In the blogpost regarding Wikipedia Services yesterday I discussed “Domoic Acid“. Domoic Acid is very well documented in the literature and I would expect the structure to be well known. On ChemSpider the structure has been curated and is believed to be that shown below.

On Wikipedia the structure is lacking the stereocenter on the side chain as shown below


The April 6, 2009 issue of C&E News had an article on Page 27 in the Science and Technology Concentrates about “Algal Neurotoxin Lingers in the Ocean“. Unless you are an ACS member and have an ACS ID you won’t be able to read the article. However, the structure from the article is shown below.  Do you notice a similarity between the structures?domoic-acid-on-cenews

Unfortunately, both are wrong. They are both lacking the side chain stereocenter for the methyl group based on my research.  Previously I had been using C&E News as a source of news about chemical compounds and association with records on ChemSpider. On a couple of occasions however I observed that the structures were wrong. Since C&E News is an ACS magazine I had assumed that the writers would have access to Scifinder to get the correct structures. Since the structure is wrong maybe it’s wrong in Scifinder (!).

In theory the presence of an article on Wikipedia means a related page will exist on Unfortunately the does NOT have all Wikipedia structures. The estimated overlap is somewhere between 50-70%. Fortunately someone had already checked the structure of Domoic Acid on Scifinder and confirmed to me that the curated structure on ChemSpider is “consistent” with that on Scifinder…let’s assume that this means its correct. I did actually confirm that structure at MANY other sites too.

So, the structure in C&E News is identical, both in layout and in chemistry, to that on Wikipedia but is NOT consistent with that in SciFinder. Surely C&E News is not sourcing their chemical structures from Wikipedia when they have access to the most highly curated compound database available?

Note to C&E News reporters…there is a LOT of work going on to validate and curate the ChemBoxes and DrugBoxes on Wikipedia but the work is not complete yet. I recommend using SciFinder to source your chemical structures for now.

Reblog this post [with Zemanta]

Stumble it!

6 Responses to “Where does C&E News Source Its Chemical Structures?”

  1. Egon Willighagen says:

    I noted that the WP entry has not been updated yet, and that made me wonder how you ChemSpider is curating WP…

  2. Egon Willighagen says:

    I saw a *very* *long* list of literature in the ChemSpider entry for domoic acid… have you spotted the primary literature about the domoic acid identity? That is, the literature where the structures and stereochemistry was published?

    That would make a very valuable piece of information in ChemSpider, if that could be singled out.

  3. Antony Williams says:

    Yes..a long list of literature…I’m amazed what’s out there.

    There are a number of articles now added to the supplementary info with DOI links just for you

    By the way…can you spot the error at this page?

  4. Steve Ritter says:

    Thanks to Antony for pointing out the mistake in the C&EN structure. I did inadvertently leave out the stereochemistry at the methyl on the side chain, and the geometry for one of the double bonds is incorrect.

    We publish several hundred structures per year in our 51 print issues and on our website, and inevitably we get some wrong–on the average five or fewer per year that I am aware of. We are grateful to our readers for pointing out the mistakes. In this case, a revised structure is being posted on our website and a correction will run in an upcoming print edition. Please check to see that the new structure is correct.

    As for where we source our structures, our primary source is the researcher and peer-reviewed papers, because many compounds are novel. For known compounds, knowing that those can sometimes be wrong in papers, we always double check them against one or more primary sources, typically Merck Index and SciFinder. Although CAS and C&EN are both part of the ACS Publications Division, we at C&EN still have to pay for our SciFinder access, strangely enough.

    To tell a woeful story, one that demonstrates it is never easy to make sure a structure is “correct,” I received a structure of domoic acid from the researcher I wrote about, as there was not one in the paper. But the structure was wrong–it was missing a methylene in one of the short carboxylic acid side chains. The researcher was not aware of that until I pointed it out, and that structure had been used in several published papers already. I noticed the error by checking the structure in the Merck Index.

    When it came time for our artist to draw the structure, I did not really like its orientation in the versions I had. I checked SciFinder, and the structure there is identical to the Merck version, but SciFinder does indicate the absolute stereochemistry. I also checked the Web, and found the Wikipedia entry and several other references with the structure. As Antony noted, domoic acid is well known in the literature, but one sees it drawn myriad ways. I liked the orientation of the Wikipedia entry the best, and used that as a model to draw out the structure by hand for our artist to redraw. I checked my version against Merck, but I was focusing on the double bond geometry and missed the stereocenter when I drew it. That’s the long-winded version.

    It’s embarassing to make any kind of mistake, especially in C&EN. But it is a bit more so for me because every structure that appears in C&EN comes across my desk for scrutiny. It’s not the first time I missed something in a structure, and probably isn’t the last. We have a great staff of writers and editors that make such mistakes rare.

    As a rule, we at C&EN don’t use Wikipedia as a primary source for structures or chemical information, and I recommend that policy to anyone. We don’t even use articles or structures previously published in C&EN as a primary source without rechecking, in case we made a mistake the first time around. The only two sources for checking structures that I really trust are Merck Index and SciFinder, with Merck being a little better because sometimes the SciFinder structures are drawn awkwardly, but that is just my personal opinion.

    It would be nice to have an authoritative web-based source of standard, well-drawn structures for chemists to go to so they can freely cut and paste structures into their papers, PowerPoint presentations, and anything else they might need. Maybe Wikipedia will be that source one day.

    As for the structure of domoic acid on the NOAA page that Antony noted, I believe the stereochemistry for each of the three ring carbons is backward.

    Apologies for rambling on, but thanks again for pointing out our mistake. We at C&EN know we are considered authoritative and held to a high standard by the chemistry community that we serve, and accuracy is paramount to maintaining that trust. We take our responsibility seriously.

  5. Antony Williams says:

    ChemSpider is doing a lot of work on the curation process. Many of these structures are very complex and after my side of the work I share them with other people on the Wikipedia chemistry team for further validation before they get uploaded. The image has now been uploaded onto Wikipedia and Wikimedia Commons. Interestingly enough a new image had already been uploaded there today and it was incorrect. It corrected the stereo for the methyl group only and not the orientation of the double bond. The full discussion is here:

  6. ChemSpider Blog » Blog Archive » Micrococcin P1 and C&E News Structures says:

    [...] is the nature of the beast. I have blogged recently about an observation of a structure drawing error in C&E News and the editor was kind enough to comment. Here’s an image of a structure from a C&E News [...]

Leave a Reply