Processing and classification of chemical data inspired by insect olfaction

• Published in: Proceedings of the National Academy of Sciences - PNAS Vol. 104; no. 51; pp. 20285 - 20289
• Main Authors: Schmuker, Michael, Schneider, Gisbert
• Format: Journal Article
• Language: English
• Published: United States National Academy of Sciences 18.12.2007; National Acad Sciences
• Subjects: Animals; Antennal lobe; Architecture; Biochemistry; Bioinformatics; Biological Sciences; Computer Simulation; Dimensionality; Insecta - physiology; Insects; Ligands; Machine learning; Mathematical vectors; Molecules; Neural Networks (Computer); Neurons; Odors; Olfactory Receptor Neurons; Pharmaceutical Preparations - chemistry; Principal components analysis; Receptors; Sensory perception; Smell; Statistical median; Structure-Activity Relationship
• DOI: 10.1073/pnas.0705683104
• ISSN: 0027-8424

The chemical sense of insects has evolved to encode and classify odorants. Thus, the neural circuits in their olfactory system are likely to implement an efficient method for coding, processing, and classifying chemical information. Here, we describe a computational method to process molecular representations and classify molecules. The three-step approach mimics neurocomputational principles observed in olfactory systems. In the first step, the original stimulus space is sampled by "virtual receptors," which are chemotopically arranged by a self-organizing map. In the second step, the signals from the virtual receptors are decorrelated via correlation-based lateral inhibition. Finally, in the third step, olfactory scent perception is modeled by a machine learning classifier. We found that signal decorrelation during the second stage significantly increases the accuracy of odorant classification. Moreover, our results suggest that the proposed signal transform is capable of dimensionality reduction and is more robust against overdetermined representations than principal component scores. Our olfaction-inspired method was successfully applied to predicting bioactivities of pharmaceutically active compounds with high accuracy. It represents a way to efficiently connect chemical structure with biological activity space.