I do not remember exactly, but I guess we must have met on the 28th and 29th? Maybe already on Wednesday. During this meeting we discussed a common data model (yes, Jmol used the CDK data model at some point) and somewhere during the meeting we wrote down a name for the project. There was the Java Development Kit, so this could be the Chemistry Development Kit. The name stuck.

A quick post like this cannot do credit to the history of the CDK, nor of everyone involved in the past or still is. You can browse some of the history of the CDK in my blog and in Chris’ blog. It has been an amazing journey and with a small grant just behind us (with Alyanne de Haan, René van der Ploeg, and Marc Teunis from Hogeschool Utrecht), and all the awesome things ongoing (new JChemPaint, various extensions, upgraded downstream tools), the CDK is alive and kicking.

A huge congrats and thanks to everyone (and every company and organization) who contributed code to the CDK with this huge milestone. There are a few people that I want to particularly thank (see the AUTHORS file for all names): Chris, who in the late nineties made a difference with open source in chemistry, Dan, for Jmol and hosting this memorable meeting at Notre Dame University, Rajarshi Guha, who operated CDK Nightly for many years, well before Travis and Google Actions, Stefan, Miguel, Gilleain, and Christian, for many years of contributions to the CDK, and John Mayfield, the current CDK release manager.

Now, addressing the limitations of the current citation databases is technically simple, and purely blocked by social and commercial aspects. The Citation Typing Ontology by David Shotton defines the framework to define citation types, independent from any existing database. The semantic web technologies will take it from there, and allow aggregation etc.

There are some things to think about on how to use such citation networks, though. If we calculate the impact of the CDK project, we should combine citation counts to the website(s), papers, etc, after removal of duplicates, etc. The cito:cites does link to resources, and the CDK paper resources is not the same as the CDK website resource. But, we could define a Project Class, where both are foo:partOf. Then, we could define that the triple chain the:citingWork cito:cites the:CDKArticle foo:partOf the:CDKProject would imply the triple the:citingWork cito:cites the:CDKProject.

Typed Citations

Now, while writing up this blog, I realize that my fork of this morning, A BIBO Citation Typing Ontology, might actually be counter-productive in the long run, as I was only working out a solution to a simpler, but different problem, which the CiTO also addresses: a citation is not typed. When a paper does cite the CDK paper, we still do not know if it uses the CDK, or merely mentioned it as related-but-unused, or even refuted work.

Now, as I am leaning towards the Biobliography Ontology as RDF-based system for my references, and been using this already in the RDF store hosting the ChEMBL data, I forked the CiTO to define rdfs:domain and rdfs:range on bibo:Document. The CiTO 1.5 actually defines a large set of document types too, and I rather see BIBO reused.

This indeed has the downside that the bibocto:cites cannot be used for the above chaining, and this might bite me seriously later. Well, nothing wrong with a failing experiment, right? For now, it will serve my purpose: setting up a citation database for the CDK project papers.

The CDK citation database

So, here goes (it’s RDFa-enabled; check this RDF pulled out):

@prefix bibo: <http://purl.org/ontology/bibo/>.
@prefix bibocto: <http://github.com/egonw/bibo-cto/>.

<urn:doi:10.1186/1758-2946-2-1> a bibo:Article ;
  bibocto:cites <urn:doi:10.1021/ci025584y> .

I am not entirely happy about the error-prone XHTML+RDFa of the above example, and filed a question of better solution on SemanticOverflow.

While the above example merely defines the citation of Peter Ertl’s article to the CDK (whether that is valid or not… would he have cited the other paper perhaps?), the citation typing allows me to state how the CDK paper is cited. Now, Peter states:

It is also gratifying to see the advent of open source movement in cheminformatics on the Internet, as advocated for example by the Blue Obelisk Group (40) and witnessed by collaborative projects like Chemistry Development Kit CDK (41), Jmol (42), Bioclipse (43) and several others.

So, I think it is fair to state that:

<urn:doi:10.1186/1758-2946-2-1> bibocto:credits <urn:doi:10.1021/ci025584y> .

which is very much appreciated!

The SWAT4LS was fun, interesting, perhaps to short, but very much appreciated! Thanx to all organizers! During the day various people tweeted the meeting, using the #swat4ls2009 hashtag (forwarded to a FriendFeed room), while Nico covered things in various blog posts which I’ll link to below where appropriate. Summaries I have seen so far are from Nico and Duncan (again :), and the organizers.

The day kicked off with a presentation by Alan Ruttenberg (Nico’s coverage). It nicely demonstrated where the semantic web for life sciences is going too. Particularly interesting was the integration of SPARQL with Jmol in ImmPort/JmolViz: it uses Jmol to visualize a PDB entry, while using SPARQL to retrieve atomic and residue annotation, using Jmol script (we have to thank another Miguel (the Jmol one) for taking the scripting and visualization capabilities to the next level in 2002). It always makes me proud to see one of the projects I have worked on to hit a prominent place in keynote talks at conferences :)

Alan also clarified that CC0 is not a license, but a statement about the public domain nature of data; there is nothing to accept, nothing to live up to. The important is, and I am sure most of my readers are well aware of that, is that it formalized the public domain concept by wrapping it in a full CC0 statement. My recommendation to all who want to make (chemical data) available as public domain, use the CC0; just because the CC0 works in any country, and it will make a lot of your users very happy. If you cannot claim CC0 because you are not really owner (as I have seen done), do not claim the data to be public domain either then (which was done)!

There was also note of the Amino Acid Ontology, which comes closer to our groups proteochemometrics work, but I have yet to look if this can be used for or linked protein descriptors. Also interesting is the idea behind RDFHerd , a project aiming to distribute RDF data sets as installable packages. If I understood correctly, only Virtuoso is yet supported, but this thing can fly, particularly, if these packages are easily converted into Debian packages.

More wrapping up will follow, but got other business to do first now.

To strengthen our intentions towards the 3D cheminformatics world, we have implemented a few new features, using CDK functionality. For example, we added Kabsch aligment and the related RMSD between molecular structures implemented as both popup menus as well as manager methods. The manager method you can see in action in MyExperiment workflow 937, which you can download directly into Bioclipse with one simple command (see Bioclipse Manager for MyExperiment.org ):

var smileses = new Array("CC(C)C", "CCCN", "CCC=O");

var unaligned = cdk.createMoleculeList();
for (i=0; i<smileses.length; i++) {
  mol = cdk.fromSMILES(smileses[i]);
  mol = cdk.generate3dCoordinates(mol)
  unaligned.add(mol);
}

var aligned = cdk.kabsch(unaligned)

jmol.load(aligned.get(0));
for (i=1; i<aligned.size(); i++) {
  jmol.append(aligned.get(i));
}

Now, we do have to update the use of Jmol in Bioclipse, and a big overhaul is scheduled for the 2.4 released in February next year. But you get the idea.

As said, there are two stories to adding this new functionality. Because we want all GUI interaction the user performs to be recordable (Scientist 1: What did you do to get those nice results? Scientist 2: I pushed that button in the that long menu. Scientist 1: What button is that? Scientist 2: Wait, I send you the BSL script with a Google Wave.)

The managers that allow this recording is Bioclipse specific, and also the reason why it would not be trivial to make a general Bioclipse plugin for Eclipse… some Spring magic is used to inject the managers into the JavaScript language. Anyway, the second thing is to add a GUI element, like popup menus. Now, this is a particular area where Eclipse excels. Now, I did have to ask for the details, as I am not using this daily (I’m doing science, not IT), but Ola was kind enough to give me the pointers for it.

The below configuration snippet links the pop up action to Bioclipse Navigator content (you know, where your MDL SD, CML, script and other files show up in Bioclipse). But only if I have selected 3 or more files! And, only if those files are actually some molecular content with 3D coordinates! And Bioclipse inherits this functionality by using the Eclipse platform.

<menuContribution
  locationURI="popup:org.eclipse.ui.popup.any?after=additions">
  <command
    commandId="net.bioclipse.cdk.ui.handlers.kabschAlignment"
    label="Perform Kabsch Alignment"
    icon="icons/molecule2D.png">
    <visibleWhen>
      <with variable="selection">
        <count value="(2-"/>
        <iterate operator="and" ifEmpty="false">
          <adapt type="org.eclipse.core.resources.IResource">
            <or>
              <test property="org.eclipse.core.resources.contentTypeId"
                       value="net.bioclipse.contenttypes.cml.singleMolecule3d"/>
              <test property="org.eclipse.core.resources.contentTypeId"
                       value="net.bioclipse.contenttypes.cml.singleMolecule5d"/>
              <test property="org.eclipse.core.resources.contentTypeId"
                       value="net.bioclipse.contenttypes.mdlMolFile3D"/>
            </or>
          </adapt>
        </iterate>
      </with>
    </visibleWhen>
  </command>
</menuContribution>

When Bioclipse is run, this looks like:

And the alignment results will nicely show up in a Jmol viewer (while it is implemented as an Eclipse editor, it is not yet):

The first screenshot also shows the new pop-up menus for calculating two matrices for 3 or more molecules. One is based on the RMSD of the 3D atomic coordinats of the atoms in the MCSS (BTW, Asad’s SMSD work is making its way into the CDK library, and will be available in a later Bioclipse version too.) and will create a distance matrix. The second new pop-up menu used the Tanimoto similarity measure based on CDK fingerprints on the selected chemical graphs. If the Bioclipse Statistics feature is installed, the created CSV files will open up in a matrix editor:

Kabsch alignment of protein backbones is planned for a later Bioclipse release, but an important feature for our groups proteochemometrics work.

Now, the several blogs and the Slashdot item contain interesting discussions on whether the ‘Dr Who’ model is the best model of how open source projects can evolve. Fact is, at least, that the model does not describe a new phenomenon; Peter merely describes a how the Blue Obelisk deals with the limited resources we have in cheminformatics, and that the succession of project leaders ensures both the scientists interest (who are generally not payed for development or, $Foo forbid, maintenance of scientific data analysis methods) as well as the project itself. This makes life science open source different from most pure-IT projects: open source academic software is always something on the side.

So, when Miguel Howard turned to Jmol, he had seemingly unlimited resources to work on Jmol and he had great ideas and made them work: Miguel is the father of the now so popular Jmol applet with scripting functionality. It did mean that the integration with the CDK I worked on, as planned by the original Jmol author Dan Gezelter, Christoph and me in 2000: the CDK data model was too slow (it is amazing how fast Jmol is, without using accelerated graphics! See this Nature Preceedings paper: DOI:10.1038/npre.2007.50.1). My attention was better spend on the CDK.

Now, if the need arises, and the current Jmol head Bob looses interest or time, I’ll be available to take over again. That is less likely to happen for an older Dr. Who actor. Several Slashdot commenters also pointed out that the model also matches the ‘drummer-in-a-band’ model. I guess, or lead-singer… This moved the discussion of what the model exactly models. Peter writes:

“Instead the Blue Obelisk community seems to have evolved a “Doctor Who” model. You’ll recall that every few years something fatal happens to the Doctor and you think he is going to die and there will never be another series. Then he regenerates. The new Doctor has a different personality, a different philosophy (though always on the side of good). It is never clear how long any Doctor will remain unregenerated or who will come after him. And this is a common theme in the Blue Obelisk.”

This brings me back to the earlier observation I wrote down: science is different, and Peter is right when he says you think he is going to die and there will never be another series. This thought is justified for many open source science projects; in Glyn’s blog there is the remark of lack of data, but I think if someone would count of the number of dead open source science projects, I think the outcome will be that the fear is highly justified.

This is likely also the power of the Blue Obelisk: it creates a lively and rewarding community with equally minded people, forming an eco-system where the individual projects can flourish. Maybe someone can come of with a good metaphore for the Blue Obelisk, matching the Dr Who model? BBC comes to mind: is the BBC an eco-system where small TV series can survive?

Anyways, my father used to watch Dr Who, and being compared to Dr Who is much more rewarding than being compared to a drummer in some band.

Today, Neil asked me to look at another Nature Chemistry paper (DOI:10.1038/nchem.100), and in particular its Chemical Compounds table. I could not directly spot the thing not in the table I discussed, but did notice the phosphate salts in the table. Not uncommonly, the counter ions are not near the phosphate in this diagram and I wondered how they did this in 3D.

Well, bringing back good memories to that internship I mentioned, the 3D model shown by Jmol actually does show the salt, and with the two sodiums near the phosphate; even better, they sit at very recognisable positions :)

Now, we’ve seen the Royal Society of Chemistry’s Project Prospect (see RSC: the first publisher to go semantic! ) and ChemSpiders recent ChemMantis system which enriches the papers with machine readable representations of the molecules discussed in those papers. The new Nature publication has been in the works for a while, and they asked the community before what a Nature Chemistry paper should like like, and I replied in Re: What should a Nature Chemistry paper look like? .

The verdict

So, have the been listening? Is the HTML they produce semantic? Is it data rich? Or is it just another hamburger? Well, I am very happy to see some of the suggestions I made picked up (though I do not fool myself in believing I am the only one that suggested those features). A tour of good things, and points for improvement.

The first impression is not shocking; it looks like any other interface, with molecules drawn as images in the paper:

All structures that are numbered and linked (as in C6-epi-stephacidin A (Compound 13) have a hover-over function to popup a drawing of the structure:

The popup image is a nice gimmick, but not really sematically useful. The link, however, is! It points to a separate supplementary page with further information which include a image of the 2D structure and, following a link, the 3D structure in Jmol. Moreover, it comes with the machine readable representations:

This is indeed interesting, and a big step forward, though please do note my comments later. For convenience, all molecules with such supplementary information is available from the special Chemical Compounds section of the paper:

Excellent! This really is a step forward towards a data-rich paper! Indeed, I will shortly write up a Bioclipse plugin for Nature Chemistry, which will download all molecular structures based on the DOI! Anyway, more on that later… For this article, that table looks like:

By now, you likely also noted the links to PubChem, and indeed, upon publication of a paper, all structures are deposited in the public domain:

At last but not least, each molecule is available in the Chemical Markup Language (with 2D coordinates)! And you know I am a very happy CML user for a long time (see e.g. Peter’s recent blog Egon Willighagen and CML ). BTW, one comment on the CML: the namespace used is the outdated namespace, not the current one (see There can be only one (namespace)). (But the CDK and Bioclipse will read it anyway.)

Details matter

So, while the first impression was not shocking, it was a bit deceptive. Nature Chemistry really changes publishing of chemistry. But I have bad news too. They need to improve the HTML they produce.

But before pointing out some missed chances, let me reply inter alia to Peter’s recent work on the Open Source plugin for including semantic chemistry in MS-Word documents (see [How can we publish semantic chemical documents? </i](https://blogs.ch.cam.ac.uk/pmr/2009/03/16/how-can-we-publish-semantic-chemical-documents/)): Nature Chemistry seems to have done a great job with existing tools. Nevertheless, I fully back up Peters comment that while the plugin is useless without Word, the results produced with the plugin are extremely Open Standard, and enormously reusable! Indeed, while the Word file format is only formally an true Open Standard, the file format is plain XML, and extracting content bearing the CML namespace is trivial.

Which reminds me, if someone from the Nature Chemistry team is reading this, please point me to a blog what tools actually are involved in publishing a Nature Chemistry paper! I think we all like to know.

Now, the HTML has room for improvement. First of all, a look at the metadata defined for the web page of the article shows a description and keywords about the journal, not the article, and the same goes for the web pages for the molecules:

Additionally, the compound details web page has no special markup for the machine readable information:

Or, if it does, it’s still mixed with markup for visual pleasing output:

Still, the HTML is clean enough to have some regular expressions extract a good deal of information, and there is also still the PubChem deposition.

Beyond connection tables

Like many other chemistry journals, Nature Chemistry does not consider properties of the molecule interesting, and NMR spectra are hidden in the Supplementary Information. This paper in particular, disregards a lot of machine readable facts by putting all experimental section bits in a PDF document. So, the next challenge for Nature Chemistry will be to get the authors of papers contribute the original spectra (JCAMP-DX, CMLSpect, etc) in the supplementary information section. Better, have the raw data or even the NMR peak-atom annotations deposited in public repositories such (see Open NMR data: raw curves and annotated peak lists ).

All in all, I am rather positive about the first Nature Chemistry issue, and like to thank the editors and paper authors for there efforts on improving publishing chemistry!

• a scripting console
• a file browser (the Eclipse way)
• an outline of the file content which allows selections
• a script editor

The underlying RCP toolkit has many other interesting features for a Jmol application, but the above is up and running:

However, when I checked the HTML of the first of the two papers (A computational investigation of the structure of polythiocyanogen, doi:10.1039/b810147g). The main HTML still links to a supplementary page. Progress, but not perfect either:

Anyway, the key thing which ONS and CDK and Jmol share, is that they use an Open Notebook. Not every Open Source or Open Data project does. Actually, many scientific Open Source are not open Projects! They are more like the Cathedral than the wished-for Bazaar (see The Cathedral and the Bazaar). So, Open Source (science) projects are certainly not ONS projects by default!

Now, the CDK actually is ONS, it is a Bazaar. The notebooks we use include:

• open project via mailing lists
• open methods/results via subversion
• informal reporting via blogs (e.g. Rajarshi, Christoph, Thomas, mine)
• informal reporting via CDK News

What more would you wish for? That’s not a rhetorical question. Remember that every reader of this blog is in my advisory board !

Unfortunately, I do not create work at a workbench myself, so I do not produce new knowledge myself, other than extracted from existing data. That’s really a shame, and I really do hope that Jean-Claude or Cameron will send me a box to measure solubilities (see here, here, and here, here for first data exploration), even though I cannot participate in the challenge. (hint, hint :)

From Cathedral to Bazaar in Life Sciences

One Cathedral we ran into with Bioclipse was BioCatalogue, which will serve as website where people can annotate and categorize (web) services. While the project has been around for a while, the website was rather uninformative. Fortunately, the projects is going to open up, and be more Bazaar-like. For example, they now started a wiki and a mailing list. I hope these efforts will continue, so that I can contribute from my point of view!

The EMBRACE Registry is a project with similar goals and a rather nice outcome (which I learned about on Monday ). It is actually anticipate to be replaced by or merge with BioCatalogue. So, all data I entered, cheminformatics workflows (look, no ‘o’ ), will later be available from BioCatalogue too. That is already my first contribution to BioCatalogue. One enormously interesting feature of the Registry, is that is allows uploading of code to test the service. This will mean the Registry will not only poll if the service is still online (by checking the WSDL file), it will also test if the service behaves properly. Now, immediate thoughts are mashups with MyExperiment. Each WSDL entry in the Registry points to MyExperiment workflows that use them, and the workflow page would indicate the status of all used WDSL services. This integration was already anticipated long before I thought about it, as the involved Cathedrals were nicely located in the same floor in Manchester.

Below is a screenshot from the EMBRACE Registry for the ChemSpider WDSL entry for a workspace I uploaded about a year ago to MyExperiment:

BTW, ChemSpider has an Advisory Board of which I am member, but it is also a classical (and intentional) Cathedral project. We do share common interests though, which makes us collaborate.

Why Important?

One recurrent theme in Open Source is given enough eyeballs, all bugs are shallow. This surely applies to science as well. The difference between the two is that in current science the eyes only inspect with a delay of at least 6 months. Current practice is that research is finished (delay), and when decided publishable written up a paper (delay, and loosing valuable information in the process, as you can read in my blog all the time), and published (even more delay). ONS changes that, and so do Bazaar-like open source projects, such as the CDK, Jmol and Bioclipse. They bugs are present, whether we like it or not, not just in source code, but in science too. Theories get overthrown, but why should we like the long delays current scientific good practice? Hate it! Work around it. Use the Bazaar. Use ONS!

Now, ONS actually needs Open Source, allowing them to deal effectively with the data they produce; to allow extraction of new scientific knowledge from the measurements. If Rajarshi and Pierre would not have made their efforts, other could not easily join in, leading to those much hated delays. Bugs should be shallow, and openness allows us to make those bugs visible. We can prove that there is a bug, without having to reproduce data ourselves, leading to those nasty delays again. Just copy the data, compare it to your own, do your analysis.

One recent project in open source chemistry dealing with making bugs visible, is the web page set up by Andreas Tille for the DebiChem project. His page summarizes the bugs listed for the chemistry in Debian (which includes the Blue Obelisk projects Avogadro, BODR, CDK, Chemical MIME Data, Kalzium and OpenBabel):

This data analysis helps the projects being analyzed.

Packaging

This brings me to a last topic, for this blog: packaging using Open Standards. In order to allow those eyeballs to spot bugs, it is of the utmost importance to package your results in Open Standards, and not just one, but likely many. For Open Source projects this ultimately means Distribution Packages (deb or rpm). If that goal has been achieved, you know your results can be read by anyone. Software should be installable (make, ant, cmake, etc), and Data should be readable (no PDF, but RDF, XML, JSON, or whatever standard). Preferably not Excel, as this is too free format (as Rajarshi also indicated), but with some added conventions it may do well. Blue Obelisk project are generally doing well in terms of packaging.

For the CDK, which already is reasonably well packaged, I am currently working on Eclipse and Maven2 packages. The former is already being used by Bioclipse, while the second aims at Jumbo (which has just seen a new release. Jim, I’m happy to see the CMLDOM/Jumbo split!), CDK-Taverna, and possibly a third (Paula, what for do you plan to use it?). The POM export is not fully working yet, but with four research sites involved in this Open Project, I’m sure we’ll work it out.

The bottom line is, scientific progress would benefit so much from a Bazaar approach. And the key thing is not collaboration; that’s something you can do in a Cathedral-like fashion too. No, the key thing is to be Open and allow anyone, even your worst nightmare, to comment on what you do. Let him prove you wrong, openly, that is.

OK, there it is. My open notebook entry for this week. Now you know what I have been up to this week.