On Mon, Dec 02, 2013, Mahendra B Thapa wrote:
>
> I used the following input to calculate NMR relaxation rates (R1 and R2)
> of backbone N-H bonds of a protein using amber 12 with cpptraj :
>
> trajin test_file1.pdb
> vector v0 .15 ired .16
> vector v1 .107 ired .108
> ...............................................
> matrix ired name matired order 2
> diagmatrix matired vecs 72 out ired.vec name modes_name
> ired relax freq 500000000 NHdist 1.02 order 2 tstep 1.0 tcorr 4000.0
> out v0.out norm modes modes_name noefile noe_NH_test_Protein
> go
>
> I am interested to find these relaxation rates at 500MHz. Though R1 values
> are comparable to the experimental results but *R2 value is quite high for
> all vectors* , such as
>
> vector R1 R2 NOE
> 0 47.73492 22659.00760 1.00000
> 1 39.87131 29025.85386 1.00000
>
> I did 50ns all-atom simulation, so I choose tcorr as 4000.0 . Any
> suggestion and comment will help me.
I'll just point out that it is very common for R2 values calculated from a
simulation to be a lot different that experimental values. The overall
tumbling behavior (which affects R2 but not R1) is often badly predicted,
usually because of limitations of the water model and the water-protein
interactions.
This doesn't mean there isn't some other problem, but you might get insight
by computing the rotational diffusion tensor from your simulation, and
comparing it to experimental estimates from NMR, if they are available.
...good luck...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Dec 03 2013 - 19:00:03 PST
