I am trying to estimate the tissue-plasma partition coefficient for liposomes but I do not have any in vitro data. I only have physicochemical data like Log D, pKa, etc. But I don't have any properties like fraction unbound to plasma.

I have been looking into predictive models for the tissue-plasma partition coefficients (Poulin and Theil 2002) and they usually use fraction unbound to plasma. Now, I am currently assuming that the liposomes, given that they are pretty big, distribute mainly into the extracellular space (I am not sure if this is a fair assumption). The equation for non-adipose tissue is:

Pt:p=(Veist/Veisp)×(fup/fut)

Please read the publication for more information. Finally, it comes down to only needing the fup (fraction unbound to plasma). This value I am estimating to be about 0.2 or 0.3 since lipopihlic and acidic compounds have much greater protein binding (Ghafourian and Amin 2013). 

Is there a better and more accurate way of estimating the partition coefficients given only physicochemical data of the drug? Also, how can I add some size dependence to my calculation of the partition coefficient? After all, the size of the liposome will greatly affect the tissue-plasma distribution in the organs.

Article Prediction of Pharmacokinetics Prior to In Vivo Studies. 1. ...

Article QSAR Models for the Prediction of Plasma Protein Binding