Structure Learning in Bayesian Sensorimotor Integration

Previous studies have shown that sensorimotor processing can often be described by Bayesian learning, in particular the integration of prior and feedback information depending on its degree of reliability. Here we test the hypothesis that the integration process itself can be tuned to the statistical structure of the environment. We exposed human participants to a reaching task in a three-dimensional virtual reality environment where we could displace the visual feedback of their hand position in a two dimensional plane. When introducing statistical structure between the two dimensions of the displacement, we found that over the course of several days participants adapted their feedback integration process in order to exploit this structure for performance improvement. In control experiments we found that this adaptation process critically depended on performance feedback and could not be induced by verbal instructions. Our results suggest that structural learning is an important meta-learning component of Bayesian sensorimotor integration.     

Size Dependent Male Reproductive Tactic in the Two-Spotted Goby (Gobiusculus flavescens)

Male investment in testes and sperm duct gland in the polygamous nest breeding two-spotted goby Gobiusculus flavescens (Fabricius) was investigated in relation to time in reproductive season and individual physical parameters. This small teleost fish is most likely the most abundant species found along the rocky shores of the North East Atlantic. The two-spotted goby has a single reproductive season, during which nest-caring males can raise several clutches of offspring. According to the literature the males are on average larger than the females. Here we report for the first time a population showing a reversal of this trend, with males on average being smaller than females, a difference likely caused by a large proportion of small males. Early in the breeding season these small males have typical sneaker characters, with relatively large testes and small seminal duct glands compared to the larger dominant territorial males. The presence of these two alternative male reproductive tactics is confirmed by histological studies, which shows the presence of sperm in the sperm duct glands (SDG) of smaller males, but not in the SDG of intermediate and larger males. To our knowledge, males with typical sneaker characters have not been reported in earlier studied populations of two-spotted goby. Interestingly we found that testes investment declined significantly over the course of the breeding season, and that this reduction was significantly more pronounced in small compared to the large males. Further, a significant increase in seminal duct gland (SDG) mass was observed for the smaller males over the breeding season. We propose that this indicates a possible shift in mating tactic by smaller males from a parasitic to a nest-holding tactic over the course of the breeding season. Thus, the observed size dependent plasticity in investment in SDG over time suggests that the reproductive tactic of G. flavescens is conditional, and possibly influenced by mate availability and male—male competition.     

Testing generalized allometries in allocation modeling within an individual-based simulation framework

The allocation of carbohydrates from photosynthesis to various plant compartments is a key process in ecophysiology and consequently an important element in process-based ecosystem modeling. In this study, we tested generalized empirical equations in a widely applied partitioning concept based on compartment-specific biomass allometries. For an 88-year chronosequence of European beech (Fagus sylvatica L.) in Austria, we used the individual-based hybrid forest model PICUS v1.4 to compare simulations employing foliage biomass functions at different levels of generalization against runs with site-specific parameterization and observations. Sensitivities of the individual tree model were generally in line with the original stand-level partitioning concept and ecological process understanding. While stand-level leaf area increased with increasing allocation to foliage, net primary productivity showed no significant response due to saturated radiation interception in the dense chronosequence stands. Strong sensitivities were revealed at the individual tree level, where favoring allocation to the foliage compartment resulted in increasing asymmetry of competition and height–diameter relationships. Applying a generalized parameterization based on data from the full range of continental species distribution resulted in a significant overestimation of mean tree height and subsequently standing volume stock at the chronosequence. At a lower hierarchical level of generality, however, simulations with a representative regional parameterization performed satisfactorily compared to model runs using the site-specific allometry. In relation to common accuracy demands, e.g., in forest management decision support, the study suggests the rejection of a generic parameterization while corroborating the use of regional generalizations in ecosystem models.     

Inhibition of Bfl-1 with N-aryl maleimides

High-throughput screening of 66,000 compounds using competitive binding of peptides comprising the BH3 domain to anti-apoptotic Bfl-1 led to the identification of 14 validated 'hits' as inhibitors of Bfl-1. N-Aryl maleimide 1 was among the validated 'hits'. A chemical library encompassing over 280 analogs of 1 was prepared following a two-step synthesis. Structure-activity studies for inhibition of Bfl-1 by analogs of N-aryl maleimide 1 revealed a preference for electron-withdrawing substituents in the N-aryl ring and hydrophilic amines appended to the maleimide core. Inhibitors of Bfl-1 are potential development candidates for anti-cancer therapeutics.     

A library of linear undecapeptides with bactericidal activity against phytopathogenic bacteria

A 125-member library of synthetic linear undecapeptides was prepared based on a previously described peptide H-K¹KLFKKILKF¹⁰L-NH₂ (BP76) that inhibited in vitro growth of the plant pathogenic bacteria Erwinia amylovora, Xanthomonas axonopodis pv. vesicatoria, and Pseudomonas syringae pv. syringae at low micromolar concentrations. Peptides were designed using a combinatorial chemistry approach by incorporating amino acids possessing various degrees of hydrophobicity and hydrophilicity at positions 1 and 10 and by varying the N-terminus. Library screening for in vitro growth inhibition identified 27, 40 and 113 sequences with MIC values below 7.5 μM against E. amylovora, P. syringae and X. axonopodis, respectively. Cytotoxicity, bactericidal activity and stability towards protease degradation of the most active peptides were also determined. Seven peptides with a good balance between antibacterial and hemolytic activities were identified. Several analogues displayed a bactericidal effect and low susceptibility to protease degradation. The most promising peptides were tested in vivo by evaluating their preventive effect of inhibition of E. amylovora infection in detached apple and pear flowers. The peptide H-KKLFKKILKYL-NH₂ (BP100) showed efficacies in flowers of 63–76% at 100 μM, being more potent than BP76 and only less effective than streptomycin, currently used for fire blight control.     

Pattern storage in a sparsely coded neural network with cyclic activation

We investigate an artificial neural network model with a modified Hebb rule. It is an auto-associative neural network similar to the Hopfield model and to the Willshaw model. It has properties of both of these models. Another property is that the patterns are sparsely coded and are stored in cycles of synchronous neural activities. The cycles of activity for some ranges of parameter increase the capacity of the model. We discuss basic properties of the model and some of the implementation issues, namely optimizing of the algorithms. We describe the modification of the Hebb learning rule, the learning algorithm, the generation of patterns, decomposition of patterns into cycles and pattern recall.