A molecular electronegativity distance vector based on 13 atomic types (MED
V-13), is a descriptor for predicting the biological activities of molecule
s based on the quantitative structure-activity relationship (QSAR). The MED
V-13 with 91 descriptors is employed to describe the structures of a series
of selective cyclooxygenase-2 (COX-2) inhibitors including 16 indomethacin
and its amide and ester derivatives (ImAE). A principal component regressi
on (PCR) is used to derive a QSAR model relating the biological activities
expressed by pIC(50) values to the MEDV-13. With the number of principal co
mponents of 6, the correlation coefficient (R) and the root mean square err
or (RUS) are 0.9245 and 0.1682 in modeling stage, and 0.8417 and 0.2389 in
leave-one-out prediction step, respectively.