Virtual Computational Chemistry Laboratory

Software Groups of Indices Contributor

GSFRAG Descriptor block for the VCCLab Project

General information. The GSFRAG program allows one to calculate the occurrence numbers of certain special fragments on k=2,...,10 vertices in a molecular graph G that can be used as molecular descriptors in quantitative structure-property/activity studies.

Description of fragments. We consider molecular fragments consisting of one or more disconnected components. Each component is a path (of length 9 or less), a cycle (on 10 or less vertices), or a path (cycle) with a number of attached chains of unit length. We call such fragments "primary". Each fragment has unique name which is formed by the following rule. We denote a path of length n by symbol "pn" and a cycle on m vertices by "cm". We also label each atom of the path (cycle) by the Latin capital A, B, etc. If the path (cycle) has some attached chains, labels of the corresponding atoms are listed next. In this manner one may get the name of each component of a given fragment. The name of a complete basic fragment is then formed by concatenating the names of each individual component (primary fragment). This naming scheme is illustrated by the picture below.

Applications. Using the GSFRAG program many numerical experiments were done. It was established that occurrence numbers of considered fragments in a graph produce a unique code of chemical structure for many large classes of compounds. Model QSPR/QSAR equations constructed from these descriptors usually provide good statistical characteristics and high predictive ability.


  1. Skvortsova M.I., Baskin I.I., Skvortsov L.A., Palyulin V.A., Zefirov N.S., Stankevich I.V. Chemical graphs and their basis invariants. J. Mol. Structure (Theochem). 1999. 466 (1-3) 211-217.
  2. Skvortsova M.I., Fedyaev K.S., Baskin I.I., Palyulin V.A., Zefirov N.S. A New Technique for Coding Chemical Structures Based on Basis Fragments. Doklady Chemistry. 2002. 382 (4-6) 33-36. Translated from: Doklady Akademii Nauk (Russ.), 2002. 382 (5) 645-648.
See FAQ if you have questions. How to cite this applet? Are you looking for a new job in chemoinformatics?

Copyright 2001 -- 2023 All rights reserved.