Thursday, August 31, 2006

Depth First Blog

Rich Apodaca runs Depth-First, a cheminformatics blog where he publishes code that may be quite handy for UsefulChem. For example, here is a post on creating an API to PubChem.

Even though he is using Ruby on Linux, we can probably adapt it to our VB work on Excel.

Wednesday, August 30, 2006

CML RSS Reader in Java

In addition to a Java version of MoleculeBlogInfo, I've also developed a first cut of an RSS reader devoted to RSS (1.0) feeds containing Chemical Markup Language (CML). A zip file containing this software can be found here. Although not (yet) very sophisticated, particularly in its user interface, this software has several useful features, which I expect to build upon. Like all feed readers, it will run in the background, periodically checking for and downloading new feed items as they appear (in fact, the user interface can be turned off, allowing the program to be run on a server machine). All CML contained in items is parsed out into separate XML files, which can be viewed with a molecular viewer such as Jmol, or processed in other ways. Other chemical information in the feed, such as SMILES, InChIs, molecular weights, and suppliers, are also extracted and displayed in the user interface as they are found. The most interesting feature of this software, however, and closely in keeping with the vision of chemical information in RSS, is the ability to configure program "hooks" to perform any additional processing desired. For example, a "post-download" program could be configured to perform any desired calculations on a molecule, or perform a web search for information; the results could then be automatically added to a database, uploaded to a web site, or e-mailed to all interested parties. An example of a configuration file for the software is given below:

<?xml version="1.0" encoding="utf-8" ?>
<Configuration>
<Feeds>
<Feed>
<DownloadURL>http://showme.physics.drexel.edu/usefulchem/Software/MoleculeBlogInfo/usefulchem-molecules/usefulchem-molecules.rss</DownloadURL>
<DownloadURLLastModified>Fri, 25 Aug 2006 23:47:26 GMT</DownloadURLLastModified>
<DataDirectory>usefulchem-molecules</DataDirectory>
<DownloadFileName>usefulchem-molecules.rss</DownloadFileName>
<Hooks>
<PostDownload>java UsefulChemMoleculesPage D:\Drexel\Cheminformatics\Ajax\UsefulChemistryMolecules</PostDownload>
</Hooks>
<RefreshInterval>
<Days>0</Days>
<Hours>1</Hours>
</RefreshInterval>
</Feed>
</Feeds>
<FeedsFrame>
<Display>true</Display>
<Width>800</Width>
<Height>600</Height>
</FeedsFrame>
<MoleculeViewer>java -Xmx512m -jar Jmol.jar</MoleculeViewer>
<OutputFile>CMLRSSReader.out</OutputFile>
</Configuration>
Note the <PostDownload> entry in the <Hooks> section; the program UsefulChemMoleculesPage generates a dynamic HTML page which allows you to select the <description> fields from the RSS item entries (this page, incidentally, is a demonstration of the use of Ajax and Javascript to dynamically load data from a server without refreshing the entire page).


Working on the Ugi

Going over some of the other papers that Chris Hulme wrote regarding the Ugi reaction and diketopiperazines, there are a few more answers but a few more questions

1) We now know that he used 1,2-dichloroethane in the 10%TFA solution for cyclization from a review paper
2) In the same review paper, there are some small changes to the procedure that make it safer and less complicated than we originally thought. While it does not specify that it uses MeOH (looks like they used ethyl glyoxalate) in the Ugi reaction it says that it was under reflux over night not just stirring at room temperature.
3) the cyclization step can be done with 10% of AcCl in MeOH (AcCl reacts immediately with methanol to generate anhydrous HCl when prepared separately, as done by Tenderbutton, it would then be added to the mixture) or TFA in 1,2-dichloroethane and it was done at room temp over night
4) the review paper mentioned above looks like it gives HMR info of two diketopiperazines which can be used for comparing NMRs when we are fairly certain we have obtained our product.

Some of the questions that we have brought up now that we have compared some NMRs of Boc-protected amino acids and from reading the paper mentioned above are as follows

1) can we have tautomerization in the cyclized diketopiperazine ring since we have amides? We have observed in our HMR of our Boc protected amino acids (Boc-Gly-OH and N-Methionine) there is some peak splitting which we can interpret to be tautomerization.
2) there are complications using dichloroethane in that it can react with our amine and other amine byproducts so why did Hulme used this particular solvent?
3) since most of Hulme's experiments were automated and purified by HPLC, is the product diketopiperazine difficult to purify by chromatography?

MoleculBlogInfo RSS Feed Now Generated by Java

A few tardy posts on recent work.  The RSS feed for usefulchem-
molecules
is now being generated automatically by a java version of
MoleculeBlogInfo.xls.  The java software currently runs as a scheduled service on my machine, downloading and processing the usefulchem-molecules blog every night, and uploading a new feed only when necessary.  A new feed is generated only if the blog has changed.  The software maintains a log file so that it can be monitored.  A zip file containing this software can be found at here.


There are two advantages to working with Java over Excel VBA.  One is portability to other platforms -- Excel of course only runs on Windows.   Another is the object-oriented features of Java, which make large software projects easier to design and maintain.  (A third advantage is that Java is free, of course.)  Finally, as there are freely available Java libraries which can generate Excel files, there is really no need to use VBA (okay, four advantages!).  No, not finally, as Java's extensive support for internet protocols and classes (sockets, HTTP, FTP, etc.) is also extensive.

Tuesday, August 29, 2006

Following UsefulChem Experiments

If you have been wondering why there has been a drop in the activity on the UsefulChem Experiments Blog, remember that we have moved our group lab book to the wiki.

The best way to find out who is doing what on an hourly basis is to click on the Recent Changes link on the left of the wiki. Clicking on the pages that come up will then take you to a particular experiment in progress (for example). From there the History button will show the most recent changes to that experiment. You can compare what was added (in green) and what was deleted (in red) by clicking on the pics in the list (for example). That way it is possible to find out not only what happened but also HOW the student, supervisor and colleagues arrived at the results, presented arguments in the discussion and came to their conclusions.

This is just not possible to do with a blog. However, we will be migrating over completed (or as close as completed as we think they will get) experiments to the experiments blog for a convenient way to be alerted to just the completed experiments via a clean and simple RSS feed (click on the subscribe with Bloglines button at the top of the blog page). The wiki does have an RSS feed but it gets updated every time there is a minor change, which makes it harder to use practically for anyone outside of our immediate group.

I have also created a contents page to quickly brief anyone on all of our projects. There are links from there to each experiment in progress or completed. We appreciate the feedback we have been getting so far and we welcome future comments on any aspect of our work from chemists anywhere.

Sunday, August 27, 2006

Synaptic Leap Presentation

Ginger Taylor gave a presentation about The Synaptic Leap at the August 2006 Science Foo Camp. She has a lot of text with her slides that explains the history of TSL and her vision for the future. Issues brought up include patents, funding, virtual pharma, collaboration and publication. Anyone with a stake in Open Source Science should take a good look.

As she mentions, UsefulChem can connect with TSL at a our milestones and broad questions, with links back to the raw experimental data in our blogs and wiki.

Tuesday, August 22, 2006

Chemistry Central

A new open access site for chemists - Chemistry Central - launches today as part of the newly announced Open Access Central group of sites from the makers of BioMedCentral.

CC collates peer-reviewed research from a range of open-access journals and makes available the original research articles as soon as they are published.

They're also planning to launch a new OA chemistry journal, boss Bryan Vickery tells me, you can get the low down on Chemistry Central in my science blog.

Monday, August 21, 2006

Synthetic Pages

In a letter to Nature in response to a publication concerning publishing scientific results online, Stephen Caddick mentions Synthetic Pages. This looks like it could be a really good vehicle for publishing our organic chemistry reactions in UsefulChem after we have all the kinks worked out. Each reaction is a separate publication and is open to public comment.

Thanks to nsaunders for the link.

CML Explained Blog

Egon Willighagen has just created a new blog to explain how to use Chemical Markup Language. Great idea! If you are working on the cheminformatics aspect of UsefulChem make sure to subscribe and contribute.

Tuesday, August 08, 2006

RRResearch

I just came across the RRResearch blog from a post by Coturnix. This is the kind of science blog that I like to see - lots of discussion about experimental details that can immediately empower other researchers in a specific field.

Creative Commons Attribution Share-Alike 2.5 License