Re: [DIYbio] Protein Categorization

Kind of both, look at it as trying to come up with a globally unique identifier for proteins in various conditions without the risk of collision beyond something really low like 0.0001%.  I'm not aiming for perfection, if on average the same exact folded protein is derived from 500 GUIDs after months of intensive number crunching, meaning 500 duplicates on average, then that's fine, the objective is to limit the number of GUIDs a protein may cover to the minimum practical without collisions (e.g. 1 GUID being able to identify two different thermodynamically stable configurations should not happen more than 0.0001% of the time, to reuse the same arbitrary percentage.)

On Friday, December 8, 2017 at 11:20:53 AM UTC-5, John Griessen wrote:

On 12/08/2017 09:36 AM, Cory J. Geesaman wrote:
> I may have worded this poorly, I'm currently just looking for the core dimensions and their scale+span.

So you mean molecule dimensions in aggregate, like mol. weight vs. volume?  and how far that ranges over the conditions?

   If it hits 99.9%
> of protein conformations for a given protein within a bioengineering application it will pass the good enough test in my opinion.


--
John Griessen
industromatic.com  Austin TX  building lab gear for biologists