I have recently docked molecules in AutoDock Vina and to my surprize Vina produces only binding affinity scores and nothing else. Hence any way to get Ki values from Vina scores (any additional program you can suggest) (please don't suggest to use formula for each value to derive the Ki value). Or is deriving from the formula Ki = exp (deltaG/ (R*T) the only way here.
as per my experience,i suggest you to use autodock instead of autodock vina as it gives you more parameters than autodock vina like inhibition constant Reference and cluster RMSD etc.
As far as I know predicting absolute Ki values is not so straightforward, since the conversion of score to Ki can heavily depend on the system (protein) you are dealing with. Why do you need Ki?
Hi Seshank,
Not sure why you don't want to use the formula, as this is exactly the way one would do such a thing (and what is taught in biochemistry classes). If you can't write code (and can't find someone to write you a short Python or Perl script) you could still put the formula into an Excel sheet and convert long lists of Vina scores (kcal/mol) to Ki (=Kd):
If you need the concrete formula to achieve this, I think this may be of help:
http://mgldev.scripps.edu/pipermail/autodock/2009-April/005619.html
HTH,
Rob
Robert Rentzsch thanku...can you tell me what would the values of R and T would be in this scenario( i mean docking)
T is basically room temperature (298.15 K = 25 C)
R is the gas constant (see 5th row of table here http://en.wikipedia.org/wiki/Gas_constant)
Both appear in the formula in the link I sent you above. You will probably find some basic biochemistry lectures on these topics on the web. Or ask some chemistry people here on RG.
All the best,
Rob
I agree with with Mrs. Schafferhans, it is impossible to predict/estimate Ki through docking. While it is true that Autodock & Autodock Vina can give you result in kcal/mol (and you can convert it to Ki) but there are many thing to consider. As you can see, the protein is static, the whole long range electrostatic is not taken into account, no water molecule involved, it is additive in nature (failed to take into account the potential energy cliff). All you can predict is whether it is active or not. I think to check the validity of a docking tool enrichment factor (EF) & ROC with the appropriate decoys would be much more appropriate.
Regards,
Radif
@muhammad radifar... plz tell me how to calculate or convert to ki (inhibition constant, in case of autodock vina)
You could use our DelG-to-Kd and IC50-to-pIC50 converter scripts:
http://shityakovlab.uphero.com/hardware-and-software.html
http://autodock.scripps.edu/faqs-help/faq/how-autodock-4-converts-binding-energy-kcal-mol-into-ki
See equation 1 in this paper.Article Structure Based Virtual Screening of the Ebola Virus Trimeri...
In autodock4 *.dlg files, there are Ki values but... but it does not have to do with experimental Ki value. Because, in real life experiment if ligand is super-active, it can bind to another protein and you can see in the experiment this ligand is not active. Also solution, catabolisma and many many things will affect your results.
Respected Sergey Shityakov
One of reviewer asked us what is standard deviation in Ki value. I wonder it may not possible here when we use a single Ki value for one type of ligand. We used your script in Ki calculations. Any comments please.
- Badges
- Science topic
- Similar topics
- Filing