- Posts: 5
insane.py rotate usage
- Helsell
-
Topic Author
- Offline
- Fresh Boarder
Less
More
10 years 1 month ago #3594
by Helsell
insane.py rotate usage was created by Helsell
Hey,
I'm a bit confused by the inputs for -rotate. How would I tell insane.py to rotate my protein 90 degrees?
On a seperate note, the basic topology file generated by insane.py defines the main chain as PROTEIN, but it appears PROTEIN is undefined in the martini .itp files I downloaded here. Am I missing something? Do I have to define the protein by amino acid composition for the topology to point to the correct forcefields?
Forgive my ignorance, I'm an extremely new GROMACS user and an even newer Martini user.
I'm a bit confused by the inputs for -rotate. How would I tell insane.py to rotate my protein 90 degrees?
On a seperate note, the basic topology file generated by insane.py defines the main chain as PROTEIN, but it appears PROTEIN is undefined in the martini .itp files I downloaded here. Am I missing something? Do I have to define the protein by amino acid composition for the topology to point to the correct forcefields?
Forgive my ignorance, I'm an extremely new GROMACS user and an even newer Martini user.
Please Log in or Create an account to join the conversation.
- Helsell
-
Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 5
10 years 1 month ago #3627
by Helsell
Replied by Helsell on topic insane.py rotate usage
I've been at the code for -rotate, but I still can't get it to return anything but "KeyError: 'myinput'". Has anyone had success with -rotate? The rest of the script seems to work beautifully, I just need a way to orient the protein perpendicular to the normal of the bilayer.
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
10 years 1 month ago - 10 years 1 month ago #3628
by Clement
Replied by Clement on topic insane.py rotate usage
Alright. I think what you want to do is -rotate princ (orient according the principal axes of the protein, works most of the time; otherwise you need to play with editconf). The flag -rotate takes only one argument, e.g. keywords "random" or "princ", or a number. The rotation is around the z axis, the axis arbitrarily chosen as perpendicular to the membrane by default.
To reply to your first question, your protein will be defined as "protein" in its topology (.itp). In your system topology (.top), you need to include this very topology file ("#include "protein.itp" "), and the protein will be recognized as it should.
To reply to your first question, your protein will be defined as "protein" in its topology (.itp). In your system topology (.top), you need to include this very topology file ("#include "protein.itp" "), and the protein will be recognized as it should.
Last edit: 10 years 1 month ago by Clement.
Please Log in or Create an account to join the conversation.
- Helsell
-
Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 5
10 years 1 month ago #3629
by Helsell
Replied by Helsell on topic insane.py rotate usage
Thank you, I appreciate the reply.
Unfortunately, I cannot get insane.py to accept any argument at all for -rotate, including random, princ, or any of the many numbers I've tried. To take the input you gave as an example, I get:
File "./insane.py", line 562, in <module>
options[ar].setvalue([args.pop(0) for i in range(options[ar].num)])
KeyError: 'princ'
But this error is what I'll get for any argument, just with a different KeyError specified. I'd love to just use insane, because excluding the orientation of the protein, it looks fine. But if I can't get this resolved, I guess it's back to editconf.
Unfortunately, I cannot get insane.py to accept any argument at all for -rotate, including random, princ, or any of the many numbers I've tried. To take the input you gave as an example, I get:
File "./insane.py", line 562, in <module>
options[ar].setvalue([args.pop(0) for i in range(options[ar].num)])
KeyError: 'princ'
But this error is what I'll get for any argument, just with a different KeyError specified. I'd love to just use insane, because excluding the orientation of the protein, it looks fine. But if I can't get this resolved, I guess it's back to editconf.
Please Log in or Create an account to join the conversation.
- Clement
-
- Offline
- Admin
Less
More
- Posts: 211
10 years 1 month ago - 10 years 1 month ago #3630
by Clement
Replied by Clement on topic insane.py rotate usage
What is the exact command line you type?
Which version of the script do you use?
Another thing you could try is the -orient flag (should be working the same as -rotate princ). I'll let know the maintainer of this script to check -rotate.
Which version of the script do you use?
Another thing you could try is the -orient flag (should be working the same as -rotate princ). I'll let know the maintainer of this script to check -rotate.
Last edit: 10 years 1 month ago by Clement.
Please Log in or Create an account to join the conversation.
- Helsell
-
Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 5
10 years 4 weeks ago #3661
by Helsell
Replied by Helsell on topic insane.py rotate usage
version = "20131112.22.TAW"
Here's the line I've been trying:
./insane.py -f sbh1_martini.pdb -pbc float -x 12 -y 12 -z 10 -l POPC:1 -sol WATER -center -orient -rotate 90.0 -o sbh1_POPC.gro -p sbh1_POPC.top
Works fine as long as I omit -rotate, though I've noticed it makes an asymmetrical bilayer. I don't really have a problem with that, though.
Thanks for your diligence, I've found the software here extremely useful so far!
Here's the line I've been trying:
./insane.py -f sbh1_martini.pdb -pbc float -x 12 -y 12 -z 10 -l POPC:1 -sol WATER -center -orient -rotate 90.0 -o sbh1_POPC.gro -p sbh1_POPC.top
Works fine as long as I omit -rotate, though I've noticed it makes an asymmetrical bilayer. I don't really have a problem with that, though.
Thanks for your diligence, I've found the software here extremely useful so far!
Please Log in or Create an account to join the conversation.
- Helsell
-
Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 5
10 years 2 weeks ago #3718
by Helsell
Replied by Helsell on topic insane.py rotate usage
Ok, so someone in my lab much more well-versed in Python than I read the code and made a change in the -orient option that fixed my problem. If anyone's interested, I could get the exact change from my peer, but he said code for -orient did a PCA that placed the longest axis of the protein in the XY plane. -rotate still doesn't seem to work, for one thing it's currently coded not to accept arguments:
("-rotate", Option(str, 0, None, "Rotate protein (random|princ|angle(float))")),
But changing that '0' to '1' didn't seem to fix the problem either.
("-rotate", Option(str, 0, None, "Rotate protein (random|princ|angle(float))")),
But changing that '0' to '1' didn't seem to fix the problem either.
Please Log in or Create an account to join the conversation.
- tsjerk
-
- Offline
- Expert Boarder
Less
More
- Posts: 103
10 years 1 week ago #3720
by tsjerk
Replied by tsjerk on topic insane.py rotate usage
The orientation functionality of insane is very rudimentary. The option -rotate should indeed take 1 argument and not 0, but then only performs a rotation in the plane. The option -orient does PCA, but in a special way: it tries to align the principal axes of the hydrophobic region in the XY plane. That's working in some cases and failing in other. The way it is now, insane is not really suitable for the orientation, but rather expects an oriented protein as input. It builds the membrane and the solvent.
If I can find the time, I will try to get the orientation in, but its not high on my priority list. I can probably extend the -rotate command to allow x/y/z rotations, but probably it's currently best to do the orientation before and pass the properly oriented molecule to insane.
I'm happy to hear the program is useful :) And I am certainly open for suggestions to improve it.
If I can find the time, I will try to get the orientation in, but its not high on my priority list. I can probably extend the -rotate command to allow x/y/z rotations, but probably it's currently best to do the orientation before and pass the properly oriented molecule to insane.
I'm happy to hear the program is useful :) And I am certainly open for suggestions to improve it.
Please Log in or Create an account to join the conversation.
- Hegedus
-
- Visitor
9 years 4 months ago #4214
by Hegedus
Replied by Hegedus on topic insane.py rotate usage
Hi,
This bug still exists - rotate should take 1 value not zero:
("-rotate", Option(str, 0, None, "Rotate protein (random|princ|angle(float))")),
("-rotate", Option(str, 1, None, "Rotate protein (random|princ|angle(float))")),
This bug still exists - rotate should take 1 value not zero:
("-rotate", Option(str, 0, None, "Rotate protein (random|princ|angle(float))")),
("-rotate", Option(str, 1, None, "Rotate protein (random|princ|angle(float))")),
Please Log in or Create an account to join the conversation.
Time to create page: 0.104 seconds