[Chimera-users] setting solvent density threshold at 2.5 times of the bulk solvent density

[Tom Goddard]

Hi Mohan,

  I think your occupancy map histogram has a peak at 0 and peak at 37 is explained as follows.  The peak near means there are many grid points in the volume that water never reaches.  This must be the volume occupied by the protein.  The peak at 37 are grid points outside the protein where usually 37 waters are found per grid point over the course of your MD frames.  So I think you are correctly interpreting the values.  Using a threshold 2.5 * 37 would show grid points where waters are found 2.5 times more often than average — you might think of those locations as “bound waters” — I’d expect them to be all at points adjacent to the protein.

  One extra bit of info about the histogram, the horizontal axis is the occupancy value (0-354 for your data) and the vertical axis is the number of grid points having a specific occupancy value.  The horizontal axis is linear, but the vertical axis is logarithmically scaled.  So on the vertical axis a slightly higher peak represents many times more grid points.

	Tom


> On Sep 10, 2015, at 9:50 AM, Elaine Meng <meng at cgl.ucsf.edu> wrote:
> 
> Dear Mohan,
> Yes, that is a very different question now that we know you calculated the map from MD.
> 
> Everything depends on how you calculated the “density" to produce the input map. Rather than an actual density, it may just be counts of how many times there was a water inside each grid cell (such as if you used the occupancy analysis in the MD Movie tool).  If so, you could divide those values by the number of MD frames that the counts came from, to make them more meaningful; for example, a count of 40 for a sample of 100 MD frames means that a grid cell had a water in it 40% of the time. MD Movie occupancy analysis:
> <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/movie/movie.html#occupancy>
> 
> Volume Viewer is just displaying this raw data, it isn’t doing anything to transform or normalize it. The data is just a grid of points with associated values.  The histogram shows the range of values horizontally, and the bar heights show how many grid points have those values.  It is not useful to show the contour surface for values near zero, and I would not read too much into the shape of the histogram.  The high value near zero just means many grid points have values of zero, and the other maximum just means many grid points have values of 37.  If you use a different size grid cell and/or sampled a different number of MD frames (if the values are simply counts), the maximum would be a different value.
> 
> I hope this helps,
> 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 Sep 10, 2015, at 9:20 AM, Mohan Pradhan <mohan.pradhan at ymail.com> wrote:
>> 
>> Dear Tom and Elaine,
>>                  Thank you both for your response. I think I was not very clear with my question. Let me rephrase my query here again in more detail.
>> 
>> I have done Molcular Dynamics simulation of my protein in explicit solvent. Following which I calculated the water density from the trajectory using grid-based approach with grid cells of size 0.5 X 0.5 X 0.5 Å. A sample of the grid-based density output file is attached here for your reference.
>> 
>> We then used Chimera to visualize the map using the volume viewer menu. We are attaching two image files here at two different isocountor threshold cutoff. One which was set to the isovalue at which we observe the maximum value of the historgram at an isovalue of 37. (image-1)
>> 
>> Second we set it to 2.5 times that isovalue 2.5 X 37 ~ 93) (Image-2).
>> 
>> The issue we wanted to clarify was what do these histogram actually depict with regard to the input grid file?
>> 
>> What is the histogram depicting at an iso-value of zero?
>> 
>> Why is there a sinusodial curve for the histogram towards the lower isovlaue?
>> 
>> Thanks,
>> Mohan
>> 
>> 
> 
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
>