Cell Biology Animation

00:00:16.00 Hello, my name is Janet Iwasa,
00:00:18.09 and I am an animator at Harvard Medical School.
00:00:20.17 I'll be talking to you today
00:00:22.07 about some of the reasons I first got interested
00:00:24.06 in animation and the ways I think animation can really give back
00:00:27.18 to the research community.
00:00:29.11 So this animation of kinesin was really one of the first
00:00:32.23 things that I saw that really got me interested
00:00:35.00 in animation. I remember I was about a first or second
00:00:38.14 year graduate student in Dyche Mullins' lab,
00:00:40.24 studying the actin cytoskeleton at the time.
00:00:42.11 And I remember looking at this animation and thinking, why aren't we all doing this?
00:00:47.08 Why are we relying on these oversimplified static illustrations
00:00:52.05 when we can really be doing something like this which shows dynamics
00:00:55.15 and a lot of things way more accurately
00:00:57.05 than we currently are.
00:00:58.13 So within a year I started taking animation courses
00:01:02.00 at a local university and also started doing animations
00:01:05.15 of some of the processes that my lab was studying.
00:01:07.23 So all of my animations are really a close collaboration between myself and the people who are
00:01:13.12 doing the research, and there's a lot of real
00:01:15.25 back and forth that goes on
00:01:18.08 in the process of creating these animations. We go through
00:01:20.13 many, many iterations and change things like color
00:01:23.15 and dynamics, and in this case, the number of filaments,
00:01:26.04 how fast they should be depolymerizing and polymerizing.
00:01:28.15 And so it really is this great collaborative process.
00:01:32.00 Animations, it is really a pretty steep learning curve
00:01:36.03 to learning animations, and it was also...each animation really took me
00:01:39.25 quite a long time, especially at the beginning,
00:01:41.29 but I found that I really loved the process.
00:01:44.01 I loved sitting down with my labmates and trying to figure out what
00:01:48.07 these kind of processes should look like visually.
00:01:51.16 And I also found that it could really give back to the research
00:01:53.05 quite a bit. So, when you are creating an animation,
00:01:55.22 you are really grappling with a lot of issues that don't necessarily
00:01:58.10 come up by any other means.
00:02:00.28 For example, you have to think about not only stoichiometry,
00:02:03.13 but also dynamics and crowding,
00:02:05.17 things that may not come up if you are really creating
00:02:07.23 these very simple illustrations.
00:02:09.15 So towards the end of graduate school
00:02:12.17 I became interested in trying to look for opportunities
00:02:15.12 that I might be able to do animation as a postdoctoral fellow.
00:02:18.08 And I was really lucky to find an opportunity that was offered by the National Science Foundation
00:02:23.02 called the Discovery Core.
00:02:25.11 And so, as a Discovery Core fellow, I worked for two years with Jack Szostak
00:02:30.03 at Mass General Hospital and the Museum of Science in Boston
00:02:33.18 to create a multimedia exhibit on the origins of life.
00:02:36.00 And this used a number of animations that
00:02:38.20 explored ideas around the RNA World hypothesis
00:02:42.05 as well as some ideas of what the early Earth might have looked like.
00:02:45.04 So many of these animations really served a dual purpose.
00:02:49.05 First they were incorporated into a multimedia exhibit that included
00:02:52.15 a website as well as a kiosk on the museum floor
00:02:57.11 that is currently there right now,
00:02:59.00 as well as a number of live presentations that I gave at the Museum of Science.
00:03:05.09 And they were also used by researchers in the lab
00:03:08.07 to talk about their science to other researchers
00:03:11.02 And what we found was that just by altering the context, as well as
00:03:14.12 the narration for these animations, you could really
00:03:16.29 use the same animations for two different audiences.
00:03:19.15 I also grew quite a lot as an animator during this
00:03:23.14 time, so I was really lucky to be able to take kind of a
00:03:27.15 crash course in animation in Hollywood for the summer before my postdoc started.
00:03:32.16 And this really allowed me to hit the ground running.
00:03:35.00 And I also found that there are a lot of tricks that you have to use in animation to
00:03:38.14 create molecular animations. So a lot of the animation packages that we use today
00:03:43.01 are really more for animating things like Buzz Lightyear
00:03:45.15 rather than molecules like actin.
00:03:47.14 So for example for this animation,
00:03:49.12 I used cloth simulation to try to create a simulation of RNA folding.
00:03:55.09 So what I also found was that animation could be a
00:03:58.18 great handle that the public could use to try to grasp
00:04:01.12 and understand complex molecular ideas.
00:04:04.04 So animation can really eliminate the need for jargon
00:04:07.29 in many cases and is also quite approachable.
00:04:10.10 And so for these kinds of reasons I think animation
00:04:13.16 can really be a great tool to try to make a positive impact
00:04:18.16 on science education, as well as the public perception of science
00:04:22.20 and ultimately, and hopefully, science policy.
00:04:25.15 After my postdoc, I started working at Harvard Medical School in the Cell Biology department.
00:04:31.15 One of the things that I am really interested in understanding
00:04:34.09 is how animation can be used by researchers to better explore and communicate
00:04:39.13 their kind of molecular hypotheses.
00:04:42.11 And so this is an example of this kind of project, this is an animation
00:04:46.06 clathrin mediated endocytosis
00:04:48.10 that I worked on with Tom Kirchhausen.
00:04:50.10 And what I found was that animations can really synthesize a
00:04:53.18 great deal of information, including...you can include molecular structures
00:04:59.09 from crystal structures and EM sources,
00:05:02.21 as well as you can include dynamics from light microscopy data,
00:05:06.24 as well as how protein-protein interactions
00:05:09.14 that are derived from genetics and biochemical experiments.
00:05:12.06 And another thing to note is that these kind of animations
00:05:17.01 are really kind of a visualization of a hypothesis.
00:05:19.10 So they may not, everything in the animation may not be completely supported by
00:05:25.01 experimental evidence, while other things may be,
00:05:27.14 and I think that's the same for any kind of hypothesis.
00:05:29.19 I also found that animations can really be
00:05:34.03 quite interesting for researchers, so the process of
00:05:36.23 creating an animation really makes you, as I mentioned before,
00:05:39.26 grapple with different issues that may not come up
00:05:41.22 when you are making other types of illustrations
00:05:44.15 so it is really kind of putting together a three dimensional
00:05:47.13 or a four dimensional puzzle. And by manipulating these different pieces you can really come up with
00:05:51.17 a lot of different ideas that
00:05:53.09 may not come up otherwise.
00:05:55.06 And I think for that reason,
00:05:57.16 these kind of animations and animation tools
00:05:59.28 may be able to really give back to research
00:06:02.12 and the research process.
00:06:04.02 So this final project that I would like to talk to you about
00:06:07.01 is a project that I am working on with Sam Reck-Peterson
00:06:10.00 at Harvard Medical School. Her lab is interested in
00:06:12.11 understanding the motor protein dynein and how
00:06:15.04 it walks along microtubules.
00:06:16.14 So as you might imagine this is a really
00:06:18.12 three dimensional and dynamic process.
00:06:20.15 And so using a 2D illustration such as this one really can be of limited utility.
00:06:26.29 when thinking about these kinds of problems.
00:06:29.03 So what we did together was to create a 3-dimensional
00:06:31.27 model of dynein that the lab could manipulate and move around
00:06:37.00 and really kind of use to explore the molecule in ways that they really
00:06:41.10 weren't able to before.
00:06:43.08 So really the goal of this project wasn't to create a finished and polished
00:06:47.02 animation, but really to create a new tool
00:06:49.15 that the lab could use to better study this process.
00:06:52.13 And using these kind of models, you can really start to
00:06:55.14 visualize the walk cycle of dynein, for example.
00:06:58.29 And each person in the lab, you can imagine, might have a different idea
00:07:02.11 of how this happens, and you can also start to try to predict
00:07:05.02 what might happen if, for example, you mutated the protein, or you put it in
00:07:08.19 to different conditions. And you can compare those things side by side
00:07:11.05 and potentially start designing experiments around these kinds of ideas.
00:07:15.21 And finally I wanted to leave you with a list of resources
00:07:20.10 that I hope might be useful for those of you who are interested
00:07:23.02 in learning more about animation.
00:07:25.01 The first is my website at Harvard Medical School
00:07:27.04 where you can download many of the animations that I showed you today
00:07:30.12 as well as new animations and new projects as they come up.
00:07:34.19 The second link here is called molecularmovies.org, which is
00:07:38.13 a really excellent resource. It has a gallery
00:07:41.09 of animations that you can download, a number of tutorials specific for molecular animation.
00:07:46.18 And there's also, in the past few years, been a number of
00:07:50.13 efforts to try to create new toolkits within
00:07:53.07 animation programs that can be used by researchers
00:07:55.10 to more easily upload and manipulate molecules.
00:07:59.15 And so this includes Molecular Maya, which is available at molecularmovies.org,
00:08:05.01 And this is for the program Maya.
00:08:06.18 There's also EPMV, which is available for Cinema4D, Maya, and Blender.
00:08:11.15 And there's also BioBlender, which is a tool kit for Blender.
00:08:14.08 So I really encourage you to take a look at some of these websites
00:08:18.06 if you are interested in learning more, and also feel free to contact me as well.
00:08:21.24 And also keep in mind that many of the software
00:08:26.02 that is out there you can download an educational version
00:08:29.04 and start exploring these animation tools for pretty much no cost.
00:08:34.01 And so that is something you also might consider doing as well.
00:08:37.02 Thank you.

This Talk

Speaker: Janet Iwasa

Audience:

General Public
Educators of H. School / Intro Undergrad
Student
Educators of Adv. Undergrad / Grad
Researcher

Recorded: September 2011

Talk Overview

Janet Iwasa recalls the animation that first led her to realize how much more information can be included in an animation compared to a static image. She explains that cell biology animation can provide a visualization of a hypothesis and bring together structural data, protein-protein interactions and dynamic information in a process that often helps researchers refine their models.

Speaker Bio

Janet Iwasa

While completing her PhD in cell biology at the University of California, San Francisco, Janet Iwasa began taking classes in animation. She honed her skills as a molecular animator during her post-doctoral fellowship and she has since joined forces with researchers to make animations of various biological processes. Iwasa joined the faculty at Harvard Medical… Continue Reading