Posts by Ian Davis

1) Message boards : Rosetta@home Science : The Science of Rosetta, first draft (Message 46236)
Posted 14 Sep 2007 by Ian Davis
Post:
Hi all,

As you know, I've been thinking about ways to better explain what Rosetta@home does, and especially how that contributes to curing diseases.

I've written up answers to four basic questions about the science behind Rosetta (plus a bonus fifth question), and it's available here. I'm trying to make it understandable to anyone who's interested in Rosetta@home (without talking down to anyone or being simple to the point of inaccuracy).

Any constructive criticism would be appreciated! Things that still aren't explained well enough, vocabulary, style, things that are over-emphasized, important questions that didn't make the list, etc.

Thanks,
Ian
2) Message boards : Rosetta@home Science : Help me explain the science behind Rosetta@home! (Message 46017)
Posted 11 Sep 2007 by Ian Davis
Post:
What criteria is used to decide whether the energy of the "trial" configuartion is right to make a new "accepted" one? It can not be simply that its energy is lower than the one of the "accepted" configuration it was obtained from, since otherwise the energy would be always reducing, hence the last "accepted" configuration would always be the lowest energy one, which is not what happens.


Correct. Lower energy configurations are always accepted. Higher energy configurations are sometimes accepted, based on a virtual "roll of the dice": if it's only a little higher energy than before it's much more likely to be accepted than one that's much higher energy. This helps avoid getting trapped in local minima, but means you'll almost never accept a really bad configuration. Look up "Metropolis Monte Carlo" for more info.

How does Rosetta decide that the calculation of a particular trajectory has come to an end? For the same reason as before it can not be that it has found a local minimum.


Also correct. The usual answer is that it just takes a set number of steps and then stops. That number is determined empirically for different problems based on how long the simulation takes to converge. In some cases, it may bail out early if it looks like things are going very badly, and just start over.
3) Message boards : Rosetta@home Science : Prototype of Rosetta "game" (Message 45521)
Posted 27 Aug 2007 by Ian Davis
Post:
Here's the link: the Rotamer Game


There's now a second game up on this page, in several different flavors. You can see some of the things we're trying out in terms of visual representation, and the simplified interface for the game. The new one is much harder than the first, but unfortunately there's no way to check your answer right now :( We hope to have that working in a week or so...

Best,
Ian
4) Message boards : Rosetta@home Science : Prototype of Rosetta "game" (Message 45517)
Posted 27 Aug 2007 by Ian Davis
Post:
Sometimes i got exception in java console like these:
(errors deleted


Did they cause the game to not work, or they just look ugly? :) I often see those messages too, but they're harmless; they come up because the software the game is based on also runs as a desktop application and it complains about the restricted environment of the web browser.

-Ian
5) Message boards : Rosetta@home Science : Prototype of Rosetta "game" (Message 45280)
Posted 22 Aug 2007 by Ian Davis
Post:
Based on the feedback here, I'd say that Java 1.6 (aka Java 6) isn't working with the game, at least on Windows. I don't know why, but am trying to figure it out. If I can figure it out I'll post something here and try to fix the game.


OK, I believe it's fixed now. Totally quit out of your browser and/or clear your cache and/or reboot if you're still having trouble. Does that fix the problem for everyone who tried Java 6 under Windows and got the red X?

Best,
Ian

PS *Much thanks* for helping track down this bug! I never would have caught it on my own...
6) Message boards : Rosetta@home Science : Prototype of Rosetta "game" (Message 45278)
Posted 22 Aug 2007 by Ian Davis
Post:
Yes, it's a bug in Java 6 under Windows (http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6570445, if you're interested).

The temporary fix is to remove Java 6 and fall back to Java 5 (JRE 5.0):

http://java.sun.com/javase/downloads/index_jdk5.jsp

There is no workaround listed in the bug database, but I'll see if I can figure one out...

-Ian
7) Message boards : Rosetta@home Science : Prototype of Rosetta "game" (Message 45255)
Posted 21 Aug 2007 by Ian Davis
Post:
Worked fine with Java 5 Update 4 (+ Firefox).
Had to update Java today to version 6 Update 2 and all I get to see now is that darn red X.
(On the game-screen it said: "If you're having trouble loading the game, update Java to the latest version". Oops :))


Based on the feedback here, I'd say that Java 1.6 (aka Java 6) isn't working with the game, at least on Windows. I don't know why, but am trying to figure it out. If I can figure it out I'll post something here and try to fix the game.

So, if you're having trouble, maybe try one of the older Java 1.5 versions instead of Java 1.6. If that solves the problem for you, please let me know! I apologize for the inconvenience but greatly appreciate the help from all of you in tracking down the problem.

Best,
Ian
8) Message boards : Rosetta@home Science : Prototype of Rosetta "game" (Message 45071)
Posted 16 Aug 2007 by Ian Davis
Post:
Actually your demo has helped me understand just how difficult all of this is, considering that it's only one side chain, that has already been fixed in place by the proteins / molecules placement.


Great! That was a really important goal for this project.


Will the game allow abinito structure design? A lot of the abinito designs produced by Rosetta seem a little off to me and I would like to play with them, especially if PDB's could be swapped between other players.

If so would Rosetta be able to do the full atom relax on the abinito structures? So you didn't have to set every side chain by hand.


Yes -- eventually. The "full" game will be a program you download, and should allow you to do all kinds of crazy things to the structures Rosetta is predicting :) We're planning to allow all the things you mention here. This more limited version I'm demoing on the web is intended more as a training ground, to help people understand the principles they'll need for playing the full game (when it becomes available later).
9) Message boards : Rosetta@home Science : Prototype of Rosetta "game" (Message 45070)
Posted 16 Aug 2007 by Ian Davis
Post:
I have the latest JAVA but I don't get the game...

Could you tell me which OS (Windows XP, Windows 2000, OS X, Linux, etc) you're using and what web browser (including version number) you're using? I can't promise I'll be able to fix it but I'll try...
10) Message boards : Rosetta@home Science : Prototype of Rosetta "game" (Message 44975)
Posted 14 Aug 2007 by Ian Davis
Post:
Hi R@H community,

I've been working with several others in the lab on an "interactive Rosetta" project. The vision is that one day, you'll be able to interact with Rosetta as it runs, both (1) to help it produce better results and (2) to learn about what it's doing.

We're not there yet, but we have an early prototype for #2, the educational side. This is a simple "game" you can play in your web browser (using Java) to find the missing side chain for a designed protein that might be used to fight cancer. In this case, there's a small number of choices and the right answer should be obvious. However, it illustrates one of the basic steps that Rosetta performs thousands of times every time it designs a protein.

More "levels" for the game will be coming in the future. Also we're already planning to change the interface on the Java applet and to add a "score board" so you can compete against others, but if you try it out and have other comments or ideas to share, please post them here.

Here's the link: the Rotamer Game

Thanks!
Ian
11) Message boards : Rosetta@home Science : Help me explain the science behind Rosetta@home! (Message 42370)
Posted 20 Jun 2007 by Ian Davis
Post:
Is the disease related research terminated or still going on? That question should be covered in the first place. A subpage, with a list of the diseases covered and a short progress message (1 sentence is enough) on the current stage of the research (planned, ongoing, terminated, results published, ...)
This type of progress report encourages many people to keep on supporting the project at WorldCommunityGrid and Einstein.


Just to be explicit, yes, disease-related research is still going on. The summary on the front page is the best source *I* know of right now for what diseases are currently being targeted. But the most important thing is that as Rosetta improves, it will become possible to target all disease in a much more precise way, leading to more effective treatments and fewer side effects, because we're not thrashing around in the dark. In the long run, I think projects like this will revolutionize medicine, but it's important to realize that it will unfold over a period of years and decades -- we're not walking down the hall to inject Rosetta in a sick patient :)
12) Message boards : Rosetta@home Science : Help me explain the science behind Rosetta@home! (Message 42368)
Posted 20 Jun 2007 by Ian Davis
Post:
One area of confusion has been all of the various fields of study and methods that are under investigation in BakerLab. There is abinitio, docking, symmetric; there is robetta, Rosetta@home, BakerLab's own Linux cluster; AIDS research, cancer, Alzheimer's, CO2 removal. It is often difficult to connect the dots and distinguish between these various facets.


In response to, How are all the things the Baker lab works on related? The lab does work on a huge variety of topics. What connects them is computational modeling of protein structures: all the projects require, in some way, that we use a computer to predict what shape a protein will have and/or how it will interact in a given situation (without doing "wet" experiments -- though we often test our predictions afterwards in the real world).

Also some description of other organizations and scientific areas that Dr. Baker and team are collaborating with would help us to see how the vast area of medical science research pieces fit together. A chart showing the various entities and a narrative of how you could go about getting from point A, a person with a disease, to B, disease identification, to C, docking experiments, to D, designed drug candiate, to E, drug trials would really be helpful. Perhaps use the AIDS research as a frame of reference. Seeing the whole picture would help understand where BakerLab fits in to the study of AIDS. Often people ask how close Rosetta@home is to finding a cure. Seeing where BakerLab fits in to the bigger scheme would help people see why that is not exactly what they are working on. ...and why their research is important anyway.


In response to, How does Rosetta@home research lead to cures for diseases? Proteins are involved in performing almost everything that goes on in your body. Diseases usually result from some breakdown in the system -- cancer can be caused by out-of-control signaling proteins, HIV proteins attack our immune system, and Alzheimer's is thought to be caused by damaged proteins building up in clumps in the brain. So the Baker lab is interested in developing techniques to (1) understand these processes more quickly or cheaply (e.g. structure prediction, docking) and (2) invent proteins or drugs that can fix some of these bodily processes when they start to go wrong.

Here's my attempt at a narrative: suppose I go the doctor's office for my annual checkup some years down the road. He draws a little blood and sends it to the lab, where it's tested with special sensor proteins that were designed using Rosetta. These sensors report on mundane things, like how much glucose is in my blood (i.e., am I diabetic?), and special things, like whether there are any molecules that are known signals for cancer. This could be very fast: 10 minutes later my doctor returns and says most things look good, but I probably have an early stage cancer. Other tests, MRIs, etc. confirm I have a small tumor: not good. But exquisitely sensitive designed sensor proteins (Rosetta again) are used in a few more tests, and they can tell it's a specific subtype that is known to be caused by a specific malfunctioning protein. (It took other researchers in other labs years to work that out, of course, using a wide variety of techniques.) Although no one has every determined the structure of this protein directly, they didn't have to, because Rosetta was used to predict its structure with very high accuracy. That prediction was so accurate that Rosetta could also be used to predict what sorts of molecules might bind to it and stop it from causing problems, leading to the development a drug that effectively cures my cancer. Of course, chemists at a pharmaceutical company had to work out how to manufacture the drug efficiently, and doctors and nurses did years of clinical trials to make sure it was safe and effective.

Does this help at all?
13) Message boards : Rosetta@home Science : Help me explain the science behind Rosetta@home! (Message 42340)
Posted 19 Jun 2007 by Ian Davis
Post:
Wow, great discussion! Let me try to respond to a few general points, and then a few specific things:

1. Relevance to medicine/health. This seems to be the most important issue to most folks -- seing how what they're doing with R@H is going to help cure cancer/AIDS/Alzheimers. I want to make sure we're enthusiastic while remaining honest. You can't say the Wright Brother's single-handledly enabled your trans-Atlantic vacation, but you can say their contribution was vital :) So it is with R@H. I think clarifying #3 below would help a lot with this too.

2. Appropriate level for materials. 9th grade level isn't so far off -- many people don't have much if any science after high school, and probably forget a lot of what they knew. It also plays well with the other face of my project, which is to do science outreach to high school kids via Rosetta. Demand for more advanced materials seems to be either about specific projects now in progress, or technical facets about computing power etc.

3. Diversity of projects. The Baker lab works on a *lot* of things -- there were 30+ people at group meeting yesterday, and they all have their own research projects. That's why it seems like Rosetta does so many different things. They *are* all connected, it's just hard to see unless you're very familiar with the field. I agree that some overview of what all Rosetta@home works on would be useful.

4. Communication with users. I hear that people want to know more about what their computers are doing at any given moment, and more (understandable) info from the screen saver. I know a little work is being done on the screen saver at least, so I'll mention it, but it is outside my domain.

5. Previous threads. I know there have to be other great threads from the past with explanations you found helpful. If we could organize some of them by topic that might help answer some of the higher-level questions about Rosetta itself. Post links to your favorites?

And now the specific things:

Feet1st, thanks for the link to Laura's thread. Some great ideas in there too.

greg_be, thank you for the suggestions. Many sound quite good, though as Mod.Sense point out, some may be beyond my scope.

The_Bad_Penguin, your questions are good ones, but more technical than the level I had been thinking of addressing. The CASP competitions are always discussed extensively in the scientific literature -- I think there's usually a special issue of some journal devoted to each one, the results, how different labs performed. Would that be an appropriate place to direct you?

Tom P, I like that list. I'm definitely envisioning something structured, not free-form, I just wondered if a collaborative model could be used to create it. The more I think about it, though, the more I think not.
14) Message boards : Rosetta@home Science : Help me explain the science behind Rosetta@home! (Message 42292)
Posted 18 Jun 2007 by Ian Davis
Post:
Hi, Rosetta@home community!

I just joined the Baker lab as a new postdoc, and I'll be spending part of my time developing materials to help participants understand the science behind Rosetta@home -- what it does, why it does it, and how the research relates to chemistry, biology, and medicine.

Here's what I need from you:

1. Questions! What do you wish you knew about Rosetta@home? What things are hard to understand, or badly explained?

2. How would like to learn about it -- writen explanations, info-graphics and pictures, video, interactive "games" and graphics, or something else?

3. What resources have you already used to help understand R@H? What did you like and dislike about them? (I'm going to be creating some new materials, but I also hope to take advantage of some of the great science resources already on the web.)

4. I would love help with this project! If you would be interested in helping, what would you like to contribute? Should there be a R@H wiki (like Wikipedia) or just submit contributions by email?

Be patient with me as I get up to speed in the lab, but I'm really looking forward to contributing to the Rosetta@home community!

Best,
Ian






©2024 University of Washington
https://www.bakerlab.org