...just wanted to say thanks to Rhiju for all the information he provided in the Workunits and Release Log threads. This is very much appreciated and I want you to know that there are eager readers of all the information you provide. On reading through the info on the different strategies that are being tested, I actually wondered whether I am informed equally well about the activities of my colleagues sitting in the offices next door. Perhaps we also should place our work logs on the Web... ;-)

---

Team betterhumans.com - discuss and celebrate the future - hoelder1in.org

...just wanted to say thanks to Rhiju for all the information he provided in the Workunits and Release Log threads. This is very much appreciated and I want you to know that there are eager readers of all the information you provide. On reading through the info on the different strategies that are being tested, I actually wondered whether I am informed equally well about the activities of my colleagues sitting in the offices next door. Perhaps we also should place our work logs on the Web... ;-)

I agree.

Keep up the good work and this projekt will last long.

Anders n

---

Absolutely agree, it was very interesting read.

Some things I'd like to see discussed, are:

1/ how the "random seed" (now provided with each WU, used to be created on our PC on the fly) is used to provide completely independant runs for a WU that might run from 2hr to 4days.

2/ crunchers loading their PC's BOINC queue with 50-60 WUs at a time (of the default 2hr duration) to "last" for many days.

3/ Ideal WU round-trip times, considering "smart resampling strategy". I mean, if one "loads the boat" downloading 60 * 2hr (default WU runtime) FARELAX_NOFILTERS_xxxx WUs to "last" him for 3-4 days, EVEN IF HIS PC REPORTS BACK EVERY 2hr after each WU is finished, many won't be useful because after the first 10k results we may focused on a different part of conformational space, right? And the least desirable "behaviour" would be reporting all 60 WU together after 4 days (because that PC will probably have spent its time looking randomly, rather than being "smart" and focusing on the most "promising" space).

1. Testing a smart resampling strategy:

FARELAX_NOFILTERS_xxxx
FACONTACTS_RECENTER_NOFILTERS_xxxx
Have you ever wondered whether there's a better way to fold proteins than have 100,000 clients do completely independent runs, without talking to each other?
These workunits are testing a new strategy that may be smarter. We look at the full-atom scores and conformations from the first 10,000 models (that's the first workunit). For generating the second round of 10,000 models we then adjust the initial stages of the search (which uses a low resolution score function) to favor contacts that gave low energies in the first round.

If I got it right, I think this is another thing to be handled by BOINC server scheduler/feeder, so it sends WUs which necessitate a fast round-trip-time only to PCs which work 24/7 with permanent Internet connection and report back every few hours.

---

Best UFO Resources
Wikipedia R@h
How-To: Join Distributed Computing projects that benefit humanity

Ty for the Information.

---

This and no other is the root from which a Tyrant springs; when he first appears he is a protector.”
Plato

If I got it right, I think this is another thing to be handled by BOINC server scheduler/feeder, so it sends WUs which necessitate a fast round-trip-time only to PCs which work 24/7 with permanent Internet connection and report back every few hours.

I think you have a good point. You might consider using Truxoft's enhanced BOINC available in the BOINC related downloads. It has an option to force the WU to be reported back immediately after computation. I have found it does not report back immediately unless the report_result_imediately parameter is the first parameter in the config file.

I hesitate to mention the above "solution" because it is not the ideal solution, perhaps not even a good solution unless ALL Rosetta hosts adopt it AND Rosetta limits the number WUs a host can have in its possession at any given time.

Rosetta has some rather unique and interesting needs that challenge the capabilities of the current Distributed Computing system. We all have much to learn.

--

Have you ever wondered whether there's a better way to fold proteins than have 100,000 clients do completely independent runs, without talking to each other?
These workunits are testing a new strategy that may be smarter. We look at the full-atom scores and conformations from the first 10,000 models (that's the first workunit). For generating the second round of 10,000 models we then adjust the initial stages of the search (which uses a low resolution score function) to favor contacts that gave low energies in the first round.

God damn it...
I was sure you are using some sort of 'genetic algo' for the project...
What you are saying is that till now we've been using a 'brute force'?
Cheers.

---

I was sure you are using some sort of 'genetic algo' for the project...
What you are saying is that till now we've been using a 'brute force'?
Cheers.

There was a thread about using GA a few months ago:

https://boinc.bakerlab.org/rosetta/forum_thread.php?id=780

Apparently it also requires some tweaking of the BOINC server code.

PS: Perhaps there should be a separate thread to discuss "Active WU(s) log", like we do about the science journal, and keep this one clean for project scientist posts only?

---

Best UFO Resources
Wikipedia R@h
How-To: Join Distributed Computing projects that benefit humanity

PS: Perhaps there should be a separate thread to discuss "Active WU(s) log", like we do about the science journal, and keep this one clean for project scientist posts only?

I agree! Let's keep this thread posting limited to Project people only, so as to keep it focused for fast and easy reference!

---

Regards,
Bob P.

This thread is for discussion of the proteins and Work Units listed in this thread.

---

Moderator9
ROSETTA@home FAQ
Moderator Contact

I hope this is the place to ask this. I've seen the size of proteins discussed in terms of how many amino acids and seen models with say 1.5 Angstrom resolution. A friend was watching Rosetta run a model of protein 1tul on my PC today and asked how small it was as in the diameter of the native form. I couldn't answer the question. Does someone know the approx. distance across one of these proteins that Rosetta is using these days such as 1tul in Angstroms or Nanometers? I've searched the site and couldn't find anything.

Thanks,
Bill

I hope this is the place to ask this. I've seen the size of proteins discussed in terms of how many amino acids and seen models with say 1.5 Angstrom resolution. A friend was watching Rosetta run a model of protein 1tul on my PC today and asked how small it was as in the diameter of the native form. I couldn't answer the question. Does someone know the approx. distance across one of these proteins that Rosetta is using these days such as 1tul in Angstroms or Nanometers? I've searched the site and couldn't find anything.

Thanks,
Bill

I am not a bio science guy but, If I read this description of 1tul correctly, the reference to "Resolution" gives some idea of the size. I can't tell if this is the field size required for the thing to fit into to be resolved or the actual size of the protein itself.

1tul is for a virus so it must be vary small.

---

Moderator9
ROSETTA@home FAQ
Moderator Contact

I am not a bio science guy but, If I read this description of 1tul correctly, the reference to "Resolution" gives some idea of the size. I can't tell if this is the field size required for the thing to fit into to be resolved or the actual size of the protein itself.

As a rule of thumb the size of an atom is about 1 Angstrom; so if the resolution isn't much larger than 1 Angstrom this is close-to-atomic resolution and it can't get any better than that. I would guess that the size of a small protein in its native state must be something like 50 atoms across, so the size of a protein would be a few Nanometers or more (I am not a bio science guy either, I am sure the experts will be able to give you more accurate numbers).

---

Team betterhumans.com - discuss and celebrate the future - hoelder1in.org

I would guess that the size of a small protein in its native state must be something like 50 atoms across, so the size of a protein would be a few Nanometers or more (I am not a bio science guy either, I am sure the experts will be able to give you more accurate numbers).

a little bigger, you have to take into account the size/length of the bonds (between atoms and between amino acids) as well as the fact that there are quite a few atoms per amino acid, and quite a few aminos per protein
for example a carbon-carbon bond is something like 150-250 nanometres, so an average protein is probably a few thousand nonometres across, however i don't claim to be a true expert, i'm sure one of the project scientists could give a definate answer

there are also probably more than just 50 atoms in even the simplest of proteins

---

Want to search the BOINC Wiki, BOINCstats, or various BOINC forums from within firefox? Try the BOINC related Firefox Search Plugins

I did some poking about, and here are some sizes from hydrogen to a eukaryotic cell, to give ideas of sizes... amino acids and simple globular proteins are in bold. I'd guess that the globular protein has about 70-110 amino acids in it.


# 0.1 nm (nanometer) diameter of a hydrogen atom
# 0.8 nm Amino Acid
# 2 nm Diameter of a DNA Alpha helix
# 4 nm Globular Protein
# 6 nm microfilaments
# 10 nm thickness cell membranes
# 11 nm Ribosome
# 25 nm Microtubule
# 50 nm Nuclear pore
# 100 nm Large Virus
# 200 nm Centriole
# 200 nm (200 to 500 nm) Lysosomes
# 200 nm (200 to 500 nm) Peroxisomes
# 1 um (micrometer)
# (1 - 10 um) the general sizes for Prokaryotes
# 1 um Diameter of human nerve cell process
# 2 um E.coli - a bacterium
# 3 um Mitochondrion
# 5 um length of chloroplast
# 6 um (3 - 10 micrometers) the Nucleus
# 9 um Human red blood cell
# 10 um
# (10 - 30 um) Most Eukaryotic animal cells
(Source http://ccgb.umn.edu/~mwd/cell_www/cell_intro.html)

And I think the smallest protein I've ever heard of is 8 amino acids long (it's involved in labour and delivery so it has to be produced FAST)

---

Questions? Try the Wiki!
Take a look inside my brain

I did some poking about, and here are some sizes from hydrogen to a eukaryotic cell, to give ideas of sizes... amino acids and simple globular proteins are in bold. I'd guess that the globular protein has about 70-110 amino acids in it.

...And I think the smallest protein I've ever heard of is 8 amino acids long (it's involved in labour and delivery so it has to be produced FAST)

VERY good info. Perhaps I can find some way to put this in the FAQs (with credit to you for your research of course). This puts the size of what the project is dealing with in human terms.

---

Moderator9
ROSETTA@home FAQ
Moderator Contact

I did some poking about, and here are some sizes from hydrogen to a eukaryotic cell, to give ideas of sizes... amino acids and simple globular proteins are in bold. I'd guess that the globular protein has about 70-110 amino acids in it.

...And I think the smallest protein I've ever heard of is 8 amino acids long (it's involved in labour and delivery so it has to be produced FAST)

VERY good info. Perhaps I can find some way to put this in the FAQs (with credit to you for your research of course). This puts the size of what the project is dealing with in human terms.

Thanks for the insight everyone. Getting ones head around the tiny scale of the models we're dealing with here and the huge impications of such small things is difficult when you don't deal with them every day. It makes it that much more interesting.

Bill

Well some of the more recent proteins seem large enough to put on the table in front of you to bend into shape with your bare hands. I have one here that looks like a three year old got loose with a crayon.

---

Moderator9
ROSETTA@home FAQ
Moderator Contact

I would guess that the size of a small protein in its native state must be something like 50 atoms across, so the size of a protein would be a few Nanometers or more (I am not a bio science guy either, I am sure the experts will be able to give you more accurate numbers).

a little bigger, you have to take into account the size/length of the bonds

I was speaking of the diameter of the folded, native state of proteins, not the length of the chain. It seems eberndl's list confirms my crude estimate. :-; BTW: I am amazed how small some prokaryotes seem to be, just hundreds of small proteins across - it must be terribly cramped in there...

---

Team betterhumans.com - discuss and celebrate the future - hoelder1in.org

3. More aggressive full atom sampling:

HBLR_1.0_xxxx_ROT_TRIALS_TRIE
The final stage Rosetta's folding strategy consists of fine movements that try to fit the protein pieces togeth iner atomic detail (the "fullatom" stage, often abbreviated FA). These simulations use David Baker's latest energy terms (the "HBLR_1.0" refers to the weight on long-range hydrogen bonding) using an aggressive minimization protocol ("rotamer trials") that is made efficient with a neat graph representation within rosetta (the "trie").

I think I am understanding that but, exactly, what does the type of WU that goes HBLR_1.0_xxxx_ROT_TRIALS_TRIE_CHECKPOINTS xxxxx checks for? To say the least the graphics of a single piece in full atom relax are intriguing and so are the relatively high number of decoys (models)and very little number of steps per model that are showing up?

TIA

[Newbie here explain in as simple language as humanly possible :) ]

---

I just found this web page that shows some graphical information about the amino acids that make up proteins and how they fit togeather, and thought some of you might be interested in the information. Foxfire and Safari users can display it full screen. Be certain to roll your mouse around a bit on the images for more information about what is being shown.

---

Moderator9
ROSETTA@home FAQ
Moderator Contact