eScience Workshop for Cheminformatics and Drug Discovery

The call for papers for the eScience for Cheminformatics and Drug Discovery Workshop is now open. It will take place in Indianapolis on Dec 11, 2008 as part of the 4th IEEE International Conference on e-Science. The deadline for submission is Aug 10, 2008.

Organizers

David Wild, Indiana University, USA
Rajarshi Guha, Indiana University, USA
Marlon Pierce, Indiana University, USA

Organizing and Program Committee

Dr. Ajay, NIH (P.C. only)
Dr. Jean-Claude Bradley, Drexel University
Dr. Curt Breneman, RPI
Dr. Jeremy Frey, University of Southampton
Dr. Phil Hajduk, Abbott Laboratories
Dr. Michael Lajiness, Eli Lilly
Dr. Peter Murray-Rust, University of Cambridge
Dr. Kerby Shedden, University of Michigan
Dr. Alex Tropsha, UNC School of Pharmacy
Dr. Ian Watson, Eli Lilly

Abstract

Cheminformatics is becoming an increasingly important part of drug discovery, chemistry, biology, toxicology and environmental science. There are now huge volumes of information about chemical compounds and their related biological properties available in the public domain, as well as in proprietary databases, and a large number of computations that can be applied to this information. There is thus much interest in the application of eScience and cyberinfrastructure to such information mining and computation problems. We are requesting papers that will cover a variery of aspects of this emerging application of eScience and grid computing in chemistry, cheminformatics, biology, toxicology, environmental science and computer-aided drug discovery. Papers should address one or more of the following areas:

* Computation on extremely large chemical datasets (>10m chemical structures)
* Distributed data mining of large chemical and biological datasets
* Grand challenges in cheminformatics
* High impact opportunities for cheminformatics and eScience
* Cloud computing infrastructure requirements for cheminformatics
* Chemistry-aware search engines and journal article processing tools
* Natural Language Processing and Semantic Web applications

Cameron Neylon's Interactive Talk Tomorrow

There's going to be tons of liveblogging and FriendFeed action at Cameron's talk on Open Notebook Science tomorrow. For those on the US East coast that is 7:45 AM:

Tomorrow I am giving a talk at the UKOLN Institutional Web Managers workshop starting at 12:45 British Summer Time (GMT+1). In principle you will be able to see the talk video cast at the links on the video streaming page. The page also has a liveblogging tool (OpenID enabled apparently!). I won’t be liveblogging my own talk but I will be attempting to respond to comments or questions either on that tool, via FriendFeed, or @cameronneylon on Twitter. I make no promises that this will work but if it all fails then I will record a live screencast.

CDD meeting at UCSF for Drug Discovery Oct08

Here is the call for poster presentations at the Collaborative Drug Discovery meeting on October 7, 2008 in San Francisco. I look forward to seeing Barry Bunin there and hopefully some of our collaborators.

Hosted by Gladstone Institute and QB3 at the UCSF, Mission Bay Campus

2nd Annual CDD Community Meeting (a free, 1-day event) on October 7th is the best place for the CDD community and all scientists to get together, discuss their research projects, and showcase their work to others in the community.

Early List of Confirmed Speakers:

* David Sullivan, MD, Johns Hopkins
* Fred Buckner, MD, University of Washington
* Janice Kranz, PhD, Stanley Center for Psychiatric Research, Broad Institute of MIT & Harvard
* Conor Caffrey, PhD, Sandler Center, UCSF
* Jean Claude-Bradley, PhD Dept. of Chemistry, Drexel University
* Sean Ekins, PhD, ex-Pfizer/Lilly Computational Chemist
* Andrea Altieri, PhD, ASINEX
* Robert Damoiseaux, PhD, UCLA Molecular Screening Shared Resource (MSSR) Center
* Moses Hohman, PhD, Head of Software, Collaborative Drug Discovery (CDD roadmap)
* J. Leighton Read, MD (Alloy Ventures) - Financing Biomedical Innovation from academia to the market: From FluMist™ to Priority Review Vouchers and other stories
* Russell "Rusty" Bromley, Chief Operating Officer, Myelin Repair Foundation

Timely panel topics:

* Panel #1: What Foundations are looking for from Academic researchers?
* Panel #2: What makes collaborations successful within and between groups?

Who should attend?

* Scientists working in drug discovery
* Researchers who are involved in collaborations or generate complex drug discovery data
* Students who want to learn more about the drug discovery industry
* People who wish to network, establish new or strengthen existing collaborations
* Current CDD community members to set the collaborative platform's future direction
* Anyone interested in new drugs for diseases of the developing world, private public partnerships, and/or global health

Stay for the post-symposium reception from 6 -7pm!

Einstein Rejected Peer Review

According to Silvan Schweber (page 9 of Einstein & Oppenheimer):

By 1930, every European scientific journal would automatically accept and publish any paper that Einstein had submitted.

When the Physical Review dared to submit his paper for peer review, Einstein responded:

We (Mr. Rosen and I) had sent you our manuscript for publication and had not authorized you to show it to specialists before it is printed. I see no reason to address the - in any case erroneous - comments of your anonymous expert. On the basis of this incident I prefer to publish the paper elsewhere.

There are many ways to look at this, depending on one's agenda.

On the positive side, it looks like Einstein was able to contribute to science, despite using the publication system much like we would now use Nature Precedings or a blog. But did the readers know his papers were not peer reviewed? At least with our current Science2.0 tools the assumptions are more explicit. And it is much easier for the community to comment.

Cameron's July 22 talk on Open Notebook Science

Cameron Neylon will be giving a talk "Science in the YouTube Age" on July 22, 2008.

The talk will be webcast and he will be taking questions by FriendFeed.

He recorded a brief screencast as a preview. It is a very nice overview of the current opportunities and challenges to doing science in a much more open way.

JoVE shoot at Drexel

On Saturday July 12, 2008, in my lab at Drexel, we recorded our JoVE shoot for the use of Mettler-Toledo's MiniBlock and automated liquid handler for optimizing a Ugi reaction.

It took about 6 hours and I was very impressed with the professionalism of Ian, the camera man JoVE sent us. The preparation time was much more than that and I appreciate JoVE's flexibility in compressing their regular operating timeline because of our limited time with the equipment.

One of the main challenges was finding an appropriate level of detail and selecting illustrative video sequences to describe the protocol. After working with Tom, Khalid and Aaron from JoVE I think the resulting script should be helpful for people wishing to repeat the experiment. Khalid and I recorded the intro and conclusion and the rest of the script will be recorded by JoVE.

The good news is that the shoot went well I think. Multiple takes were recorded for most sequences so it should be possible to piece together a decent video. However, we discovered flaws in the way we have been programming the MiniMapper, which would explain the strange and inconsistent results we obtained at first.

Thanks to Tom Osborne digging through the log files of our runs to identify the problems, I think we have a liquid handling protocol that is correct now. The first experiment using it (EXP199) is showing reasonable results so far, with the higher concentrations 0.2M and 0.4M and methanol or methanol/ethanol methanol/acetonitrile mixtures showing the higher yields. We'll see if we can get good reproducibility in next few runs. Once that is done we can wrap up the JoVE paper.

Older Males Prefer NMR of Ester

I was just looking at the YouTube Insight feature showing demographic and access info on my uploaded videos. Since I use my YouTube account mainly to provide solutions to organic chemistry problems in my undergrad classes it was surprising to see that the most active group of viewers were 45-55 year old men.


And the most popular video is the NMR of an ester, where I explain the effect of a chiral center on the splitting pattern of methylene groups.

How should Open Notebook Science be used?

Maxine Clarke highlights a bit of recent controversy regarding Open Notebook Science that has been bouncing around the blogosphere and FriendFeedosphere.

There are some who interpret the ongoing publication of our laboratory notebook as an expectation for the world to read it like a magazine. For someone who is not a collaborator or working in a related area that would make about as much sense as reading the phone book.

Here is an example of how an Open Notebook should be used:

Based on our Sitemeter records, this morning someone from the UK seached for "purification of phenylacetaldehyde" on Google. The first hit was the UsefulChem experiment EXP037 performed by undergrad student James Giammarco on October 6, 2006.

This happens to be a "failed experiment". But here is the information that can be obtained from the wiki:

1) The picture of the experimental set up:



2) The results provide an NMR of the first distillate, showing exactly the nature and amount of impurities. This is before we started using JSpecView systematically so this spectrum is not interactive, it is just an image.



3) The discussion addresses a major problem with the first distillate and a suggestion is made about what might be the problem:

Although the HMR of 37-F1 shows reasonable purity for use as a reagent, it appears contaminated with droplets floating and sticking to the sides of the flask. This may be water condensed by the distillation in open air. Future distillation should be done under nitrogen.
Phenylacetaldehyde is reported to boil at 195C. Since 37-F1 was collected at 162C (approx), perhaps this corresponds to an azeotrope with water and would explain the contamination in the first fraction. In the future more careful measurement of the Ubergang temperature over the course of the distillation would help to clarify this issue.

4) The history of this wiki page reveals who made each contribution. A major advantage of using a hosted wiki as a lab notebook is that a reference to every version of the notebook can be made, marked with a third party time stamp:

On October 27, 2006 James recorded his log, the NMR results and a conclusion.
On October 29, 2006 I added some questions in bold, re-interpreted the results, added a detailed discussion and changed the conclusion.
On January 3, 2007 James corrected a spelling mistake I made.
On January 9, 2007 my graduate student Khalid added a link for phenylacetaldehyde.
On January 23, 2007 Khalid added InChI tags.
On November 22, 2007 James answered my questions in the log.

5) This is the kind of information that someone experiencing problems with purifying phenylacetaldehyde could use. In the experimental section of a chemistry article the most you are likely to get is "phenylacetaldehyde was purified by distillation", with no proof as to the actual purity.

Notice that the experiment is never really "completed". There is always more that someone could add in terms of the discussion or add a question anywhere in the document. That is how science actually happens for the most part and we can forget that when listening to a slick presentation at a conference or read a well-written article making it look like the project unfolded effortlessly from a perfect plan.

Coming back to the person who searched for and found this page. This lab notebook page may or may not help solve their problem directly. The most valuable information is probably the contact information for the people who contributed to the execution and analysis of the experiment.

We aren't hiding behind pseudonyms. We use our real names because we want to fully participate with the research community and we stand behind what we report. Shoot us an email for help if the notebook isn't enough.

InkSpot and Open Drug Discovery

I had a nice long chat with David Leahy from InkSpot yesterday. There is actually a good discussion in the comments of David's most recent post.

What I gathered from our conversation is that InkSpot will be providing transparency on the computational side of the drug discovery process. This is something that could be very valuable to the Open Science community in general and to our group in particular as we look for new anti-malarial agents.

We've tried to record the docking results we've obtained from Rajarshi Guha using a wet-lab style report (for example D-EXP014). Hopefully there is enough information there for the docking experiment to be replicated but clearly it would be better if a workflow system were in place to automatically record the steps and make those records public and easily searchable.

David said he would start with a QSAR-style exploration of our Ugi product precipitate prediction problem. It will be interesting to see how this plays out.