Binding pocket surface command script not working
Hello,
I'm trying to reproduce the Binding Pocket Surface image from the Gallery with a different pdb file.
I can successfully reproduce with the pdb used in the example (2zcp), but not with 2v5z. Running latest Chimera.
Steps to reproduce:
open 2v5z delete :.b split atoms :sag surface measure contactArea #0.2 #0.1 3 slab -.1,0 color tan
gives an error:
error: Array must be 2-dimensional, got 1-dimensional
File "C:\Program Files\Chimera 1.8.1\share\MeasureVolume__init__.py", line 60, in surface_area area = surface_area(varray, tarray)
See reply log for Python traceback.
Why, and would it be helpful to file a bug report?
Thanks and Cheers Markus
-- Markus Heller, Ph.D. NMR Scientist CDRD - The Centre for Drug Research and Development 2405 Wesbrook Mall, Fourth Floor | Vancouver, BC V6T 1Z3 | Main: (604) 827-1147 Direct: (604) 827-1122 | F: (604) 827-1299 | E: mheller@cdrd.camailto:mheller@cdrd.ca | www.cdrd.ca [cid:image007.png@01CF49BD.0097D1F0]
Follow us: [Description: cid:image003.jpg@01CD86A6.39B41A10] https://twitter.com/C_D_R_D [Description: cid:image005.jpg@01CD86AE.B197F5D0] http://www.linkedin.com/groups/CDRD-3186016/about?report%2Esuccess=3NcE4iZo5QzagkO261M3aXe-B-TlARKBdjF7DAAAM-pSOa9_fAg67zA-NqOzVcb8I4zDMyK-MLWY5h9uI01j7OVI_QTVGWfuFtRDBJJle1wzFYTM5AvuRGUAMRWIpWKMI2g6aOg4NN_PXv5DAAP6RyF-BnszFYXufzF6aCzz7m6PKKGjfCN_NfYdaZsPILTDlgIuiVd-MLWqO-rDz4duiGFAR-wIFaXsWXUEldO [Description: cid:image009.png@01CD86AE.B197F5D0] https://www.facebook.com/pages/The-Centre-for-Drug-Research-and-Development-CDRD-Ventures-Inc/465976796791713?ref=hl [Description: cid:image010.png@01CD86AE.B197F5D0] http://www.youtube.com/user/CDRDCVI?view_as=public
This email and any files transmitted with it are confidential and intended solely for the addressee. If you are not the named addressee you should not disseminate, distribute, copy, or alter this email.
Hi Markus, It is a mystery to me… I tried it in both 1.8.1 and the current pre-1.9 daily build and didn't get the error. I'd say go ahead and submit a report… we can always close it if it's not reproducible.
However, the result I got was somewhat ugly because there were several holes in the pocket. I then tried a larger cutoff, as in the folowing:
open 2v5z delete :.b split atoms :sag surface measure contactArea #0.2 #0.1 4 slab -.1,0 color tan measure contactArea #0.2 #0.1 3.99 slab -.11,-.01 color cornflowerblue ~surf #0 ~disp; ~ribbon; disp :sag sop hideDust #1,2 size 1 metric sizeRank
which I think got rid of as many of the holes as reasonably possible, but there were still a couple of ugly places that might be surface calculation artifacts: where the blue shows through the tan near the middle of the attached image, and the little funnel on the upper right. There is a small opening through the funnel, so I'm not sure whether it's an artifact or the actual shape of the pocket. This general process would be expected to make a better-looking figure for pockets with fewer openings.
This pocket could alternatively be considered one large pocket containing both the SAG and FAD. For example, in a separate Chimera session I tried:
open castp:2v5z
which reports the two largest pockets as the SAG/FAD pockets of the A and B chains, respectively. If you are using the "measure contactArea" approach to get the appearance of different inside and outside pocket colors as shown in the image gallery figure, it could be done for this collective pocket with "split atoms :sag,fad" (well, assuming you didn't get that mysterious error again!). Sorry for the weirdness, Elaine ---------- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Mar 27, 2014, at 1:05 PM, Markus Heller mheller@cdrd.ca wrote:
Hello,
I’m trying to reproduce the Binding Pocket Surface image from the Gallery with a different pdb file.
I can successfully reproduce with the pdb used in the example (2zcp), but not with 2v5z. Running latest Chimera.
Steps to reproduce:
open 2v5z delete :.b split atoms :sag surface measure contactArea #0.2 #0.1 3 slab -.1,0 color tan
gives an error:
error: Array must be 2-dimensional, got 1-dimensional
File "C:\Program Files\Chimera 1.8.1\share\MeasureVolume__init__.py", line 60, in surface_area area = surface_area(varray, tarray)
See reply log for Python traceback.
Why, and would it be helpful to file a bug report?
Thanks and Cheers Markus
-- Markus Heller, Ph.D. NMR Scientist CDRD - The Centre for Drug Research and Development 2405 Wesbrook Mall, Fourth Floor | Vancouver, BC V6T 1Z3 | Main: (604) 827-1147 Direct: (604) 827-1122 | F: (604) 827-1299 | E: mheller@cdrd.ca | www.cdrd.ca <image007.png>
Follow us: <image003.jpg> <image004.jpg> <image005.png> <image006.png>
This email and any files transmitted with it are confidential and intended solely for the addressee. If you are not the named addressee you should not disseminate, distribute, copy, or alter this email.
Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
Hi Markus,
I did not get your error message when I tried your commands with Chimera 1.8.1 on Mac. But I bet you are not on Mac, and the surface calculation gave a different result that led to the error. I'd like to see a bug report so I can fix this. Use the "Report Bug" button that appears after the error occurs.
Tom
On Mar 27, 2014, at 2:39 PM, Elaine Meng wrote:
Hi Markus, It is a mystery to me… I tried it in both 1.8.1 and the current pre-1.9 daily build and didn't get the error. I'd say go ahead and submit a report… we can always close it if it's not reproducible.
However, the result I got was somewhat ugly because there were several holes in the pocket. I then tried a larger cutoff, as in the folowing:
open 2v5z delete :.b split atoms :sag surface measure contactArea #0.2 #0.1 4 slab -.1,0 color tan measure contactArea #0.2 #0.1 3.99 slab -.11,-.01 color cornflowerblue ~surf #0 ~disp; ~ribbon; disp :sag sop hideDust #1,2 size 1 metric sizeRank
which I think got rid of as many of the holes as reasonably possible, but there were still a couple of ugly places that might be surface calculation artifacts: where the blue shows through the tan near the middle of the attached image, and the little funnel on the upper right. There is a small opening through the funnel, so I'm not sure whether it's an artifact or the actual shape of the pocket. This general process would be expected to make a better-looking figure for pockets with fewer openings.
This pocket could alternatively be considered one large pocket containing both the SAG and FAD. For example, in a separate Chimera session I tried:
open castp:2v5z
which reports the two largest pockets as the SAG/FAD pockets of the A and B chains, respectively. If you are using the "measure contactArea" approach to get the appearance of different inside and outside pocket colors as shown in the image gallery figure, it could be done for this collective pocket with "split atoms :sag,fad" (well, assuming you didn't get that mysterious error again!). Sorry for the weirdness, Elaine
Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
<2v5zpock4.png>
On Mar 27, 2014, at 1:05 PM, Markus Heller wrote:
Hello,
I’m trying to reproduce the Binding Pocket Surface image from the Gallery with a different pdb file.
I can successfully reproduce with the pdb used in the example (2zcp), but not with 2v5z. Running latest Chimera.
Steps to reproduce:
open 2v5z delete :.b split atoms :sag surface measure contactArea #0.2 #0.1 3 slab -.1,0 color tan
gives an error:
error: Array must be 2-dimensional, got 1-dimensional
File "C:\Program Files\Chimera 1.8.1\share\MeasureVolume__init__.py", line 60, in surface_area area = surface_area(varray, tarray)
See reply log for Python traceback.
Why, and would it be helpful to file a bug report?
Thanks and Cheers Markus
-- Markus Heller, Ph.D. NMR Scientist CDRD - The Centre for Drug Research and Development 2405 Wesbrook Mall, Fourth Floor | Vancouver, BC V6T 1Z3 | Main: (604) 827-1147 Direct: (604) 827-1122 | F: (604) 827-1299 | E: mheller@cdrd.ca | www.cdrd.ca <image007.png>
Follow us: <image003.jpg> <image004.jpg> <image005.png> <image006.png>
This email and any files transmitted with it are confidential and intended solely for the addressee. If you are not the named addressee you should not disseminate, distribute, copy, or alter this email.
Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
Hi Elaine,
Thanks for your reply. I just filed a bug report on this, as per Tom's request.
I should also mention that I did get some result with your version, but not what I expected: essentially a couple of surface blobs, but nothing like your figure.
Anyhow, this seems to be a genuine bug, so I'll just hang tight and wait for a fix.
Cheers Markus
From: Elaine Meng [mailto:meng@cgl.ucsf.edu] Sent: Thursday, March 27, 2014 2:39 PM To: Markus Heller Cc: chimera-users@cgl.ucsf.edu Subject: Re: [Chimera-users] Binding pocket surface command script not working
Hi Markus, It is a mystery to me... I tried it in both 1.8.1 and the current pre-1.9 daily build and didn't get the error. I'd say go ahead and submit a report... we can always close it if it's not reproducible.
However, the result I got was somewhat ugly because there were several holes in the pocket. I then tried a larger cutoff, as in the folowing:
open 2v5z delete :.b split atoms :sag surface measure contactArea #0.2 #0.1 4 slab -.1,0 color tan measure contactArea #0.2 #0.1 3.99 slab -.11,-.01 color cornflowerblue ~surf #0 ~disp; ~ribbon; disp :sag sop hideDust #1,2 size 1 metric sizeRank
which I think got rid of as many of the holes as reasonably possible, but there were still a couple of ugly places that might be surface calculation artifacts: where the blue shows through the tan near the middle of the attached image, and the little funnel on the upper right. There is a small opening through the funnel, so I'm not sure whether it's an artifact or the actual shape of the pocket. This general process would be expected to make a better-looking figure for pockets with fewer openings.
This pocket could alternatively be considered one large pocket containing both the SAG and FAD. For example, in a separate Chimera session I tried:
open castp:2v5z
which reports the two largest pockets as the SAG/FAD pockets of the A and B chains, respectively. If you are using the "measure contactArea" approach to get the appearance of different inside and outside pocket colors as shown in the image gallery figure, it could be done for this collective pocket with "split atoms :sag,fad" (well, assuming you didn't get that mysterious error again!). Sorry for the weirdness, Elaine ---------- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
[cid:image001.png@01CF4A77.ED08BC10]
On Mar 27, 2014, at 1:05 PM, Markus Heller <mheller@cdrd.camailto:mheller@cdrd.ca> wrote: Hello,
I'm trying to reproduce the Binding Pocket Surface image from the Gallery with a different pdb file.
I can successfully reproduce with the pdb used in the example (2zcp), but not with 2v5z. Running latest Chimera.
Steps to reproduce:
open 2v5z delete :.b split atoms :sag surface measure contactArea #0.2 #0.1 3 slab -.1,0 color tan
gives an error:
error: Array must be 2-dimensional, got 1-dimensional
File "C:\Program Files\Chimera 1.8.1\share\MeasureVolume__init__.py", line 60, in surface_area area = surface_area(varray, tarray)
See reply log for Python traceback.
Why, and would it be helpful to file a bug report?
Thanks and Cheers Markus
-- Markus Heller, Ph.D. NMR Scientist CDRD - The Centre for Drug Research and Development 2405 Wesbrook Mall, Fourth Floor | Vancouver, BC V6T 1Z3 | Main: (604) 827-1147 Direct: (604) 827-1122 | F: (604) 827-1299 | E: mheller@cdrd.camailto:mheller@cdrd.ca | www.cdrd.cahttp://www.cdrd.ca <image007.png>
Follow us: <image003.jpg> <image004.jpg> <image005.png> <image006.png>
This email and any files transmitted with it are confidential and intended solely for the addressee. If you are not the named addressee you should not disseminate, distribute, copy, or alter this email.
_______________________________________________ Chimera-users mailing list Chimera-users@cgl.ucsf.edumailto:Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
Hi Markus,
This error occurs when the two specified surfaces have no points within the specified distance range. This is a bug -- it should just report the contact area as zero in that case. I've fixed that in tonight’s daily build.
http://plato.cgl.ucsf.edu/trac/chimera/ticket/12966
But there is a second problem, since your commands work on Mac with 64-bit Chimera 1.8.1. The trouble is almost surely that the surface calculation for one of the two sets of atoms failed on 64-bit Windows. Unfortunately the 3rd part surface calculation code we use (MSMS) often fails. Some ideas about how to try to get surfaces are here
http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/surfprobs.html
although often nothing works. The surface calculation fails especially often in 64-bit Windows Chimera. You might try 32-bit Windows Chimera. Our next generation Chimera will have surface calculation that always works, but is not expected out for at least 6 months.
Tom
On Mar 28, 2014, at 11:23 AM, Markus Heller mheller@cdrd.ca wrote:
Hi Elaine,
Thanks for your reply. I just filed a bug report on this, as per Tom’s request.
I should also mention that I did get some result with your version, but not what I expected: essentially a couple of surface blobs, but nothing like your figure.
Anyhow, this seems to be a genuine bug, so I’ll just hang tight and wait for a fix.
Cheers Markus
From: Elaine Meng [mailto:meng@cgl.ucsf.edu] Sent: Thursday, March 27, 2014 2:39 PM To: Markus Heller Cc: chimera-users@cgl.ucsf.edu Subject: Re: [Chimera-users] Binding pocket surface command script not working
Hi Markus, It is a mystery to me… I tried it in both 1.8.1 and the current pre-1.9 daily build and didn't get the error. I'd say go ahead and submit a report… we can always close it if it's not reproducible.
However, the result I got was somewhat ugly because there were several holes in the pocket. I then tried a larger cutoff, as in the folowing:
open 2v5z delete :.b split atoms :sag surface measure contactArea #0.2 #0.1 4 slab -.1,0 color tan measure contactArea #0.2 #0.1 3.99 slab -.11,-.01 color cornflowerblue ~surf #0 ~disp; ~ribbon; disp :sag sop hideDust #1,2 size 1 metric sizeRank
which I think got rid of as many of the holes as reasonably possible, but there were still a couple of ugly places that might be surface calculation artifacts: where the blue shows through the tan near the middle of the attached image, and the little funnel on the upper right. There is a small opening through the funnel, so I'm not sure whether it's an artifact or the actual shape of the pocket. This general process would be expected to make a better-looking figure for pockets with fewer openings.
This pocket could alternatively be considered one large pocket containing both the SAG and FAD. For example, in a separate Chimera session I tried:
open castp:2v5z
which reports the two largest pockets as the SAG/FAD pockets of the A and B chains, respectively. If you are using the "measure contactArea" approach to get the appearance of different inside and outside pocket colors as shown in the image gallery figure, it could be done for this collective pocket with "split atoms :sag,fad" (well, assuming you didn't get that mysterious error again!). Sorry for the weirdness, Elaine
Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
<image001.png>
On Mar 27, 2014, at 1:05 PM, Markus Heller mheller@cdrd.ca wrote:
Hello,
I’m trying to reproduce the Binding Pocket Surface image from the Gallery with a different pdb file.
I can successfully reproduce with the pdb used in the example (2zcp), but not with 2v5z. Running latest Chimera.
Steps to reproduce:
open 2v5z delete :.b split atoms :sag surface measure contactArea #0.2 #0.1 3 slab -.1,0 color tan
gives an error:
error: Array must be 2-dimensional, got 1-dimensional
File "C:\Program Files\Chimera 1.8.1\share\MeasureVolume__init__.py", line 60, in surface_area area = surface_area(varray, tarray)
See reply log for Python traceback.
Why, and would it be helpful to file a bug report?
Thanks and Cheers Markus
-- Markus Heller, Ph.D. NMR Scientist CDRD - The Centre for Drug Research and Development 2405 Wesbrook Mall, Fourth Floor | Vancouver, BC V6T 1Z3 | Main: (604) 827-1147 Direct: (604) 827-1122 | F: (604) 827-1299 | E: mheller@cdrd.ca | www.cdrd.ca <image007.png>
Follow us: <image003.jpg> <image004.jpg> <image005.png> <image006.png>
This email and any files transmitted with it are confidential and intended solely for the addressee. If you are not the named addressee you should not disseminate, distribute, copy, or alter this email.
Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
participants (3)
-
Elaine Meng -
Markus Heller -
Tom Goddard