Stuck on Refinement Starting

7 views
Skip to first unread message

Sherik726

unread,
Jul 7, 2020, 8:01:45 PM7/7/20
to BioXTAS RAW
Hello,

I am currently attempting to create 3D reconstructions with bead models using DAMMIF/N and DAMAVER in RAW. After creating the reconstructions, the refinement process does not start (even though in the status bar it says "Starting Refinement". In the Refinement tab, the page is clear. I have been waiting for an hour or so. Is this a bug or should I be waiting longer?

Thank you,
Mustafa


Jesse Hopkins

unread,
Jul 7, 2020, 8:07:36 PM7/7/20
to bioxt...@googlegroups.com
Hi Mustafa,

Refinement should start immediately after damaver finishes, though it can take quite a while to complete.

What version of RAW are you using, and on what platform?

All the best.

- Jesse

----
Jesse Hopkins, PhD
Beamline Scientist
BioCAT, Sector 18
Advanced Photon Source

--
You received this message because you are subscribed to the Google Groups "BioXTAS RAW" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bioxtas_raw...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bioxtas_raw/5c77d9f7-54d1-4ffb-a1a4-e2a41f5a33b5o%40googlegroups.com.

Sherik726

unread,
Jul 7, 2020, 8:12:03 PM7/7/20
to BioXTAS RAW
Hello Jesse,

 I am running RAW 2.0.1 on my Macbook Air (MacOS Catalina Version 10.15.5). Perhaps I should try it on a Windows computer?

Best,
Mustafa


Jesse Hopkins

unread,
Jul 7, 2020, 8:20:34 PM7/7/20
to bioxt...@googlegroups.com
Mustafa,

What version of ATSAS are you using? It could have something to do with that.

Refinement should work on any platform, I’ve run refinements with 2.0.1 on 10.15.5 (I just started one now to double check). You can try on Windows if you like, it would be another data point.

It would be helpful if you could send me the .out file you’re using, it might be something particular to that dataset. I can test it on my own machine and see what happens.

Also, I’ve attached a standard .out file from the RAW tutorial data. Try running 3 dammif reconstructions in fast mode, and then doing damaver and refinement with this. It should only take a few minutes to get to the refinement step. If this runs okay, it’s either something with your settings or your dataset.
lysozyme.out

Sherik726

unread,
Jul 7, 2020, 8:31:50 PM7/7/20
to BioXTAS RAW


Hello,

I am using ATSAS 3.0.1. I will install the most recent version of ATSAS and try it out.  

Attached below is the out. file. Also, I will try using the lysozyme data you just sent me.

Thank you for your help

Regards,
Mustafa
GalB for Jesse.out

Sherik726

unread,
Jul 7, 2020, 8:44:38 PM7/7/20
to BioXTAS RAW

Hello Jesse,

I just tried the file you sent me and it works. There must be something wrong with my data set then. Let me know if my data set worked on your system or not.

Best,
Mustafa

Sherik726

unread,
Jul 7, 2020, 9:18:08 PM7/7/20
to BioXTAS RAW
Hello Jesse,

I tried another data set and refinement still did not work. I will review my settings to make sure they are correct.

Regards,
Mustafa 

Jesse Hopkins

unread,
Jul 7, 2020, 9:36:11 PM7/7/20
to bioxt...@googlegroups.com
Hi Mustafa,

I’ll test the dataset you sent me tomorrow.

It would be good to know what your dammif/n settings are (number of reconstructions, mode, symmetry, etc), so I can test under the same conditions.

All the best.

- Jesse

----
Jesse Hopkins, PhD
Beamline Scientist
BioCAT, Sector 18
Advanced Photon Source
--
You received this message because you are subscribed to the Google Groups "BioXTAS RAW" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bioxtas_raw...@googlegroups.com.

Sherik726

unread,
Jul 7, 2020, 9:42:12 PM7/7/20
to BioXTAS RAW
Hello Jesse,

No rush at all and thank you for your help so far. My settings are below.

Number of Reconstructions: 15
Number of simultaneous runs: 3
Use: DAMMIF 
Mode: Slow
Symmetry: P1
Anisometry: Unknown
Align and average envelopes (damaver): YES
Refine average with dammin: YES
Align and cluster envelope (damclust): YES
Align output to PDB: NO

Thank you in advance for trying this out for me on your system. I really appreciate it.

Best,
Mustafa

Jesse Hopkins

unread,
Jul 8, 2020, 11:26:32 AM7/8/20
to bioxt...@googlegroups.com
Hi Mustafa,

I'm testing your dataset now, I'll let you know what I figure out regarding the problem you reported.

 I did want to note that I have a couple of concerns about the data. First, your Dmax is too small, you're artificially truncating the P(r) function, which you can see in how it's forced to zero. I'd put the Dmax closer to 125-130. Second, I'm concerned there might be a bit of aggregation (and that you've already cut some off in the low q, I don't have access to the original data, just what's in the .out file). Third, the ambiguity (as estimated by ambimeter) is quite high, indicating that even if you do have a good P(r) function, your reconstructions may not be trustworthy.

I've been working on some documentation on best practices for P(r) functions and bead model reconstructions that I'd recommend you read:

All the best.

- Jesse

----
Jesse Hopkins, PhD
Beamline Scientist
BioCAT, Sector 18
Advanced Photon Source

--
You received this message because you are subscribed to the Google Groups "BioXTAS RAW" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bioxtas_raw...@googlegroups.com.

Jesse Hopkins

unread,
Jul 8, 2020, 12:01:30 PM7/8/20
to bioxt...@googlegroups.com
Hi Mustafa,

The problem is with your filename. If I give it the original filename (I'm assuming S_A_GalB_END_00657_1_data_000001.out), it gets truncated when running dammif (you should see a warning message pop up when you click start). There's a bug, which I thought I had previously fixed, where if the filename is truncated the refinement doesn't run correctly. Just change your output prefix in the DAMMIF window to something like 'GaIB' and you should be fine. I'll make sure this bug is (actually) fixed in the next release.

Let me know if you have questions about how to implement this, and whether or not it works for you.

All the best.

- Jesse

----
Jesse Hopkins, PhD
Beamline Scientist
BioCAT, Sector 18
Advanced Photon Source

Jesse Hopkins

unread,
Jul 9, 2020, 4:58:16 PM7/9/20
to bioxt...@googlegroups.com
Hi Mustafa,

I've just released a new version of RAW, 2.0.2, which fixes this issue.

All the best.

- Jesse

----
Jesse Hopkins, PhD
Beamline Scientist
BioCAT, Sector 18
Advanced Photon Source

Sherik726

unread,
Jul 9, 2020, 9:17:06 PM7/9/20
to BioXTAS RAW
Hello Jesse,

Thank you for your help and fixing the issue. It works now!

Cheers,
Mustafa
Reply all
Reply to author
Forward
0 new messages