ICMS Edinburgh Workshop on multiscale methods for stochastic dynamical systems in biology

IMG_8566.JPG

The International Centre for Mathematical Sciences (ICMS) held a fantastic workshop over the last week here in Edinburgh, covering multiscale methods for stochastic dynamical systems in biology.  It's phenomenal that there are organizations that are strongly committed to supporting the exciting interface between mathematics and the biological sciences, and the enthusiastic discussions at this meeting were a reflection of the enormous potential that work at this interface holds for both fields. These workshops also attempt to engage the public to communicate the importance of this interdisciplinary work through a public lecture series, with Sarah A. Harris delivering a talk on the interface of physics and biology.

PDF slides from my talk are available online, and the talks were all recorded to be posted online shortly.

CECAM Workshop on developing interoperable and portable molecular simulation software libraries

I'm excited to be participating in the CECAM Workshop on Developing Interoperable and Portable Molecular Simulation  Software Libraries this week at the Forschungszentrum Jülich, organized by Julien Michel, along with Christopher Woods, Peter Eastman, and Gareth Tribello.

Slides and materials used during the workshop can be found here:

  • Examples and materials: https://github.com/choderalab/cecam-2015-julich-workshop

Science Communication Boot Camp: The Experience

Guest blog post from postdoc Sonya Hanson.

Earlier this month, I went to Science Communication Boot Camp. It was at the 'Alan Alda Center for Communicating Science' at Stony Brook University. We did not get to meet Alan Alda. That was disappointing. But everything else was really, super awesome. We played a lot of improv games, we did a lot of woodshedding explaining our own science, we learned about how to make stories, we learned about metaphors, and at the end we taped three-minute mock media interviews and talks to try and implement what we had learned throughout the week. It was exhausting, it was embarrassing, it was hard, but it was a blast.

Baseball

I first realized we were someplace special when we started talking about baseball. Now, I am not a baseball person, but that's okay: I know enough. But in the beginning of the first day, they asked us to explain the following to someone who knew nothing about baseball:

'In the bottom of the ninth, Jeter worked a one-out walk and stole second. But the Red Sox's ace reliever got Ellsbury and Teixeira to strike out swinging to end the game.'

And it was super hard. We started explaining baseball: there are three bases, there are these things called outs, you get a point by... That just wasn't working. And then... they put up this explanation:

'The game was almost over, and the home team was losing to its most hated rival. The beloved captain of the home team, playing in his last season, made a last-ditch effort to win. He took a big risk, and it looked like it might pay off. But when his teammates tried to help him score, a key player on the other team shut them down. The game ended, and the home team went down to bitter defeat.'

Suddenly you understand the stakes. Suddenly you understand why someone might care about baseball.

Yes, and...

Don't worry, the rest of the camp was not about baseball. Time-wise, the plurality of the boot camp was spent doing improv games. Why, you may ask? What do science and improv have in common? Why would you do improv, where the whole point is that you're just making stuff up, to become a better scientist, where the whole point is precisely that you do not just make stuff up? Because improv is about connecting. Improv is about 'Yes, and..'

Not 'No, but...', not 'Yes, but...', but 'Yes, and...'. Agree and add something. Find how to connect to someone and then add to that. At the beginning the improv games were not science related. Zip-zap-zop, the mirror game, the ball game, the positive side of ranting, etc. My hypothesis is that they're to get us talking. To get us comfortable with talking about anything, anything at all. To find our own rhythm, and try to connect that rhythm to whoever you're communicating with. In one of the most powerful games, we were told to take a blank piece of paper and describe a picture. It was intense. Almost everyone talked about something deeply personal. A picture of an important family space or pet that got you through hard times. There were no instructions to 'do your best to make everyone cry', but somehow that's what happened. And this was all without any preparation. Somehow, we already had these stories inside of us, but how could we use them *dun dun dun* FOR SCIENCE?

STORIES, DISTILLING, AND METAPHORS

The first night, guest speaker Carl Safina told us about (among other things) his 'Spray can theory of science communication': he used to be pissed off that you buy a can of spray paint and it's only 2% paint, but then he realized that the paint is no good without the 98% propellant. The story is the 98% propellant. The science is the 2%. Sorry, guys. That means it pays to find that 2% of your science that is really what you want to communicate. Unsurprisingly a lot of what we learned when we weren't doing improv was how to 'distill our message'. We would go around in small break-out sessions and have one minute to describe our work. One of the first things I learned was, "bring cancer up front". Apparently in my first run-through I left it to the very last sentence. Another was "tell them what you're going to tell them, tell them, then tell them what you told them". I think this is more powerful than it seems. It forces you to decide what you're going to say, and decide your goal instead of just rambling off a laundry list of facts that probably don't mean anything to the person you're talking to, anyway.  Relatedly, getting rid of jargon was surprisingly difficult. In retrospect, it shouldn't have been surprising, but also I now feel bad for all the people I have explained my science to in the past that had to deal with all those meaningless (to them) words. If anything came out of this camp, I hope I am now better at recognizing when this when it happens.

One of the things that helped a lot in explaining more difficult concepts was using any kind of comparison to a real life thing. Why do we care about semiconductors? How small is an atom? What analogies can you make to other complex, but more commonplace things? Coming up with these kinds of comparisons, I think, is often scary for scientists: we don't want to lose reality in an imperfect analogy. But actually, having any kind of reality to compare to is surprisingly helpful when your other option is just an abstract concept... and you only have three minutes to get your point across. We even did a game to explore how easy it could be to find everyday things to relate to scientific topics: everyone writes down a scientific topic on a piece of paper and puts it in a pile; everyone puts a random object from their backpack or purse in the pile; pick four of each. Surprisingly, after mixing and matching, it is not hard to find reasonable pairs: swiss army knife and adaptive evolution, broken retractable badge holder and the RNA folding problem, sunglasses case and protecting DNA in epigenetics, headphones and being desensitized to the song 'Happy' for antibiotic resistance, etc. The hardest part, it turned out, was not pairing a scientific concept to an every day object, but telling a story around it.

The media interview

So that was a lot of fun. We got a bit better at improv and finding our rhythm. We got a bit better at distilling our scientific message. But then they got out the big guns: time to record it. This was definitely invigorating, partly because it was actually at the Stony Brook journalism school where they had real lighting and real cameras, and the Alda Center brought in some totally legit interviewers (they were kind of a big deal: googling Marcy McGinnis or Rory O'Connor is a good way to misplace a few hours of your day). It was also pretty nerve-wracking. I for one felt like an idiot because I was wearing a black shirt in front of a black background. What a noob! See my little interview below for your viewing pleasure:

Eh? If you liked that, it is probably just because I've heard my boss say those same things over and over and over again... Also I did do a bit of (extremely professional) editing and cut out some parts I messed up. Anyway, it's certainly not perfect, but maybe I managed to put into practice the bits and bobs mentioned above. Hopefully, now you know more about what the Chodera lab does!

Closing remarks

One of my favorite things about all this is that from these short (1-3 minute) talks, I now understand the science of other boot campers  better than the work of scientists I've seen talk for 15 minutes or more at conferences! While this is great, there are still things that bother me about a lot of these techniques of communicating science. What if what you want to communicate is the history of the material of the bases and how that has had an impact on the game of baseball. Something smaller, something that is harder to make seem important. I think most of us were able to make our science personal by talking about how it effects human health, and I would have loved to see more diverse ways of making science personal. I would have loved to take a crack at explaining why we care about the Higgs boson.

This hits on another issue we ran into several times during the course: we don't want to oversimplify. And I think we didn't cover how far is too far very well. On the first day we picked an abstract of someone in the course (Hi, Tali!) to explain to a lay audience: a study on Archer fish (the link is a sweet NYTimes ScienceTake video) that seemed to indicate that fish could do conflict resolution even though they don't have a brain with a prefrontal cortex, which is where humans and mammals do conflict resolution (while this particular abstract is not published yet, here's a link to a related study from the same lab/author). The resulting impression the audience had was: "This research says that an injury to the decision-making section of the brain, may be curable." This made the abstract's author cringe. To me this is exactly what we want to avoid, and I think I am still a little scared of this, and as a result might still fall into jargon sometimes when I explain my research because my instinct is that it is better to communicate poorly than to communicate wrongly. Ideally, we wouldn't do either

These are some of the more complex nuances that I don't think we quite got to cover and clarify in the class. But that's fine. No one said this was going to be easy. One of the most important things we learned, I think, is that we all have our own rhythm and our own stories and that tapping into those is all we need to communicate science effectively. We don't all need to be Bill Nye.

Automatic equilibration detection manuscript on bioRxiv

In molecular simulations---especially simulations of complex systems like biomolecules---it's incredibly difficult to start the simulation close enough to equilibrium to avoid initial transients in properties of interest.  As a result, it is almost universally recommended that some initial portion of the simulation be discarded to "equilibration".  Unfortunately, there hasn't been a simple, automated, and generally applicable way to do this that is standard practice in the field.

In a new manuscript draft posted to bioRxiv this morning, I show how an amazingly simple approach---simply maximizing the number of statistically uncorrelated samples in the latter part of the simulation---can lead to a surprisingly robust and useful algorithm for equilibration detection.  This is very much a work in progress, so comments and feedback is very much appreciated!

DOI: http://dx.doi.org/10.1101/021659

All code needed to grab the exact versions of the tools I used (using the conda package installer and the omnia molecular simulation suite), generate the simulation data, analyze it, and generate the figures for the paper is available on GitHub: You simply need to run

./reproduce.sh

to regenerate everything---which is exactly what I did to generate the figures in the posted version of the manuscript.  There are still a few improvements I hope to make the scripts easier to read and the data easier to deal with, but hopefully we can try to attain this level of ultra-simple reproducibility in future work as well.

Update [5 July 2015]: The manuscript has been updated based on valuable feedback I've already received!  Thanks to everyone who has made comments!

Postdoc Sonya Hanson participates in inaugural Alan Alda Center for Communicating Science Summer Bootcamp

We are beyond thrilled to congratulate Chodera lab postdoc Sonya Hanson for being accepted into the inaugural Scientific Communication Summer Bootcamp at the Alan Alda Center for Communicating Science!  The Center, founded by the veteran director/actor/writer Alan Alda (M*A*S*H, The West Wing, QED, and PBS' Scientific American Frontiers), strives to enhance public understanding of science by working with young scientists and health professionals to develop effective skills for disseminating and communicating science.

The Bootcamp is an intensive week-long program designed to aid scientists in honing their ability to communicate clearly with the public in a multitude of forums, training that is essential for supporting a healthy scientific enterprise in the United State, but which is far too often overlooked in science training programs.  For those interested in what an intensive training program in science communication can look like, the Alda Center has posted the full Bootcamp program agenda online

Sonya has been engaged in scientific communication and outreach activities throughout her career.  She is a former editor of the Oxbridge Biotech Roundtable, maintains an active twitter feed and science blog covering a newsworthy science-related topics, blogs about her research into anticancer therapeutics on the Folding@home blog, was recently a featured guest blogger for the Biophysical Society, and a champion of open source salads.

Stay tuned for more from Sonya on her experiences at the Science Communication Summer Bootcamp.

Automated forcefield benchmarking manuscript on arXiv

Molecular mechanics forcefields are an integral part of molecular simulation.  The quality of any properties computed from molecular simulations is wholly dependent on the quality of the underlying forcefield.  Quantifying how well the forcefields we use can reproduce various physical properties provides insight into expected accuracy in other properties of interest, deficiencies in the forcefield parameters or functional form, and strategies for making systematic improvements.

In a new manuscript posted to arXiv ahead of submission, postdoc Kyle Beauchamp tackles one of the most critical issues in forcefield validation: Most of the physical property information one would like to benchmark against is tied up, inaccessible, in paper databases (also known as "books" or "journal articles").  Using the ThermoML Archive from NIST TRC headed by Kenneth Kroenlein (a coauthor on the paper), Kyle is able to show that this data the computer-readable data stored in this archive in the IUPAC-standard XML-based ThermoML format contains a wealth of information useful for automated validation (and eventually parameterization) of molecular mechanics forcefields.

As usual, all code used in the production of this manuscript is made available through GitHub. The code make use of the excellent OpenEye Toolkit, which is available free for academic use that will generate data for the public domain; the GPU-accelerated OpenMM toolkit, and the free AmberTools distribution.

Kyle A. Beauchamp, Julie M. Behr, Ariën S. Rustenburg, Christopher I. Bayly, Kenneth Kroenlein, and John D. Chodera.
Preprint ahead of submission: [arXiv] [PDF] [GitHub]

OpenEye JCUP VI talk slides

I'm thrilled to have the opportunity to speak to a distinguished collection of scientists from pharma and academia here in Tokyo at the OpenEye JCUP VI drug discovery meeting.

I thought it might help to post a PDF copy of my slides where I discuss how experiments and theory can work together to gain insight into how to improve quantitative predictive models for drug discovery.

Kinase Inhibitor Dorm Room Poster!

Postdoc Sonya Hanson and grad student Julie Behr have collected all 27 FDA approved small molecule kinase inhibitors into a poster suitable for your lab wall or undergrad student's dorm room! A bit more official information on the more recently approved molecules can be found here and an interesting paper that covers the first 23 of these and the types of inhibitors found can be found here. Happy hanging!  [PDF]

[Note: This poster was updated on June 1, 2015 because @mvkrier spotted that it had an uncommon tautomer for regorafenib. This prompted a further search, and it turns out that the original version also had an uncommon tautomer for ruxolitinib. Somehow importing the molecules by their names (e.g. 'regorafenib') in marvinsketch gave us the wrong tautomers. What a great opportunity to mention the Let's not forget tautomers paper by Yvonne Martin! Also... Isn't twitter great?]

 

Chaya Stern awarded NSF Graduate Fellowship

Congratulations to Chodera lab TPCB student Chaya Stern, who was awarded an NSF Graduate Fellowship.

From the TPCB News website:
http://chembio.triiprograms.org/publications-news/news/

Michaelyn Lux and Chaya Stern Awarded NSF Graduate Fellowships

Congratulations to TPCB students Michaelyn Lux and Chaya Stern who have been awarded a National Science Foundation (NSF) Graduate Research Fellowship. They join a select group of 2,000 awardees among 16,500 applicants nationwide. Michaelyn Lux is working with Prof. Derek Tan at Sloan Kettering on the synthesis of novel anticancer agents inspired by natural products. Michaelyn received her Bachelor’s degree in Chemistry from Michigan State University and is interested in studying organic chemistry and its applications to biological problems. Chaya Stern is working with Prof. John Chodera at Sloan Kettering in collaboration with Prof. Harel Weinstein and Prof. Scott Blanchard at Weill Cornell on combining single-molecule experiments with massively distributed molecular dynamics simulations to study G-protein coupled receptor dynamics. Chaya received her Bachelor’s degree in Chemistry from Brooklyn College and is interested in combining theory and experiments within a quantitative and testable framework to study biological macromolecules.

 

Ensembler paper on biorXiv

We have posted a draft of our manuscript describing Ensembler, a new tool to enable omics-scale molecular simulations, on the biorXiv preprint server.  Check it out and let us know what you think!

Read the manuscript: http://dx.doi.org/10.1101/018036
Get the software: http://github.com/choderalab/ensembler

The tyrosine kinase dataset described in the paper will be posted to the Dryad Digital Repository shortly.

Congratulations to postdocs Daniel Parton, Sonya Hanson, Kyle Beauchamp and graduate student Patrick Grinaway for putting this together!

Chodera lab seeking awesome tech to help advance drug discovery science with robots and Python!

NOTE: This posting is out of date. Please see http://choderalab.org/jobs#technician for a current posting!


Are you someone with a life/physical sciences degree eager to help design new therapeutics for cancer and other diseases using robots and cutting-edge computing technology? Looking to do something amazing before heading to graduate school or industry? Enthusiastic about working alongside scientists that integrate computation and experiment to push the boundaries of rational drug design? Then come join us!

The Chodera lab is looking to hire a talented technician to help us maximize our ability to make the most of the unique technological resources at our disposal.

We're looking for someone with:

  • life or physical sciences BS/BA degree

  • strong organizational skills

  • most importantly, an unbounded enthusiasm for SCIENCE!

And one or more of the following:

  • biological wetlab experience

  • eagerness to learn about (or previous experience with) laboratory robotics and automation

  • coding skills (especially Python) a huge plus

In the Chodera lab, you will:

  • work in the basic science arm of a world-class cancer research institute

  • work alongside amazing postdoc and graduate students that integrate computation and experiment

  • become an integral part of several projects that push the boundaries of rational drug design

  • learn the ins and outs of our whole-wetlab automation system

  • help develop new robotic protocols

  • run assays and coordinate with instrument vendors for service

  • help organize wet and dry labs to keep everything running smoothly

  • delve into data analysis and lab infrastructure with Python

  • help blog about exciting science going on in the lab

Official application link: http://careers.mskcc.org/job/5331414/research-technician-computational-biology-new-york-ny/
Please send inquiries to apply@choderalab.org




Congratulations and farewell to postdoc Jan-Hendrik Prinz

Many congratulations to postdoc Jan-Hendrik Prinz, who has been awarded a DAAD Fellowship to work with Peter Bolhuis at the UvA in Amsterdam to continue work on the development and implementation of novel path sampling and Markov state model methods for studying biomolecular dynamics. While we will certainly miss him here in NYC, we look forward to many years of fruitful collaboration yet to come.

If you haven't yet had a chance to check out the path sampling code that Jan-Hendrik and David Swenson (UvA) have produced, you can find it at openpathsampling.org.  It is shaping up to be a fantastic tool for exploring and using path sampling algorithms in a variety of fields!

The Chodera lab is headed to the Biophysical Society Meeting in Baltimore

Four members of the Chodera lab will be headed to the Feb 7-11 Biophysical Society Meeting in Baltimore, MD!
Postdocs Sonya Hanson and Kyle Beauchamp will both be giving talks:

In addition, Sonya Hanson will serve as an official guest blogger for the BPS 2015 meeting!  Be sure to tune into the official BPS Blog to follow along!

Talk slides will be posted after the conference.

Rumor has it the lab will also make a brief appearance at the Society for Laboratory Automation and Screening SLAS2015 conference in Washington DC.

Talk slides for New York Structural Biology Discussion Group (NYSBDG) posted

I had a fantastic time hearing about all of the exciting science going on in NY at the New York Structural Biology Discussion Group 10th Winter Meeting hosted at the New York Academy of Sciences last week.  I had several requests to make my talk slides available, so I'm posting them here in PDF form:

John Chodera's talk slides from the NYSBDG meeting on Wed 21 Jan

Looking forward to many great future meetings of the NYSBDG!

Folding@home comes to Sony Android phones!

You can now run Folding@home on your Sony Android phone!

The GPU-accelerated OpenMM molecular simulation framework has been ported to run on Sony Android phones by the Pande group at Stanford.  We're excited to see Folding@home (which the Chodera lab makes extensive use of) reach a whole new class of hardware.

More information about the Sony Android roll-out can be found on the Folding@home blog.
 

Get the Android app (Sony devices only) here.