Discovery of genes involved with learning and memory: An experimental synthesis of Hirschian and Benzerian perspectives

The biological bases of learning and memory are being revealed today with a wide array of molecular approaches, most of which entail the analysis of dysfunction produced by gene disruptions. This perspective derives both from early “genetic dissections” of learning in mutant Drosophila by Seymour Benzer and colleagues and from earlier behavior-genetic analyses of learning and in Diptera by Jerry Hirsch and coworkers. Three quantitative-genetic insights derived from these latter studies serve as guiding principles for the former. First, interacting polygenes underlie complex traits. Consequently, learning/memory defects associated with single-gene mutants can be quantified accurately only in equilibrated, heterogeneous genetic backgrounds. Second, complex behavioral responses will be composed of genetically distinct functional components. Thus, genetic dissection of complex traits into specific biobehavioral properties is likely. Finally, disruptions of genes involved with learning/memory are likely to have pleiotropic effects. As a result, task-relevant sensorimotor responses required for normal learning must be assessed carefully to interpret performance in learning/memory experiments. In addition, more specific conclusions will be obtained from reverse-genetic experiments, in which gene disruptions are restricted in time and/or space.

Structure-activity relationships derived by machine learning: the use of atoms and their bond connectivities to predict mutagenicity by inductive logic programming.

We present a general approach to forming structure-activity relationships (SARs). This approach is based on representing chemical structure by atoms and their bond connectivities in combination with the inductive logic programming (ILP) algorithm PROGOL. Existing SAR methods describe chemical structure by using attributes which are general properties of an object. It is not possible to map chemical structure directly to attribute-based descriptions, as such descriptions have no internal organization. A more natural and general way to describe chemical structure is to use a relational description, where the internal construction of the description maps that of the object described. Our atom and bond connectivities representation is a relational description. ILP algorithms can form SARs with relational descriptions. We have tested the relational approach by investigating the SARs of 230 aromatic and heteroaromatic nitro compounds. These compounds had been split previously into two subsets, 188 compounds that were amenable to regression and 42 that were not. For the 188 compounds, a SAR was found that was as accurate as the best statistical or neural network-generated SARs. The PROGOL SAR has the advantages that it did not need the use of any indicator variables handcrafted by an expert, and the generated rules were easily comprehensible. For the 42 compounds, PROGOL formed a SAR that was significantly (P < 0.025) more accurate than linear regression, quadratic regression, and back-propagation. This SAR is based on an automatically generated structural alert for mutagenicity.