, assuming the science is good.

Saturday, July 15, 2006

Subscribe to CMLRSS feed

As Dave mentioned recently, we now have an automatically generated CMLRSS feed for the UsefulChem molecules blog. Using VBA, the SMILES code from the molecules blog is extracted and MW, CML and InChI codes are computed, suppliers are searched online and an image is generated. Clicking on the image shows the molecule in the interactive JMOL format. Let us know what additional information would be useful.

I have added a Bloglines button for this feed on the molecules blog to make it easier for people to subscribe.

One current limitation of using Bloglines to view the feed is that only the last 100 entries are displayed.

Thursday, July 13, 2006

Patents in Open Source Science

Anonymous brings up an interesting point about the usefulness of patents in open source science, specifically in the case of a pending Chinese patent for the synthesis of praziquantel, an anti-schistosomiasis drug:
Why would you "avoid" patents when they are an outstanding (and often overlooked) source of technical information? Assuming (quite possibly incorrectly) that this process (which is the subject of only a pending patent and only in China) might really be useful to this quest - why in the world would you avoid it? Under what conditions might you access this technology? Might simply talking with the patent applicants yield new information - even that working with them is desirable (and may accelerate delivery of this critically needed compound)? Humble suggestion: keep all eyes on the prize, be innovative, don't cast anything aside because it doesn't fit with preconceived notions...

I'll give my opinion and then I'd like to hear from others (especially from the Synaptic Leap folks) who come across this issue.

Basically, the purpose of a patent is to prevent people from doing things without permission (generally for a fee) from the patent holder. That's what gives patents monetary and strategic value. It may be possible to get a special license from the patent holder for humanitarian applications but that sounds like an added layer of complexity and cost that is not present in open source public domain solutions.

The other issue, in my experience as a chemist, is that chemical patents are often written in a way that makes it difficult to reproduce some of the experiments. There are generally some detailed examples but much of the content tends to be written to be as broad as possible, covering reactions and conditions that have not been done. I've been through the patent writing process with attorneys a few times and it is a very different mindset compared to writing an article with the aim of truly sharing knowledge. I would be curious to see the details of this Chinese patent application but not curious enough to buy it - maybe someone has looked at it and can comment.

Concerning the issue of this being a Chinese patent application and thus not applicable to the rest of the world, that is a dangerous assumption. The only way to know is to search for similar filings in other countries. I would think that there is a good chance that there is a US filing if there is a Chinese one. But different countries have different timelines for disclosing patent filings. Basically it would require some legwork to really know.

Wednesday, July 12, 2006

Using existing drugs to fight malaria

Here is a gem highlighted in The Sceptical Chymist about a new approach to find anti-malarial compounds:
The authors created a library of 2,687 existing drugs and screened them for inhibitors of the malaria parasite Plasmodium falciparum.

The main idea is that if the authors could find a relatively potent inhibitor of multidrug-resistant parasites, the compounds could be put into human clinical trials very quickly (since many of the compounds are already FDA-approved for human use).
They did find an anti-histamine (brand name Hismanal) that was effective at inhibiting the malaria parasite. (In fact I'm pretty sure I took it 15 years ago to fight allergies before it was pulled from the market.) . An interview in Chemistry World suggests that it could still take a while to be used.
The general principle of trawling through approved drugs for additional benefits is a good idea, said Peter Winstanley, a malaria researcher at the University of Liverpool, UK. "‘It means that a lot of the preliminary work has been done, and that does make it cheaper," he said. But he cautions that even if drugs are already approved, they still have to go through fresh clinical trials for additional diseases before they are used, a process that could take five years or more.
It will be interesting to see how this plays out in practice.


Thursday, July 06, 2006

Full-text journal search

Chemspy.com is now running a Google hack as a search box that allows full-text journal search from a wide range of academic publishers (excluding Elsevier, ACS, and RSC, which you can search via Scirus and their respective websites separately). It's analogous to checking only the "Journals" box in Scirus but covers Nature Publishing, Karger, Wiley, Springer, IOP, APS, and numerous other publishers' journals. Give it a try and let me know how you get on with "real-life" searches, I've only tested it with dummy keywords like viagra and benzene so far and it seems to work well.

If this is already well-known, feel free to boot this post into the recycle bin.

db

Sunday, July 02, 2006

CML In RSS Feeds

As an addition to my 6/30 post, the RSS feeds for usefulchem-molecules and malaria37 contain CML which can be processed by molecular display software such as Jmol and Bioclipse. To load these feeds in Jmol, select "Open URL" from the File menu, and enter the feed address in the input box. Processing with Bioclipse is a little more complicated. First you must create an OPML file (File/New... -> CMLRSS -> New OPML Resource), and edit it to contain the following (if you are processing usefulchem-molecules):

<opml version="1.0">
<head>
<text>Useful Chemistry Molecules</text>
<outline text="Useful Chemistry Molecules">
<outline title="Useful Chemistry Molecules"
xmlurl="RSS feed URL">
description="Useful Chemistry Molecules" />
</outline>
</opml>


RSS feed URL should be set to the address of the usefulchem-molecules feed, in this case
http://showme.physics.drexel.edu/usefulchem/Software/MoleculeBlogInfo/usefulchem-molecules/usefulchem-molecules.rss. You can, of course, simply copy/paste the above code into the working directory (e.g., as C:\bioclipse\workspace\UsefulChem-Molecules.opml) of your Bioclipse installation (remember to enter the feed address!); alternatively, simply download the entire file from here. The Bioclipse Wiki site has a screen shot of this feature of Bioclipse. A Flash screencast of creating an OPML file can be found here, and another on how to use an OPML file to load CML resources from an RSS feed can be found here.

Creative Commons Attribution Share-Alike 2.5 License