The Concord Consortium

No smoke and mirrors here: dragons are getting kids all fired up about genetics. Geniverse software engages students with compelling reasons to solve genetics problems. As they rise through the ranks of the Drake Breeders Guild, students win stars and quills for efficient experimentation and for using their own experimental results as evidence for their scientific claims. Watch how students are learning genetics while having fun—using Geniverse! Want to get your students fired up about genetics, too? Sign up to use Geniverse in your classroom next year.

Our Next-Generation Molecular Workbench (MW) software usually models molecular dynamics—from states of matter and phase changes to diffusion and gas laws. Recently, we adapted the Molecular Dynamics 2D engine to model macroscale physics mechanics as well, including pendulums and springs.

In order to scale up the models from microscopic to macroscopic, we employ specific unit-scaling conventions. The Next-Generation Molecular Workbench (MW) engine simulates molecular behavior by treating atoms as particles that obey Newton’s laws. For example, the bond between two atoms is treated as a spring that obeys Hooke’s law, and electrostatic interactions between charged ions follow Coulomb’s Law.

Dipole-dipole interactions simulated using Coulomb’s Law.

At the microscale, the Next-Generation MW engine calculates the forces between molecules or atoms using atomic mass units (amu), nanometers (10⁻⁹ meters) and femtoseconds (10^-15 seconds), and depicts their motion. To simulate macroscopic particles that follow the same laws, we can imagine them as microscopic particles with masses in amu, distance in nanometers, and timescales measured in femtoseconds. Once the Next-Generation MW engine calculates the movement of these atomic-scale particles, we simply multiply the length, mass and time units by the correct scaling factors. This motion satisfies the same physical laws as the atomic motion but is now measured in meters, kilograms and seconds.

In the pendulum simulation below, the Next-Generation MW engine models the behavior of a pendulum by treating it as two atoms connected by a very stiff bond with a very long equilibrium length. The topmost atom is restrained to become a “pivot” while the bottom atom “swings” because of the stiff bond. Once the engine has calculated the force using the atomic-scale units, it converts the mass, velocity and acceleration to the appropriate units for large, physical objects like the pendulum.

Large-scale physical behavior simulated with a molecular dynamics engine.

In order to appropriately model the physical behavior of a pendulum or a spring, we use specific scaling constants. Independent scaling constants for mass, distance and time enable us to convert nanometers to meters, atomic mass units to kilograms and femtoseconds to model seconds. Using the same scaling constants, we can derive other physical conversions, such as elementary charge unit to Coulomb. In order to make one model second pass for every real second, we adjusted the amount of model time between each page refresh. We also chose to simulate a gravitation field—a feature usually absent in molecular dynamics simulators—because it is relevant to macroscopic phenomena.

From microscale to macroscale, the Next-Generation Molecular Workbench engine is a powerful modeling tool that we can use to simulate a wide variety of biological, chemical, and physical phenomena.  Find more simulations at mw.concord.org/nextgen/interactives.

For nearly 18 years, our logo has been a beautiful and complex sunflower, created by Senior Web Developer Noah Paessel. (He was Noah Fields back in 1994 when he worked at the Concord Consortium during his first stint with us, but that’s another blog post!)

With the former logo, our founder, Bob Tinker, wanted to showcase the Fibonacci sequence in nature, which represents a fascinating link between the sublime and the natural world and invites scientific inquiry and mathematical investigation. (Sunflower seeds exhibit many different Fibonacci spirals in their close-packed patterns, as do many other things in the natural world . Bob also thought Concord as a place evoked important concepts of revolution and free thinking and that the etymology of the name “Concord” linked with the sunflower expressed the ideas of “sharing one’s heart” and being “of the same mind,” both of which resonated with his pacifist and gentle nature.

We are now proud to announce our new logo, created by Derek Yesman of Daydream Design.

This logo both simplifies and augments our original logo. It morphs the original sunflower while also referencing both technology and our core mission of generating, experimenting with and spreading important ideas.

The central star represents the initial spark of an idea, that “a-ha moment” of inspiration that can so quickly turn into extended experimentation – or possibly into a whole new research project. The light bulb surrounding it represents how we work to build these inspirational flashes into complete ideas and products and determine their potential to improve teaching and learning. The petals and radiating elements in the background represent our mission to spread the best of these ideas outward to transform learning for millions around the world.

We’ve recently modified our tag line to make this mission (and our ties to Concord’s location and history) even more explicit: Revolutionary digital learning for science, math and engineering.

By the way, for all you font geeks (don’t hide – we know you’re out there!) our logotype is rendered in Museo 500, part of Jos Buivenga’s excellent Museo family. We discovered this font when we worked with ISITE Design during our last website redesign – thanks Patrick! – and fell in love. Since then, we’ve explored the many weights of this font as well as its sans serif and slab variants. We’ve also had some early-adopter fun watching this font gain status and uptake in many print and Web locations on its way to becoming a modern classic.

We’re excited about this new logo and about how it represents an evolution we’re in the midst of as well. As we evolve toward a new phase as an organization while still embracing our legacy as pioneers in educational technology, we’re more committed every day to creating a bright future for STEM teaching and learning.

It was a great year for the Concord Consortium!

1. We won a Smaller Business Association of New England (SBANE) Innovation Award!
2. Next-Generation Molecular Workbench interactives starred in the MIT MOOC (Massive Open Online Course) “Introduction to Solid State Chemistry” through a new collaboration with edX.
3. Chad Dorsey described our vision of deeply digital education at the national Cyberlearning Research Summit.
4. Six new projects were funded by the National Science Foundation: InquirySpace, Understanding Sub-Microscopic Interactions, High-Adventure Science: Earth’s Systems and Sustainability, GeniVille, Graph Literacy, and Sensing Science.
5. The What Works Clearinghouse (WWC), a federally funded organization that scans educational research for high-quality studies, recognized our Technology Enhanced Elementary and Middle School Science (TEEMSS) software and materials.
6. The Concord Consortium Collection was accessioned into the National Science Digital Library (NSDL).
7. Our debut webcast featured Chad Dorsey, speaking about the scientific and engineering practices of the Next Generation Science Standards and our free, technology-based activities.
8. We had two fabulous Google Summer of Code students.
9. Our staff population increased by 10%, thanks to our new Software Portfolio and Project Manager Jen Goree, Web Developer Parker Morse, and Software Developer Tom Dyer, who just started (technically in 2013, but we’re so excited, we’ve included him on this 2012 list)!

2013 promises to be another great year! Follow us on Facebook, Twitter, Google+, and subscribe to our mailing list to receive print or email news updates.

Gas laws are generally taught in high school chemistry. Students learn that Boyle’s law, for instance, can be expressed as P₁V₁=P₂V₂, where P is pressure and V is volume.

From the equation, it’s clear that there is an inverse relationship between the gas pressure and volume, but do students understand the molecular mechanism behind this relationship?

Since students are programmed to plug and chug, if you give them, say, P₁, V₁, and P₂, they can find the numeric value of V₂. Although students can get the correct answer, teachers have told us that their students don’t really understand the gas laws because they don’t have a mental model of what’s happening. Gases are, after all, invisible! Nor can students see volume or pressure.

Molecular Workbench makes the gases, volume, and pressure visible. With a new set of Next-Generation Molecular Workbench interactives, students can experiment with increasing the pressure on a gas to see why the gas volume decreases.

The “What is Pressure?” interactive (above) shows the inside (yellow atoms) and outside (pink atoms) of a balloon. (Even the velocities of the individual atoms are visible with vectors!) The green barrier represents the wall of the balloon.

Students learn that pressure is nothing more than molecular collisions with a barrier. In the beginning, atoms hitting the balloon wall on either side move it just a tiny bit—transferring some of their kinetic energy to the barrier. At equilibrium, the balloon wall remains (relatively) stationary. (Go ahead and run it to see!)

But if you add atoms to the balloon, the balloon wall moves out; more atoms means that there is increased pressure pushing outwards on the barrier. Since the number of atoms on the outside of the balloon hasn’t changed, the pressure pushing inwards is the same as it was before. With unbalanced forces, you get net movement.

With barriers, we can also measure the pressure caused by those molecular collisions.

In the “Volume-Pressure Relationship” interactive (above), students see a visual representation of Boyle’s law.

Other models allow students to investigate all the relationships of Charles’s law (V₁T₂=V₂T₁), Gay-Lussac’s law (P₁/T₁=P₂/T₂), and Avogadro’s law (V₁/n₁=V₂/n₂).

And, of course, all of these relationships together make up the Ideal Gas Law (PV=nRT). Explore gas laws today with some HTML5 molecular models!

[Editor's note: Piotr Janik (janikpiotrek@gmail.com) was a Google Summer of Code 2012 student at the Concord Consortium and is now a consultant working on our Next-Generation Molecular Workbench.]

Some time ago we described the core engine used in Molecular Workbench and our attempts to speed it up. At that time we focused mainly on the low-level optimization connected with reducing the number of necessary multiplications. This promising early work encouraged us to think even more about performance.

We next reviewed existing algorithms in the core of the molecular dynamics engine. To make a long story short, atoms interact with each other using two kinds of forces:

• Lennard-Jones forces (repulsion and short-range attraction)
• Coulomb forces (electrostatic and long-range attraction)

Atomic interactions are pairwise, meaning that we have to calculate forces between each pair of atoms while using the basic, naive algorithm. Having n atoms, we must perform about n^2/2 calculations. “The Big O” notation can be used and the computational complexity can be described as O(N^2), which means that the execution time of calculations grows very fast as the number of atoms used in the simulation increases. This is definitely an unwanted effect, but fortunately there are ways to reduce the complexity.

Solutions are different for short-range and long-range forces, so let’s start with short-range. “Short-range” means that atoms interact only while they are quite close to each other. Let’s use rCut as a symbol for the interaction maximum distance. So, one obvious optimization would be to limit calculations to pairs of atoms that are closer to each other than rCut. How? There are two popular approaches—cell lists and Verlet (neighbor) list algorithms.

The cell lists algorithm is based on the concept that we can divide the simulation area into smaller boxes or cells. Each cell dimension is equal to the maximum range of interaction between atoms—rCut. So, while calculating interactions for a given atom, it’s enough to take into account only atoms from the same box and its closest neighbors. Atoms in other boxes are too far to interact with this atom. This is both simple and effective, reducing computational complexity to O(N)! Note that it’s C * O(N) with a pretty significant C, unfortunately.

However, while calculating interactions between atoms in neighboring cells, still only 16% of atoms that we take into account are interacting! This is a waste of resources and where we find room for further optimizations. So, what about creating a list for each atom, which contains only atoms actually interacting with it? This Verlet or neighbor list algorithm as it’s called works well. The only problem is that we have to be smart about updating these lists, as atoms constantly change their position and, thus, their “neighborhood.” We can slightly extend these lists to also include some atoms outside the area of interaction. So each list should include atoms closer than rCut + d from the given atom, where d defines a buffer area size. Because of that, lists need to be updated only when the maximum displacement of some atom, measured since the moment of the previous lists update, is bigger than d. If it’s smaller, neighbor lists are still valid. Lists can be updated using the normal, naive algorithm (which still leaves the complexity O(N^2)), or even better, using the cell lists algorithm presented above! This ensures complexity O(N) and greatly reduces inefficiencies of the cell lists approach.

We’re also working on long-range forces optimization. Since we can no longer use the assumption that atoms interact only when they are close to each other, we can’t rely on the optimization strategies above. The algorithms are now more complicated. The problem of the electrostatic interaction is akin to a problem of gravitational interactions (called N-body problem), popular in astrophysics. One of the most common algorithms for speed-up of such calculations is the Barnes-Hut algorithm. We tried to implement it, but the overhead connected with creating additional data structures was bigger than potential performance gains. The reason is that the number of charged atoms we use in our models is too small to see the advantage of such an approach. As a result, we left our naive algorithms for long-range interactions, which perform better due to their simplicity.

However, we successfully implemented both short-range optimizations in Next-Generation Molecular Workbench and the results are spectacular. The speed-up varies from 20% for really small models (where the number of atoms is less than 50) to 700% for bigger ones (where the number of atoms is about 250). This is the really significant improvement and made complex models usable. As you can see, conceptual, algorithmic optimizations really matter!

We’re still thinking about further optimizations, both low level and algorithmic. Stay tuned as the Next-Generation MW is getting more and more computational power!

Thanks to everyone who entered our Suggest-a-Model contest. We always enjoy hearing from teachers and love to help with hard-to-teach science concepts. If you haven’t already, please vote for the model you’d most like us to build.

To Vote

1) Go to our Facebook page (you like us on Facebook already, right?)

2) Look for the poll pinned to the top left of the page’s wall

3) Click on the idea you like most to cast your vote

Our goal is to build a custom computer model to help teach a complex, science, math or engineering concept suggested by real teachers, like YOU! We know all too well the awkwardness of jumping up and down and waving your hands to model the behavior of molecules or dancing around the classroom to model photosynthesis.

We received a lot of great ideas and whittled the list down to three concepts.

One finalist told us that her students “are always making fun of me looking like I am doing a swim stroke in front of the class” when she tries to model convection! She’d love a new set of heat transfer models!

Another finalist is looking for a model of nutrient runoff into coastal waters and how that stimulates harmful algal bloom production. Concerned about the environment? Show your support for this model!

A model of meiosis and genetic recombination (known as crossing over, when exchanges of chromosome portions occurs) also made it to the top three. If you teach biology or know a student who’s taking Bio, this may be the one for you.

Voting ends on November 30th, so please go to our Facebook page and vote now.

After voting is over, we’ll announce the winner and get started on building the model. And once it’s done, it’ll be available for free to everybody. Win-win all around! If you want to know when it’s available, be sure to like us on Facebook, follow us on Twitter and subscribe to our mailing list and RSS feed. We’ll be posting about it through all those channels.

But don’t wait to use our models. Check out our Activity Finder and Classic MW. These free resources contain lots of great examples of the models we already have available for science, math and engineering teachers at all grades. You’re sure to find an activity (or two or three!) that covers other difficult-to-teach concepts. Enjoy!

[Editor's note:  Vaibhav Ahlawat was a Google Summer of Code 2012 student at the Concord Consortium.]

At any time, the Concord Consortium runs a number of small research projects and large scale-up projects, but in the past we built each system separately and each required a separate login. Want to teach your fourth graders about evolution? Great. Log in at the Evolution Readiness portal. Wait, you also teach your students about the cloud cycle? That requires logging in at the Universal Design for Learning (UDL) portal.

Clearly, some students and educators find value across different projects, and my goal is to make it a little easier for them to sign in just once and get access to the myriad great resources at the Concord Consortium for teaching science, math and engineering. As a Google Summer of Code student, I’m working under the guidance of Scott Cytacki, Senior Software Developer, to bring different projects under a single authentication system or, in the language of software development, a Single Sign-On.

Single Sign-On will allow both students and teachers to login across different projects with a single username and password, doing away with the need to remember multiple usernames and passwords. They’ll be able to move seamlessly among projects and find the resources they need to teach and learn. I’m also working on code that will allow students and teachers to sign up and login to Concord Consortium’s portals with their existing Google+ or Facebook accounts.

For those who want technical details, read on.

I’m working on moving from Restful Authentication to Devise, both of which are authentication solutions for Rails. These days, Devise is the preferred one among the Rails community and it makes things like password resetting and confirmation email pretty easy. Once we are done with this conversion, adding the support for signup and login using Facebook and Google+ accounts should be simple. For example, to add support for Google Oauth2 authorization protocol, all we have to do is add a gem named omniauth with Oauth2 strategy, which works brilliantly with Devise, then write a couple of functions.

Here’s a snippet of my code, which adds google oauth2 support to Devise

class Users::OmniauthCallbacksController < Devise::OmniauthCallbacksController
    def google_oauth2
 
    # The User.find_for_google_oauth2 method also needs to be implemented.
    # It looks for an existing user by e-mail, or creates one with a random password
    @user = User.find_for_google_oauth2(request.env["omniauth.auth"], current_user)
 
    if @user.persisted?
      flash[:notice] = I18n.t "devise.omniauth_callbacks.success", :kind => "Google"
      sign_in_and_redirect @user, :event => :authentication
    else
      session["devise.google_data"] = request.env["omniauth.auth"]
      redirect_to new_user_registration_url
    end
  end
end

Including support for authentication using the Facebook API can be done simply. Support for Oauth, which is used by many learning management systems, is provided, making integration far more easier than it was before.

I’m happy to help make it easier for Concord Consortium’s resources to be used by many more people.

– By Vaibhav Ahlawat

At the heart of Molecular Workbench’s modeling of atomic interactions is a profoundly important but fundamentally simple concept:

At close distances, atoms attract each other until they get so close that they repel.

Here’s a demo of that concept: two atoms interacting. Drag the green atom to various locations near and far from the purple atom and watch what happens as the two atoms approach each other and move apart. (If you’re wondering why the atom slows down and stops, the answer is that we apply an artificial damping force to the green atom in order to make it easier for you to “grab” it and play with it.)

This concept is called intermolecular attraction. Molecular Workbench (MW) uses an approximate formula for calculating the intermolecular potential that was originally proposed by John Lennard-Jones (in 1924!) and is now called the “Lennard-Jones potential” or L-J for short.

Here you see the L-J potential as a graph. The horizontal axis shows distance between two atoms, the vertical axis is the net intermolecular energy, with regions of negative slope indicating that the resulting force is repulsive, and regions of positive slope indicating attraction. This graph shows that these atoms will attract to each other if they are more than 2.3 radii apart, but begin to repel sharply at distances less than that value as shown by the steep rise in the curve.

This interaction is not just a fundamental concept in physics and chemistry. It also is quite central to the MW simulation engine. We typically do this calculation tens or hundreds of thousands times per second, especially when we have many atoms interacting, such as here:

If you study our code, going deeper and deeper, you’ll peel away layers of the coding onion, until you get to the very center of this model, which calculates the L-J force between just one pair of atoms. (Did we mention that the reason the L-J approximation is used is that it’s considered relatively fast to calculate?) It does this for each of the many pairs of interacting atoms, repeating over and over, with each time-tick of our simulation.

As it turns out, the L-J formula is still computationally demanding, requiring calculating 6th and 12th powers. (That’s the theory. In practice, the form that is most convenient for our code happens to use the 8th and 14th powers. That also speeds things up–but we’ll save that for another blog post.) So when we’re looking for ways to make our code run faster and our model run better, the L-J calculation is a prime place to look!

In our first pass at improving speed (in MW Classic), we converted the 6th and 12th power calculations to simpler repetitive multiplications:

X² = X * X

X³ = X² * X

X⁶ = X³ * X³

X¹² = X⁶ * X⁶

That gave us just 4 multiplication tasks, instead of 16 (X⁶ has 5 multiplications, X¹² has 11).  This reduced the calculation demand to 25%. The model ran faster and smoother.

Recently, we tried another method to improve speed: using a look-up table, with pre-calculated values. We computed the L-J values for each element type for dozens of typical interatomic distances and put them into a table. We thought this would save computational time because the software could simply look up the values in the table without any multiplication.

We tested the two methods (multiplication vs. look-up table) over many iterations and found that the look-up table was slower! As we investigated further, we saw that accessing the look-up table had its own overhead in terms of use of cache memory and data transfer. This was evidence of a trend in improving computer speeds over the past few years: computation speeds improve at a faster rate than memory access speeds. They both are faster, but computation (i.e., all that multiplying we have to do) has gotten the improvement edge.

So, after this testing, we went back to the efficient multiplication approach. This illustrates our basic approach: creative thinking validated by empirical data.

We will continue to develop and test creative ways to speed the software and improve the user experience, especially as we move to support a greater variety of learning activities and computer platforms.

Just got back from the Biennial Conference on Chemical Education (BCCE 2012), where I participated in a symposium titled “Web-Based Resources for Chemical Education.” About 60 people attended to learn about Molecular Workbench and other online tools and resources. One of the audience questions was about future availability of Molecular Workbench on the iPad and other tablets. Our latest work on the HTML5/JavaScript next-generation MW project, generously funded by Google.org, will address exactly this. We’ll be bringing much of the Java-based Classic MW to the browser, so that any device running a modern Web browser will be able to run our newest interactives and activities.

I didn’t get to attend many of the other sessions at BCCE because much of my time was spent staffing Concord Consortium’s exhibit booth to disseminate our free software. Jeanne Hurtz and I spoke with hundreds of people who stopped by our booth to hear about the current MW capabilities and see a next-generation MW model running on a tablet. We gave away about 350 MW buttons, but have a few left. If you’d like one of your own, please stop by our office at 25 Love Lane in Concord, MA, to pick one up!

It was great to share the excitement of MW’s potential and versatility with so many new people. We heard from many (surprised) guests at our booth: “This is free?”  Yes! And so is the button.