Dear all,
I am trying to construct a model with four HIS binding to a copper. I set the distance constraints for ND1--CU as 2.0 angstrom, but the distances between ND1 and CU of the constructed models are around 2.5 angstrom, it seems that the constraints failed. Acctually, when I checked the model, I found that the distances could be around 2.0 angstrom if CU was just moved a little bit.
rsr.add(forms.gaussian(group=physical.xy_distance, feature=features.distance(at['ND1:13:X'],at['CU:43:Y']), mean=2.00, stdev=0.05)) rsr.add(forms.gaussian(group=physical.xy_distance, feature=features.distance(at['ND1:14:X'],at['CU:43:Y']), mean=2.00, stdev=0.05))
Could someone tell me what kinds of reasons could result in restraints failure? Thanks very much!
All the best, Qinghua
Dear all,
Finally I solve this problem after I set the stdev as 0.005. I thought that stdev means standard deviation, there might be some other meanings, right? I will check the manual again.
All the best, Qinghua
-------- Original Message -------- Subject: constraints failed Date: Mon, 14 Oct 2013 12:19:05 +0200 From: fantasticqhl fantasticqhl@gmail.com To: modeller_usage@salilab.org
Dear all,
I am trying to construct a model with four HIS binding to a copper. I set the distance constraints for ND1--CU as 2.0 angstrom, but the distances between ND1 and CU of the constructed models are around 2.5 angstrom, it seems that the constraints failed. Acctually, when I checked the model, I found that the distances could be around 2.0 angstrom if CU was just moved a little bit.
rsr.add(forms.gaussian(group=physical.xy_distance, feature=features.distance(at['ND1:13:X'],at['CU:43:Y']), mean=2.00, stdev=0.05)) rsr.add(forms.gaussian(group=physical.xy_distance, feature=features.distance(at['ND1:14:X'],at['CU:43:Y']), mean=2.00, stdev=0.05))
Could someone tell me what kinds of reasons could result in restraints failure? Thanks very much!
All the best, Qinghua
On 10/14/13 3:19 AM, fantasticqhl wrote: > Could someone tell me what kinds of reasons could result in restraints > failure? Thanks very much!
Modeller tries to satisfy all restraints. This includes both your restraints and the default restraints that are extracted from the alignment and forcefield. So generally a failure such as you describe is the result of a conflict - Modeller is unable to satisfy both your restraints and some other set of restraints at the same time.
Ben Webb, Modeller Caretaker
On 10/14/13 3:42 AM, fantasticqhl wrote: > Finally I solve this problem after I set the stdev as 0.005. I thought > that stdev means standard deviation
That's exactly what it means. The smaller the standard deviation of the probability distribution, the stronger the restraint. So by making the standard deviation smaller, you make your restraint stronger, so that it can overwhelm other restraints that it may be conflicting with.
Ben Webb, Modeller Caretaker
-
fantasticqhl -
Modeller Caretaker