Noticed in recently completed WUs: Tree Jump Abrelax.....
and current WU (also Tree Jump)
that the plots for RMSD and Energy seem to be on overload (anything over approx 130 energy). Does this distort the usefulness of the WU?

---

one of these went off the scale at -113 in energy. I posted a question in problems for 5.68, but no answer yet. probably won't be one. I noticed this same problem with 1gidA WU's. The answer was for the 1gidA WU's which may be the same answer for what we are seeing now: "Hi: sorry, should have posted earlier. 1gid is probably one of the biggest molecules we've run on Rosetta@home, so we didn't anticipate the scaling of the energy and rmsd axes that would make sense for the molecule. As you pointed out, the RMSD can go crazy high, and that's upsetting the automatic (and not very smart) scaling inside Rosetta. Should be fixable! Though currently our priorities are to fix the FOLD_AND_DOCK_SUBSYSTEM workunits and some similar tasks..."

Noticed in recently completed WUs: Tree Jump Abrelax.....
and current WU (also Tree Jump)
that the plots for RMSD and Energy seem to be on overload (anything over approx 130 energy). Does this distort the usefulness of the WU?

---

not quite what i hoped i was getting at.
best analogy i can think of just now is that of radio wave becoming distorted if the power is too high for the system to handle

---

ahh ok...well leave that one for the experts if there are any out there

not quite what i hoped i was getting at.
best analogy i can think of just now is that of radio wave becoming distorted if the power is too high for the system to handle

---

The WUs are useful. The energy levels they find could be better, but that is part of how you learn what approaches to finding that energy level are working well, and which approaches do not work well. Or, perhaps which situations a given approach would be the method of choice, and which situations it would not be.

So, the italicized quote below is basically just saying that the graphic could be improved to better handle the scaling of the unit of measure, to show how the task is proceeding, but the operational aspects of those tasks are just fine.

---

Rosetta Moderator: Mod.Sense

when the WU is done and is returned back to the lab, what all do you guys see for information? do you see the graphing as well as the decoys or just the decoys?
if our graphing is off the "scale" for one particular WU how does that affect the data of the results of the decoys?

The WUs are useful. The energy levels they find could be better, but that is part of how you learn what approaches to finding that energy level are working well, and which approaches do not work well. Or, perhaps which situations a given approach would be the method of choice, and which situations it would not be.

So, the italicized quote below is basically just saying that the graphic could be improved to better handle the scaling of the unit of measure, to show how the task is proceeding, but the operational aspects of those tasks are just fine.

---

Sometime when you have a chance to tinker, increase your cache size and get enough work to keep your machine busy for a while. Then suspend network activity (activity tab). Then wait for one of your tasks to complete. Have a look in the slots directory and find it's .out file. That's what gets sent back.

If you'd like a visualization of what it contains, or path names of the slots directories, you can review the info. here about viewing your predictions:
https://boinc.bakerlab.org/rosetta/rah_view_predictions.php
Then you can download one of the PDB viewers mentioned and view the 3D structures you have created.

...don't forget, when you are through, enable network activity and set your cache back as desired (3D structures and cache back, what more could you ask for?).

---

Rosetta Moderator: Mod.Sense

Edit, sorry, I guess after the task actually completes, the .out file is moved and renamed. Do that for a task where you've completed a few models, but not finished it yet.

---

Rosetta Moderator: Mod.Sense