The system:
SMIREP is a system for predicting the structural activity of chemical compounds. For that it can be categorized as a SAR/QSAR tool. The advantages of SMIREP are that it is fairly fast, due to a heuristic approach of identifying the main features of active versus inactive compounds. The system is written in Python and makes use of the OpenBabel library. In its current version it still uses an old incarnation of OpenBabel. This page provides all necessary files to compile and run the system on a Linux machine. I have not even tried to make it work on a Windows PC.
WARNING: The system is now quite old and is based on a very old version of openbabel. If there is interest in an update, please write to me an tell inform me, that there is some interest in this.
Current Version:
The program:
The OBGrep module for python and openbabel 1.100.2 (now, very old!):
The data used in our publication:
In case you need Python:
Go to and get the latest release.
Installation Instructions:
Download the files listed above. You need a working version of Python (2.7). After downloading, unzip the OpenBabel tar files (tar -xvfz openbabel-1.100.2.tar.gz) and compile the library (use the –prefix=”XXX” option to install it to your preferred system directory called XXX). After this you will need to install the python library OBGrep. For this extract the gzipped tar file as before, and go to the OBGrep subdirectory. The command “python install”  should compile the library and install it to the python lib directory. After this OpenBabel specific installation you can simply unpack the SMIREP program and data files, descend into the subdirectory SMIREP_V1.0 and read the instructions (README file).
Known Issues:
The QSAR version is slower than the SAR version. This is also mentioned in the arcticle.



SMIREP: Predicting Chemical Activity from SMILES (Journal Article)

Journal of Chemical Information and Modeling, 46 (6), pp. 2432 – 2444, 2006.

(Abstract | Links | BibTeX | Tags: cheminformatics, graph mining, machine learning, QSAR, relational learning, scientific knowledge)

Karwath, Andreas; De Raedt, Luc

Predictive Graph Mining (Inproceeding)

The 7th International Conference of Discovery Science, DS 2004, pp. 1-15, Springer-Verlag Berlin Heidelberg Springer Verlag, Berlin Heidelberg, Germany, 2004, ISBN: 978-3-540-23357-2.

(Abstract | Links | BibTeX | Tags: cheminformatics, graph mining, machine learning, QSAR)

a AT :    A. Karwath