The Rosania Lab Open Notebook Science Wiki

I recently reported on a new collaborator who agreed to work with us in the open on modelling subcellular drug transport.

I am very pleased to report that Gus Rosania has now created an entire wiki (1CellPK) for his lab to use as an open notebook. From the home page of the wiki:

Open Notebook Science is ideally suited for community-wide collaborative research projects involving mathematical modeling and computer simulation work, as it allows researchers to document model development in a step-by-step fashion, then link model prediction to experiments that test the model, and in turn, use feedback from experiments to evolve the model. By making our laboratory notebooks public, the evolutionary process of a model can be followed in its totality by the interested reader. Researchers from laboratories around the world can now follow the progress of our research day-to-day, borrow models at various stages of development, comment or advice on model developments, discuss experiments, ask questions, provide feedback, or otherwise contribute to the progress of science in any manner possible.

How's that for a Christmas present to the Open Science community?

X-Ray Crystallography Collaborator

We have another collaborator who is comfortable with working openly: Matthias Zeller from Youngstown State University.

With the fastest turnaround for any crystal structure analysis I've ever submitted, we now have the structure for the Ugi product UC-150D. For a nice picture of the crystals see here.

Although NMR supports our structure assignments, there is not enough proximity between the hydrogens to assign them all by walking through the couplings. Yes, high resolution mass spectrometry is a nice confirmation. But nothing beats the confidence a chemists gets from an X-ray crystal structure.

Here is a pic of UC-150D - for all the raw data files take a look at the lab notebook page EXP150. Notice the pi-stacking of the aromatic rings.

Subcellular Drug Transport UsefulChem Collaborator

Rajarshi Guha has yet again made a key contribution to our UsefulChem project by connecting us with Gus Rosania at the University of Michigan. Gus is interested in a fully open collaboration to help us further prioritize our drug targets based on predicted subcellular drug transport:

It is the first time I hear about Open Notebook Science, but it sounds like a fantastic idea!

My research group studies the subcellular transport of small molecules. We are interested in combining cellular distribution of small molecules together with systems biology, to analyze the pharmacological activity of small molecules in a cellular (and organismic) context. For more information about our subcellular transport lab, you can visit us at http://www-personal.umich.edu/~grosania/

Indeed, one of our objectives is to make all our cellular pharmacokinetic models open source, so that they can be modified, evolved and used for educational purposes and as virtual drug discovery tools, throughout the world. We are porting all our models to Virtual Cell (http://vcell.org/), where they are in a form that can be freely accessed and distributed on-line, though a simple graphical user interface Plus, Virtual Cell sponsors courses in computational modeling and systems biology as well as conferences, so it is more than just a modeling tool. As we work on this, we would like to explain what we are doing step-by-step, for everyone else to follow.

Gus provides more details about what his group can do:

We can readily calculate on-target lysosomal drug concentrations, vs. off-target mitochondrial or cytosolic drug concentrations. Question: are the inhibitors small "drug-like" molecules, peptides(or some other awful thing)? If they are small drug-like molecules, then we are good.

With 1cellPK we can identify the inhibitors that would lead to greatest accumulation in lysosomes of a cell surrounded by a homogeneous extracellular drug concentration (the parasite) while minimizing the accumulation in lysosomes of an off-target cell (ie the intestinal epithelial cell mediating absorption in the pesence of a transcellular concentration gradient). With 1CellPk we should also able to select those molecules with the highest transcellular permeability (for oral administration) while at the same time accumulating minimally in the cytosol of cells in the presence of a transcellular concentration gradient (to minimize metabolism andtoxicity in intestinal epithelial cells and hepatocytes, while maximizing intestinal absorption and systemic bioavailability).

There are several ways we can actually execute the collaboration. At the most basic level, if you give me the compounds' chemical structures, I can run them through 1cellPK (in Virtual Cell), and estimate their permeability, intracellular accumulations and distribution in the parasite, as well as in non-target intestinal epithelial cells. Once we give you the 1CellPK calculations, we can sit down together and figure out the best way to combine docking predictions with 1cellpK predictions.

Our falcipain-2 project is an obvious place to start.

Camphor in Second Life

This term, the students in my organic chemistry class were presented with an opportunity to do an extra credit assignment using Second Life to represent concepts they learned in the course.

When I was an undergraduate, finding molecules in articles was mainly done using the Chemical Abstracts books. A convenient way to find a specific molecule would be to look up the molecular formula and find the corresponding IUPAC name. Theoretically, one could figure out the IUPAC name from scratch but this can be very tricky for complex molecules and prone to error. With the correct name, I could look for analogues of a molecule of interest in alphabetical catalogues by understanding how the chemical name works.

But when computer databases started to be used in chemistry, using the name of a compound became far less important. Searching for molecules now comes down to drawing them on computer screens and using computer generated text representations like SMILES and InChI. Knowing how to use these tools on free software and services is key to being fluent and flexible on the chemical web. And I think that is the most important benefit that students get from doing these assignments.

As an example, take a look at the project created by my student Charles Sineri (Chaz Balbozar in SL). In the image below he is standing between two molecules of camphor that are mirror images of each other, demonstrating the concept of chirality that we covered in class. This is a particularly difficult example to demonstrate on paper.

Using molecular models that are bigger than my body is not something that I have ever done in real life and it provides an interesting perspective to what the molecule really looks like. Another advantage is that you can fly the molecule like an airplane by wearing it. Here is Chaz flying up to a buckyball on his camphor ship:


In order to get his project done Chaz had to learn about and use SMILES, InChIs, ChemSpider and ChemSketch. These are free tools that he will use again in future chemistry applications.

The main challenge in getting this implemented in Second Life is providing tools that are easy to use. We used Andy Lang's (Hiro Sheridan in SL) molecule rezzer to do this because it now has the capability of understanding InChIs and SMILES. Hiro was kind enough to make some further modifications to make it even easier to use. It was gratifying to see that it understood chiral SMILES code.

Visit Chaz's project on Second Nature island - see SLURL.

Chemistry Crowdsourcing with Open Notebook Science

I recently submitted a Letter of Intent for the NSF Cyber-Enabled Discovery and Innovation competition. Kevin Owens is a co-PI and will assist with the laboratory automation component. ChemSpider will contribute the database support. The pre-proposal is due in early January 2008 and we'll be writing it openly here. Comments are welcome.

We would ultimately like to enable the chemistry community to directly control the actions of a robot to help us understand some chemistry problems. As we make our way towards this goal, it would be very useful to start with suggestions for protocols to be executed by students we currently have in the group.

We already have a mechanism in UsefulChem to post experimental plans. In order to make the transition to full automation easier, it would be preferable if suggested protocols are even more specific than what we currently have listed. For example, instead of describing a general procedure like EXPLAN005, actually specify all of the compounds, amounts, mixing times, etc. This way the protocol can just be copied and pasted in the Procedure section of a new experiment, executed faithfully and reported in the main experiment list.

The main puzzle to solve is the prediction of which Ugi products will precipitate. A hypothesis might be that a precipitate will always occur from methanol at a certain minimal concentration of a certain reagent. Another approach might be based on the predicted molecular descriptors of the Ugi products. We might also start with as few assumptions as possible and use a genetic algorithm to evolve a solution. We'll be doing some of these but clearly there are more ways to solve this puzzle than we have resources or expertise.

So if anyone is interested in participating at this stage contact me to get access to the wiki and further discuss. Other examples of chemistry crowdsourcing : Chemmunity, The Synaptic Leap, OrgList, Chemists Without Borders and ChemUnPub.

Here is the LOI:

Chemistry Crowdsourcing using Open Notebook Science

The current system of dissemination of scientific data and knowledge is far less efficient than it needs to be to facilitate improved collaborative science, especially considering current publication vehicles and infrastructure. There is a growing movement promoting more Open Science, with the belief that a more transparent scientific process can perform far more effectively. The logical extension of this concept is full transparency - exposing a researcher's complete record of progress to the public in near real time. Not only will such a process enable ongoing data sharing it also provides an opportunity to develop collaborative communities of scientists and, at the conclusion of data acquisition, can enable communal extraction of conclusions when necessary. We have named this approach Open Notebook Science and have demonstrated its implementation and feasibility with the UsefulChem project, started in the summer of 2005, with the aim of synthesizing novel anti-malarial compounds. Our system currently uses free hosted services using general blog and wiki functions to facilitate replication across any scientific domains. These services are not chemically intelligent and are limited to text and graphic based data sharing only. For Open Notebook Chemistry the ability to intelligently manipulate, manage and search chemical structures and associated data is necessary and we have demonstrated proof of concept capabilities by integrating with the ChemSpider service, a free access online database managing chemical structures and focused on developing a structure centric community for chemists. This work will require the development of a chemically intelligent software platform to extend the capabilities of both the blog and the wiki environment for managing Open Notebook Science. The exposure of raw experimental procedures and data in a semantically rich format will enable the participation of both human and autonomous agents in the process of scientific discovery. This phenomenon of spontaneous group intelligence, referred to as "Crowdsourcing", has proven valuable in several contexts. Already, productive collaborations have been forged within the UsefulChem project with groups from Indiana University, Nanyang Technological University, the National Cancer Institute and UC San Francisco.

First Falcipain-2 Targets Shipped

We've reached an important milestone on our CombiUgi project involving the synthesis of falcipain-2 inhibitors. In my last update I described how our focus was more on doing many reactions in parallel and only looking for Ugi products that precipitate in pure form within a few days.

It took little longer than I hoped. In order to do more reactions, we reduced our efforts towards monitoring. One of the assumptions that we made was to trust a bottle's label to accurately describe its contents. That turned out to be incorrect for one of our key aldehydes, as we eventually found out by systematically taking NMRs of the starting materials. Soon after ordering a new bottle of phenanthrene-9-carboxaldehyde we were treated to the growth of beautiful crystals (see EXP150 by Khalid and Emily):

This compound was ranked 155th out of 71,000 Ugi products for docking with falcipain-2 at potential receptor site V1 (see full description by Rajarshi Guha here).

This compound, along with another (EXP148) that crystallized similarly, have been shipped to the Rosenthal group at UCSF for testing against the malarial parasite and hopefully get some falcipain-2 inhibition assay results. That way we'll be able to investigate the validity of our docking model.

I'll be posting updates on this blog but the status of any shipped compounds will be maintained on the isolated compounds table.

If anyone would like to run their own assays please contact us. We would be happy to ship any of these compounds, as long as our collaborators are willing to work under Open Notebook conditions.

The beauty of screening for products that are purified by crystallization is that, if any of these prove to be useful for any application, it should be very simple and cheap to produce several kilograms. This can come in handy for end users with very limited resources.

Chemical Blogspace Tags

InChI=1/C15H10O/c16-10-12-9-11-5-1-2-6-13(11)15-8-4-3-7-14(12)15/h1-10H
Phenanthrene-9-carboxaldehyde

InChI=1/C5H9N/c1-5(2,3)6-4/h1-3H3
tert-butylisocyanide

InChI=1/C4H6O2/c1-2-3-4(5)6/h2-3H,1H3,(H,5,6)/f/h5H
Crotonic Acid

Surfing Web2O Article on the Chemistry Blogosphere

Richard Van Noorden of Chemistry World recently published a nice little review of Web 2.0 applications in chemistry. He introduced our UsefulChem project in a creative way:

As chemical reactions go, it was a complete failure. 'Contents of the reaction flask decomposed. Aborted', Drexel University chemist Jean-Claude Bradley and students recorded the day after an attempt to synthesise a catechol aldehyde from adrenaline on 24 January 2006. But the experiment has acquired a peculiar honour: Bradley chose it to be the first written in his group's new online laboratory notebook, in which all experimental data is made public and freely available to web users - a concept he later christened 'Open Notebook Science'.

Bradley's idea is simple: most failed experiments are discarded, yet their data could be useful to someone else. Even published papers don't always sufficiently explain the workings behind a successful experiment. In contrast, all Bradley's research and raw data is now documented transparently and almost in real-time. Anyone can see it, comment on it, and use it; and the internet is the perfect vehicle for hosting it.

There are many other examples in the article - a must read for the new generation of chemists.

Nature's Role in e-Science SFLO Session

Berci Mesko will be moderating a session on Nature's Role in e-Science on SciFoo Lives On (in Second Life) tomorrow Monday December 10, 2007 at 12:00 ET/17:00 GMT.

There will be 4 talks:

Matt Brown: Nature Network
Ian Mulvany: Connotea
Hillary Spencer: Nature Precedings
Helen King: Dissect Medicine

JSpecView Article on Chemistry Central

Robert Lancashire has just published an article in Chemistry Central Journal:

The JSpecView Project: an Open Source Java viewer and converter for JCAMP-DX, and XML spectral data files

Our lab has found this software to be key for communicating organic chemistry results within an Open Notebook Science environment. All NMR raw data and metadata are automatically recorded and users from anywhere can mine the spectra by expanding and integrating at will from a browser interface. This is an enormous improvement over the traditional method of storing and publishing spectra as images that cannot be expanded.

The article describes other useful applications, such as the integration of JSpecView with Jmol, to show the assignment of specific peaks.

The other reason I really like this article is that Robert has used some UsefulChem blog posts as primary references. This is an important way for the scientific blogosphere to get incorporated and accepted by the mainstream.

Abstract

The JSpecView Open Source project began with the intention of providing both a teaching and research tool for the display of JCAMP-DX spectra. The development of the Java source code commenced under license in 2001 and was released as Open Source in March 2006. The scope was then broadened to take advantage of the XML initiative in Chemistry and routines to read and write AnIML and CMLspect documents were added. JSpecView has the ability to display the full range of JCAMP-DX formats and protocols and to display multiple spectra simultaneously. As an aid for the interpretation of spectra it was found useful to offer routines such that if any part of the spectral display is clicked, that region can be highlighted and the (x,y) coordinates returned. This is conveniently handled using calls from JavaScript and the feedback results can be used to initiate links to other applets like Jmol, to generate a peak table, or even to load audio clips providing helpful hints. Whilst the current user base is still small, there are a number of sites that already feature the applet. A tutorial video showing how to examine NMR spectra using JSpecView has appeared on YouTube and was formatted for replay on iPods and it has been incorporated into a chemistry search engine.

Run for Malaria in Philly

For anyone in the Philadelphia area who cares about malaria:

Drexel University Crossings Stair Run

Beta Beta Beta and the Office of Residential Living will sponsor a stair run Saturday, December 8, 2007, from 9 a.m. to noon in University Crossings (101 N. 32nd Street).

Registration is $3 per person, $5 if two people sign up together. Sign-up in the lobby of University Crossings.

All the proceeds generated from the event will purchase mosquito netting to be placed over beds for an African village. Each net costs $10, and can potentially save three people, as children in the villages typically share beds.

More information about this initiative is available at http://malarianomore.org/.

More Proteins in Second Life

For those of you following chemistry in Second Life, take a look at Peter Miller's Tidal Blog.

Peter has been posting a lot lately about rendering proteins in Second Life. For example, here is malarial enoyl reductase, an enzyme that has been a target of the UsefulChem group for some time.

Although we did demonstrate the docking of a molecule in a pocket of enoyl reductase, it is nice to see the whole enzyme.