Hepatoprotector Effect and Relationship Between Structure and Detoxicating Activity of Adamantane Derivatives

Relationship Between Structure and Hepatoprotector Activity of Indole Derivatives

“…The drugs were introduced with a starch jelly via a gastric tube 1 h after hepatotoxin injection. The bile-expelling activity was expressed in terms of the cholate -cholesterol index (CCI) calculated as the ratio of cholate concentration to that of cholesterol in the bile samples [2].…”

Relationship Between Structure and Hepatoprotector Activity of Adamantane Derivatives. Part 2. Application of Autocorrelative, Substructural, and 3D Molecular Descriptors

“…From the standpoint of predictions based on each particular class of descriptors used separately, 2D descriptors have proved to be generally more effective than 3D ones. It should also be noted that autocorrelative topological descriptors based on topological characteristics of the vertices of a molecular graph, which have been proposed and studied for the first tine, lead to the linear discriminant functions whose predicting ability (87 -89%) is much higher than that (70 -80%) of the discriminant functions calculated for the best combinations of usual topological descriptors [1]. The maximum predicting ability (92 -96%) with respect to the hepatoprotector activity of adamantane derivatives was provided by the discriminant functions obtained for the combinations of descriptors belonging to different groups: 2D, 3D, or combined 2D/3D descriptors).…”

“…Previously, we reported [1] on the experimental estimates of hepatoprotector activity for 30 adamantane derivatives selected using a three-stage primary screening system. The screening was based on classification rules determined by means of linear discriminant analysis (LDA).…”

“…In this study, we used a learning set of 30 compounds employed in [1], including 12 active compounds (i.e., substances producing a hepatoprotective detoxicating action and ensuring a percentage survival of experimental animals not less than 50%) and 18 inactive compounds of the same class, for constructing new classification rules based on molecular descriptors of the following types: (i) autocorrelative, (ii) substructural (fragmentary), (iii) geometric (topographic) [2 -5], (iv) GETAWAY [6,7], and (v) WHIM [8 -10]. The last three types of descriptors (calculated using the DRAGON program package [11]) represent the so-called 3D descriptors treating molecules as three-dimensional objects.…”

“…The last three types of descriptors (calculated using the DRAGON program package [11]) represent the so-called 3D descriptors treating molecules as three-dimensional objects. The structural formulas of compounds constituting the learning set are presented in [1].…”