I have 42 wus pending for validation, but server status says that there are only 2 pending validation...

I have 42 wus pending for validation, but server status says that there are only 2 pending validation...

Have you checked to see if they are still waiting for another computer to complete the WU ?
These tasks are minimum quorum 2 now.

Have you checked to see if they are still waiting for another computer to complete the WU ?
These tasks are minimum quorum 2 now.

Yes, i have a lot of wingmans who have not complete the wus

I wonder if this is only temporary, because everything appeared to be OK before when there was no second validation required?

I am seeing credits coming through.

Fistly, the server status webpage is not in real time (1 refresh every few hours).

Secondly, "pending for validation" on this page means validator was not executed on a task. But validation have multiple steps and sometimes results are different and need to be checked/compared multiple times, so not a very accurate number...
From the server side, there is a lot of pending for validation... 16671 at the moment ! a slight deviation from the server status page...

I have 42 wus pending for validation...

Now 67.
Where are my wingmans?

Now 67.

You are in luck. I have:
Validation pending (606) · Validation inconclusive (1) · Valid (11167) · Invalid (1107)

I expect that the "Invalids" will be checked again, since it appears that all of the other (usually multiple) wingmen also so invalid.
That is, I don't see any "valids" in that group.

It looks like it will take time to tweak the validator.

I tweak the validator on H2O and it's clearly too strict.
Now, I lowered the threshold to make validation more permissive and I'll launch a re-validate on all WU.

I'll launch a re-validate on all WU.

To be clear: for me the problem is not points, it's that without wus validated there is no science.

To be clear: for me the problem is not points, it's that without wus validated there is no science.

And without validation, there is no data for science ;)
By the way, the validation problem raises interesting scientific questions.

...the validation problem raises interesting scientific questions.

Do tell.

---

The validation process is based on a comparison on results (energy, conformation and so on).
The thing is that since we generate new and unknown molecules, we see that the optimized geometry can vary depending on the machine that made the calculation.
We need to find the correct parameters to discriminate "similar" results that do correspond to the same chemical object but that are not identical results and false results (cheated ones that can mimic a real result)...

I know with proteins that there can be many amino acid substitutions that can made with minimal change in the energy of the combination, e.g. a cell receptor docking with a signalling molecule. I don't have a conception of what you're doing as a long list of possibilities for these databases was listed in the abstract. I would think that a cohort of candidates would be found requiring close human inspection.

---

The validation process is based on a comparison on results (energy, conformation and so on).
The thing is that since we generate new and unknown molecules, we see that the optimized geometry can vary depending on the machine that made the calculation.
We need to find the correct parameters to discriminate "similar" results that do correspond to the same chemical object but that are not identical results and false results (cheated ones that can mimic a real result)...

Perhaps you've already done this but It might be a good idea to pick 2 or 3 'in-house" computers and run one of your unknown molecules 10 or more times on each computer. The results may help in selecting validation parameters.

Also, do any of your molecules have chiral centers or rotatable bonds. If so, multiple valid conformations for the same molecule could be possible.

They explain there how to solve the discrepancies made by different computers.
The idea is to sort volunteer's computers by types or by classes to do the comparison result for the validation.
They give a wcg example .
May it help you.

I am seeing WUs that are marked as a validate error, but my wingman and I each received credit for them. I've never seen that before in any project.
Is this part of doing a re-validate on all the WUs?

I was a little surprised too, but it seems so.

Now 67.

Now 107.
Seem that the situation has not an easy solution (to find correct parametes of simulations).

I have some wus pending since 5 of february.
I'm losing hope :-P