[Ingeniosuccinimide]

I'm currently attending one medicinal chemistry course in which a very common exercise task is that a couple of molecules are given, e.g.:

http://en.wikipedia.org/wiki/Hydroxyzine

http://en.wikipedia.org/wiki/Loratadine

and I have to estimate the pK_a of a given drug, determine what will be the predominant form at physiological pH and whether the molecules will penetrate the BBB.

What is the "theoretical" way of estimating the pK_a for organic compounds? Does it involve some kind of structural contribution factors and their addition/subtraction, or solely the groups that can be dissociated are taken into consideration and other groups are omitted? How would I estimate the pK_a for the compounds mentioned above?

[AromaticAcrobatic]

When I took Organic, I never had to look up Pka values. What helped me was understanding the more a molecule can hold the charge, that is the more resonance opportunities a molecule has with the charge, the more acidic it is. In example, compare the acidity of phenol with ethanol. There is also some rules you just have to remember such as with alcohols. So if I was given this task I would first look for the acidic areas within the molecule and then think about resonance opportunities that are available for that position within the molecule. Next I would think about the position in question and see if it maybe resembles a structure that already has a Pka given to it and then +/- Pka according to its resonance opportunities.
And lastly, if none of these suggestions work I'm sure we can make some theoretical calculations to come up with an estimated Pka.
Looking at hydroxyzine, I see some acidic positions that resemble molecules that already have a Pka assigned to them, as well as, a basic center that I'm pretty sure you can find the Pka or at least find a Pka that is similar to the basic area.
Hope this helps in your quest for medicinal enlightenment!
 

[Ingeniosuccinimide]

Thanks for the answer, but if someone could help me by practically solving e.g. hydroxyzine, it would be slightly clearer. The task that is given states: "predict % of ionization of given compounds at physiological pH".

[kriggy]

I think you need to look at the structure and compare with other structures.
If I take Loratidine as an example then I would suggest that its slightly more basic than pyridine. The carboxy group is in form of ester so no acidity here and the tertiary amine is IMO much less basic than pyridine due to steric efects and conjugation with carboxy group. Pyridine has 2 alklyl substituents which increase the electron density so thats why I think its gonna be more basic that pyridine.

For hydroxyzine I would expect similar basicity to 1,4-diazabicyclo[2.2.2]octane given the structure. The hydroxy group can disociate too but but I think It would require very basic conditions

[Ingeniosuccinimide]

I think you need to look at the structure and compare with other structures.

That's something I was considering.

I'm wondering if there is some strictly empirical method that adds or subtracts the contribution of individual groups/fragments (analogous to the methods of approx. estimation of logP where fragments and moieties are taken into account)?