No, both methods can be applied seperately

Thanks Mr Serdar Durdagi ,

What is the best method?

and what are the criteria to choose one of two methods?

CoMFA approach has been widely accepted and a useful technique in 3D-QSAR, however, in particular, both steric and electrostatic potential functions are very steep near the van der Waals surface of the molecule, causing rapid changes in surface descriptions, and they require the use of cut-off values so calculations are not done inside the molecular surface. In

addition, a scaling factor is applied to the steric field, so both fields can be used in the same PLS analysis.

for detailed explanations about these techniques you may check my PhD thesis from following link:

http://www.diss.fu-berlin.de/diss/receive/FUDISS_thesis_000000010355

but CoMFA does'nt take into account the H-bond and hydrophobic properties, whereas with CoMSIA we can have 5 properties. when I do analysis by both methods for a series of molecules I found the same results for CoMFA and CoMSIA with regard to the steric and electrostatic properties. But the good model it was found by CoMSIA and only by 3 properties (h-bond, hydrophibic and steric). in this case it is only that I limit myself to the results found by CoMSIA or I will include in my article the results of CoMFA even if it is not good.

thanks for sharing me your PhD thesis

Dear Akabli Taoufik,

As the comparative molecular field analysis (CoMFA), comparative molecular similarity indices analysis (CoMSIA) are totally different approaches, there are many difference between these two. Basically, CoMFA enables the calculation of steric (Lennard−Jonnes potentials) and electrostatic (Coulombic potentials) fields in 3D grid, whereas CoMSIA is use to evaluate the similarity indices between the compounds. Moreover, hydrophobic and hydrogen-bond donor and acceptor fields were derived using the CoMSIA method.

As per my opinion, you should include the results of CoMFA (even if it is not good) and CoMSIA studies. Further, you may interpret the contours of the best model in details. I have attached some references for your better undersatnding.

http://pubs.acs.org/doi/abs/10.1021/ci500216c

http://www.tandfonline.com/doi/full/10.1080/1062936X.2017.1302991

http://www.tandfonline.com/doi/full/10.1080/1062936X.2017.1406986

http://www.tandfonline.com/doi/full/10.1080/07391102.2017.1418679

https://link.springer.com/article/10.1007%2Fs11224-017-1028-6

https://www.sciencedirect.com/science/article/pii/S0022286017316253?via%3Dihub

Thanks