Hi, I'm from Norway, English isn't my strongest side. I just wondering: When I get some new work, are,(is?), there any way of telling EXACTELY what my computer are,(is?), checking.

And What ever REALLY hapens to the resulst I'm returning????

---

Random Silliness

Brynjar, I`ll ask Trondos our Norwegian team member to come and help you out, alternately you can post here in Norwegian and I`m sure he will get back to you.

---

Not all Czech`s bounce but I`d like to try with Barbar ;-)

Make no mistake This IS the TEDDIES TEAM.

Brynjar, I`ll ask Trondos our Norwegian team member to come and help you out, alternately you can post here in Norwegian and I`m sure he will get back to you.

Bryjar, you are also very welcome to post *here* in Norwegian; and Trondos, is welcome to reply to your question in Norwegian, and so is anyone else who knows how.

This is an international project and nobody should feel that they can't get peer support because of language difficulties. The project team is English speaking, so obviously postings in other languages will be relying on peer support -- but in my experience on several projects is that the peer support is better than the project support... ;-)

m8

Thank for answers. Thanks for trying to find "Trondos", but You see, I got a nice opportunity to brush up on my rusty English. We do have english as a side language on shool from the 4.ft grade, but I ended my school for approx 20 years ago.

I tryed to upload a pic of my son as an avatar, but the web page just got blanck, and thats it.
Any suggesions??

---

Random Silliness

Brynjar the picture must be avatar sized to 100 x 100 or less

You can download a free avatar sizer here which will create the correct size for you.

Trondos will get here, as with most Norwegians he`s busy keeping warm probably ! Normally with a Swedish girl ;-)

---

Not all Czech`s bounce but I`d like to try with Barbar ;-)

Make no mistake This IS the TEDDIES TEAM.

Hurray!!!

The Avatar shrinking-picture-program was quite nice. Thanks!

Swedish girls??? Oh, no, the Norwegian girls are much better!! :)

---

Random Silliness

When I get some new work, is there any way of telling EXACTLY what my computer is checking?

And What ever REALLY happens to the results I'm returning????

If you're coming from a different project (such as FaD) where each job was for a different disease, then you need to realize that we're working at improving the Rosetta client so that it can be used by other projects in the future to find cures/treatments.

If you're just curious about what each Rosetta job is working on, and what type of molecule that job is dealing with, then someone else can help demonstrate how to tell what each job is dealing with. And tell how to find our best score, and the best score for that job by the whole Rosetta community.

So, the results you're returning are being used to find out which improvements to the Rosetta client have the greatest benefit.

(And to those with more experience with Rosetta - please correct me if I'm wrong.)

---

Thank You!!
But as You try to explain, some new questions. Wath are "FaD"???

Anyone knows how the "Score" are calculated. And what do I do with all the points I have earned????? Can I use them to purcare some other colours at the animated screensaver???

---

Random Silliness

FaD= Find a Drug. It's a DC (distributed computing) project that recently closed down.

And which score are you interested in (are all of them?) Under your avatar on this forum, it lists your total score, your RAC (Recent Average ?? Score - although the half life mathmatical explanation for the RAC score doesn't sound like "average" to me.) Are you interested in determining the score for every uploaded result? Or how close we're getting to the actual molecule/protein we're working on? (The thread labeled similar to "Is 1.07 Angstroms for xxxxxxx a mistake?") On Distributed Folding, the distance from the actual molecule/protein was calculated using Root Mean Square Deviation - which I pictured as them taking each of the 40 nodes of a 40 node molecule and comparing how far off each predicted node was from the actual node's location. Since we were dealing with 3 dimensional space, they'd square the difference in the X axis, add it to the difference in the Y axis, and add that to the square of difference on the Z axis. Then take the square root of the result of (x^2 + y^2 + z^2) and you'd get how close the prediction was from the actual location for that node. Basically, the final distance reported was the average of the differences for every node.

There's a recent thread dealing with the RAC score, but the others are either in older threads or discussed in the BOINC Wikipedia. Mention which score or scores you're interested in, and some of us can track down links to the appropriate discussions.

If there's an option for setting colors in the screen saver, perhaps someone who's played with it will post a solution.

---

FaD= Find a Drug. It's a DC (distributed computing) project that recently closed down.

And which score are you interested in (are all of them?) Under your avatar on this forum, it lists your total score, your RAC (Recent Average ?? Score - although the half life mathmatical explanation for the RAC score doesn't sound like "average" to me.) Are you interested in determining the score for every uploaded result? Or how close we're getting to the actual molecule/protein we're working on? (The thread labeled similar to "Is 1.07 Angstroms for xxxxxxx a mistake?") On Distributed Folding, the distance from the actual molecule/protein was calculated using Root Mean Square Deviation - which I pictured as them taking each of the 40 nodes of a 40 node molecule and comparing how far off each predicted node was from the actual node's location. Since we were dealing with 3 dimensional space, they'd square the difference in the X axis, add it to the difference in the Y axis, and add that to the square of difference on the Z axis. Then take the square root of the result of (x^2 + y^2 + z^2) and you'd get how close the prediction was from the actual location for that node. Basically, the final distance reported was the average of the differences for every node.

There's a recent thread dealing with the RAC score, but the others are either in older threads or discussed in the BOINC Wikipedia. Mention which score or scores you're interested in, and some of us can track down links to the appropriate discussions.

If there's an option for setting colors in the screen saver, perhaps someone who's played with it will post a solution.

Mmmmmmmmmmmm..................
Yes, right. I only had 9 years of ordinary school, and then 4 years of car teck education. So Your math seems pretty good, I think. Yeah. Uhm......
Can You please say that again, so I understand that?? The colouration of the screensaver was a kind of joke, referring to the "Credit" that I earning. Just wandered what that "Credit" was good for....

But Your math exempel... Sorry, that was so high over my head that I couldn't see it, even if I were to use the Hubble. Sorry again. (",)

---

Random Silliness

RMSD is used by Statistics people. I'm not a statistician. (Is this the time to claim that I stayed at a Holiday Inn last night? - reference to a joke about a advertisement in the USA) http://mathworld.wolfram.com/Root-Mean-Square.html isn't really helpful to me; but this is what I think it means.. (I didn't find a well written and illustrated description for us non-statisticians..)

Here's a picture of something we worked on at Distributed Folding. The actual item is in yellow, and the prediction is in blue. http://www.blueprint.org/proteinfolding/distributedfolding/distfold_1enh.html
It's a 54 (atom/protein) length structure; and you can tell that the structures do not match perfectly. For each of the 54 atoms/proteins that make up this structure, there's a difference from where the predicted atom/protein is and where the actual atom/protein is. There's the distance left and right (the x axis), the distance up and down (the y axis), and the distance in and out (the z axis) since this is a 3 dimensional object. If for Atom#1, the predicted atom was 1.5 Angstroms off from the actual atom on the X axis, 1.25 Angstroms off from the actual atom on the Y axis, and 0.25 Angstroms off from the actual atom on the Z axis then we'd square the X value, square the Y value, square the Z value, and add them together. Then take the square root of the result.
Sqr root (X*X + Y*Y + Z*Z) or Sqr root (1.5*1.5 + 1.25*1.25 + 0.25*0.25) which equals Sqr root (3.875) or 1.9685019685029528

That would be done for all 54 atoms.. then they'd add up all 54 values, and divide it by 54 to get the average.
Then subtract the average from each of the atom values and get the "deviation from the mean" for each atom.
Then they'd square all 54 deviations, add them together, and take the square root of the total. I think. A statistician will hopefully come by and correct me or point out a web page that has it illustrated so it's easier to understand.
It's just a mathematical method of describing how close something is to another object or value.

On the other hand.. When we were at Distributed Folding, a RMSD of 4 Angstroms was considered good (and supposedly "useable" by the industry; a RMSD of 2 Angstroms was considered fantastic; and a RMSD of 1 Angstrom was a wishful dream for larger molecules in the 100-150 length range. So the Rosetta client is producing some impressive results, if we're working with structures near the 100 length range or larger.

As for the use of the Credits.. on other projects, it's been claimed it was for Imagined Penile Enhancement. *grin* It shows how much you've contributed to the project, and can be used for competition. If you leave your machine running longer, then you can catch up to someone slightly ahead of you, for example. A way of having fun while benefiting the science. Rivalries between teams are usually just to encourage each team to produce more which ends up benefiting the science.

Did I sidetrack you from finding out where a particular score you were interested in came from?

---

It's a 54 (atom/protein) length structure; and you can tell that the structures do not match perfectly. For each of the 54 atoms/proteins that make up this structure, there's a difference from where the predicted atom/protein is and where the actual atom/protein is. There's the distance left and right (the x axis), the distance up and down (the y axis), and the distance in and out (the z axis) since this is a 3 dimensional object. If for Atom#1, the predicted atom was 1.5 Angstroms off from the actual atom on the X axis, 1.25 Angstroms off from the actual atom on the Y axis, and 0.25 Angstroms off from the actual atom on the Z axis then we'd square the X value, square the Y value, square the Z value, and add them together. Then take the square root of the result.
Sqr root (X*X + Y*Y + Z*Z) or Sqr root (1.5*1.5 + 1.25*1.25 + 0.25*0.25) which equals Sqr root (3.875) or 1.9685019685029528

That would be done for all 54 atoms.. then they'd add up all 54 values, and divide it by 54 to get the average.
Then subtract the average from each of the atom values and get the "deviation from the mean" for each atom.
Then they'd square all 54 deviations, add them together, and take the square root of the total. I think. A statistician will hopefully come by and correct me or point out a web page that has it illustrated so it's easier to understand.
It's just a mathematical method of describing how close something is to another object or value.

On the other hand.. When we were at Distributed Folding, a RMSD of 4 Angstroms was considered good (and supposedly "useable" by the industry; a RMSD of 2 Angstroms was considered fantastic; and a RMSD of 1 Angstrom was a wishful dream for larger molecules in the 100-150 length range. So the Rosetta client is producing some impressive results, if we're working with structures near the 100 length range or larger.

Thank You for REALLY trying!! But I don't have the proper education to understand this math. Sorry.

I guessed that the "credit" was for competition among the participents, without any "grand price" waiting at the other end of the tunel. Hehehe.
That guesture are a typical US phenomen i guess... Hehehe.

---

Random Silliness

---

---

Random Silliness

Sorry that I couldn't explain that better.

Math and Chemistry were fun for me.. which is why I've got my 24 year old paper with a proposed "safe" way of making nitroglycerin. Fortunately, I never tried it. Black powder was dangerous enough. :) I have had to find ways of visualizing the chemistry/biology concepts in some of the DC projects I've been involved in, so understand your difficulty. Wish I could have helped more.. (I'm used to others jumping in and helping explain things like this, especially when I'm failing. :)

---

Sorry that I couldn't explain that better.

Math and Chemistry were fun for me.. which is why I've got my 24 year old paper with a proposed "safe" way of making nitroglycerin. Fortunately, I never tried it. Black powder was dangerous enough. :) I have had to find ways of visualizing the chemistry/biology concepts in some of the DC projects I've been involved in, so understand your difficulty. Wish I could have helped more.. (I'm used to others jumping in and helping explain things like this, especially when I'm failing. :)

Nah! Thanks for trying! I guess that nobody could explain this chemi thingies to me. You see, I one of thouse who wonder: If all them factories are pumping out all that Co2, and thats bad. That I get, but the main ingredience in soda water is Co2,(istn't it????), why havent anybody tryed the connection??
Saved the enviroment, and make all then Cola drinkers happy, and perhaps earned some money too....

---

Random Silliness