Processing and classification of chemical data inspired by insect olfaction

Pattern orthogonalization via channel decorrelation by adaptive networks

The early processing of sensory information by neuronal circuits often includes a reshaping of activity patterns that may facilitate further processing in the brain. For instance, in the olfactory system the activity patterns that related odors evoke at the input of the olfactory bulb can be highly similar. Nevertheless, the corresponding activity patterns of the mitral cells, which represent the output of the olfactory bulb, can differ significantly from each other due to strong inhibition by granule cells and peri-glomerular cells. Motivated by these results we study simple adaptive inhibitory networks that aim to separate or even orthogonalize activity patterns representing similar stimuli. Since the animal experiences the different stimuli at different times it is difficult for the network to learn the connectivity based on their similarity; biologically it is more plausible that learning is driven by simultaneous correlations between the input channels. We investigate the connection between pattern orthogonalization and channel decorrelation and demonstrate that networks can achieve effective pattern orthogonalization through channel decorrelation if they simultaneously equalize their output levels. In feedforward networks biophysically plausible learning mechanisms fail, however, for even moderately similar input patterns. Recurrent networks do not have that limitation; they can orthogonalize the representations of highly similar input patterns. Even when they are optimized for linear neuronal dynamics they perform very well when the dynamics are nonlinear. These results provide insights into fundamental features of simplified inhibitory networks that may be relevant for pattern orthogonalization by neuronal circuits in general.

Fast and Robust Learning by Reinforcement Signals: Explorations in the Insect Brain

We propose a model for pattern recognition in the insect brain. Departing from a well-known body of knowledge about the insect brain, we investigate which of the potentially present features may be useful to learn input patterns rapidly and in a stable manner. The plasticity underlying pattern recognition is situated in the insect mushroom bodies and requires an error signal to associate the stimulus with a proper response. As a proof of concept, we used our model insect brain to classify the well-known MNIST database of handwritten digits, a popular benchmark for classifiers. We show that the structural organization of the insect brain appears to be suitable for both fast learning of new stimuli and reasonable performance in stationary conditions. Furthermore, it is extremely robust to damage to the brain structures involved in sensory processing. Finally, we suggest that spatiotemporal dynamics can improve the level of confidence in a classification decision. The proposed approach allows testing the effect of hypothesized mechanisms rather than speculating on their benefit for system performance or confidence in its responses.

Maximum unbiased validation (MUV) data sets for virtual screening based on PubChem bioactivity data

Refined nearest neighbor analysis was recently introduced for the analysis of virtual screening benchmark data sets. It constitutes a technique from the field of spatial statistics and provides a mathematical framework for the nonparametric analysis of mapped point patterns. Here, refined nearest neighbor analysis is used to design benchmark data sets for virtual screening based on PubChem bioactivity data. A workflow is devised that purges data sets of compounds active against pharmaceutically relevant targets from unselective hits. Topological optimization using experimental design strategies monitored by refined nearest neighbor analysis functions is applied to generate corresponding data sets of actives and decoys that are unbiased with regard to analogue bias and artificial enrichment. These data sets provide a tool for Maximum Unbiased Validation (MUV) of virtual screening methods. The data sets and a software package implementing the MUV design workflow are freely available at http://www.pharmchem.tu-bs.de/lehre/baumann/MUV.html.

Chemosensor-driven artificial antennal lobe transient dynamics enable fast recognition and working memory

The speed and accuracy of odor recognition in insects can hardly be resolved by the raw descriptors provided by olfactory receptors alone due to their slow time constant and high variability. The animal overcomes these barriers by means of the antennal lobe (AL) dynamics, which consolidates the classificatory information in receptor signal with a spatiotemporal code that is enriched in odor sensitivity, particularly in its transient. Inspired by this fact, we propose an easily implementable AL-like network and show that it significantly expedites and enhances the identification of odors from slow and noisy artificial polymer sensor responses. The device owes its efficiency to two intrinsic mechanisms: inhibition (which triggers a competition) and integration (due to the dynamical nature of the network). The former functions as a sharpening filter extracting the features of receptor signal that favor odor separation, whereas the latter implements a working memory by accumulating the extracted features in trajectories. This cooperation boosts the odor specificity during the receptor transient, which is essential for fast odor recognition.

Self-organizing molecular fingerprints: a ligand-based view on drug-like chemical space and off-target prediction

Background: Reliable prediction of multiple ligand–receptor interactions for a given bioactive compound helps recognize and understand off-target effects, and enables drug re-purposing and scaffold-hopping in lead discovery. We developed a ligand-based computational method for drug-target prediction that is independent from protein structural analysis. Method: The idea is to infer drug targets from the pharmacophoric feature similarity of known ligands, and define functional target similarity from a ligand perspective, which also provides access to targets with unknown structures. First, known ligands were represented by topological pharmacophoric features. Then, the self-organizing map technique was used to generate fingerprint patterns for similarity analysis, where each resulting fingerprint represents a drug target. Target fingerprints were clustered and analyzed for correlations. Well-structured dendrograms were obtained presenting interpretable and meaningful relationships between drug targets. Conclusion: Self-organization of fingerprints reduces noise from molecular pharmacophore descriptors, captures their essential features, and reveals potential cross-activities of ligand classes and off-target effects of bioactive compounds.

Olfactory information processing in the Drosophila antennal lobe: anything goes?

When an animal smells an odor, olfactory sensory neurons generate an activity pattern across olfactory glomeruli of the first sensory neuropil, the insect antennal lobe or the vertebrate olfactory bulb. Here, several networks of local neurons interact with sensory neurons and with output neurons--insect projection neurons, or vertebrate mitral/tufted cells. The extent and form of information processing taking place in these local networks has been subject of controversy. To investigate the role of local neurons in odor information processing we have used the calcium sensor G-CaMP to perform in vivo recordings of odor-evoked spatiotemporal activity patterns in five genetically defined neuron populations of the antennal lobe of Drosophila melanogaster: three distinct populations of local neurons (two GABAergic and one cholinergic), as well as sensory neurons and projection neurons. Odor-specific and concentration dependent spatiotemporal response patterns varied among neuron populations. Activity transfer differed along the olfactory pathway for different glomerulus-odor combinations: we found cases of profile broadening and of linear and complex transfer. Moreover, the discriminability between the odors also varied across neuron populations and was maximal in projection neurons. Discriminatory power increased with higher odor concentrations over a wide dynamic range, but decreased at the highest concentration. These results show the complexity and diversity of odor information processing mechanisms across olfactory glomeruli in the fly antennal lobe.

SuperScent--a database of flavors and scents

Volatiles are efficient mediators of chemical communication acting universally as attractant, repellent or warning signal in all kingdoms of life. Beside this broad impact volatiles have in nature, scents are also widely used in pharmaceutical, food and cosmetic industries, so the identification of new scents is of great industrial interest. Despite this importance as well as the vast number and diversity of volatile compounds, there is currently no comprehensive public database providing information on structure and chemical classification of volatiles. Therefore, the database SuperScent was established to supply users with detailed information on the variety of odor components. The version of the database presented here comprises the 2D/3D structures of approximately 2100 volatiles and around 9200 synonyms as well as physicochemical properties, commercial availability and references. The volatiles are classified according to their origin, functionality and odorant groups. The information was extracted from the literature and web resources. SuperScent offers several search options, e.g. name, Pubchem ID number, species, functional groups, or molecular weight. SuperScent is available online at: http://bioinformatics.charite.de/superscent.

Mimicking the olfactory system for the classification of chemical data

Various computational techniques have been used in pharmacological research to classify chemical compounds based on their physicochemical properties and putative biological activity. The recent publication by Schmuker and Schneider describes a new approach for the processing and classification of chemical data. Their study was motivated by nature's solution for detection and discrimination of chemical data, which is manifested in the olfactory systems of vertebrates and invertebrates.