Time Series Input Selection using Multiple Kernel Learning

In this paper we study the relevance of multiple kernel learning (MKL) for the automatic selection of time series inputs. Recently, MKL has gained great attention in the machine learning community due to its flexibility in modelling complex patterns and performing feature selection. In general, MKL constructs the kernel as a weighted linear combination of basis kernels, exploiting different sources of information. An efficient algorithm wrapping a Support Vector Regression model for optimizing the MKL weights, named SimpleMKL, is used for the analysis. In this sense, MKL performs feature selection by discarding inputs/kernels with low or null weights. The approach proposed is tested with simulated linear and nonlinear time series (AutoRegressive, Henon and Lorenz series).

Electrical neuroimaging of memory discrimination based on single-trial multisensory learning.

Multisensory experiences influence subsequent memory performance and brain responses. Studies have thus far concentrated on semantically congruent pairings, leaving unresolved the influence of stimulus pairing and memory sub-types. Here, we paired images with unique, meaningless sounds during a continuous recognition task to determine if purely episodic, single-trial multisensory experiences can incidentally impact subsequent visual object discrimination. Psychophysics and electrical neuroimaging analyses of visual evoked potentials (VEPs) compared responses to repeated images either paired or not with a meaningless sound during initial encounters. Recognition accuracy was significantly impaired for images initially presented as multisensory pairs and could not be explained in terms of differential attention or transfer of effects from encoding to retrieval. VEP modulations occurred at 100-130ms and 270-310ms and stemmed from topographic differences indicative of network configuration changes within the brain. Distributed source estimations localized the earlier effect to regions of the right posterior temporal gyrus (STG) and the later effect to regions of the middle temporal gyrus (MTG). Responses in these regions were stronger for images previously encountered as multisensory pairs. Only the later effect correlated with performance such that greater MTG activity in response to repeated visual stimuli was linked with greater performance decrements. The present findings suggest that brain networks involved in this discrimination may critically depend on whether multisensory events facilitate or impair later visual memory performance. More generally, the data support models whereby effects of multisensory interactions persist to incidentally affect subsequent behavior as well as visual processing during its initial stages.

An Automatic Approach for Learning and Tuning Gaussian Interval Type-2 Fuzzy Membership Functions Applied to Lung CAD Classification System

The potential of type-2 fuzzy sets for managing high levels of uncertainty in the subjective knowledge of experts or of numerical information has focused on control and pattern classification systems in recent years. One of the main challenges in designing a type-2 fuzzy logic system is how to estimate the parameters of type-2 fuzzy membership function (T2MF) and the Footprint of Uncertainty (FOU) from imperfect and noisy datasets. This paper presents an automatic approach for learning and tuning Gaussian interval type-2 membership functions (IT2MFs) with application to multi-dimensional pattern classification problems. T2MFs and their FOUs are tuned according to the uncertainties in the training dataset by a combination of genetic algorithm (GA) and crossvalidation techniques. In our GA-based approach, the structure of the chromosome has fewer genes than other GA methods and chromosome initialization is more precise. The proposed approach addresses the application of the interval type-2 fuzzy logic system (IT2FLS) for the problem of nodule classification in a lung Computer Aided Detection (CAD) system. The designed IT2FLS is compared with its type-1 fuzzy logic system (T1FLS) counterpart. The results demonstrate that the IT2FLS outperforms the T1FLS by more than 30% in terms of classification accuracy.

Learning ability and longevity: a symmetrical evolutionary trade-off in Drosophila.

Learning ability can be substantially improved by artificial selection in animals ranging from Drosophila to rats. Thus these species have not used their evolutionary potential with respect to learning ability, despite intuitively expected and experimentally demonstrated adaptive advantages of learning. This suggests that learning is costly, but this notion has rarely been tested. Here we report correlated responses of life-history traits to selection for improved learning in Drosophila melanogaster. Replicate populations selected for improved learning lived on average 15% shorter than the corresponding unselected control populations. They also showed a minor reduction in fecundity late in life and possibly a minor increase in dry adult mass. Selection for improved learning had no effect on egg-to-adult viability, development rate, or desiccation resistance. Because shortened longevity was the strongest correlated response to selection for improved learning, we also measured learning ability in another set of replicate populations that had been selected for extended longevity. In a classical olfactory conditioning assay, these long-lived flies showed an almost 40% reduction in learning ability early in life. This effect disappeared with age. Our results suggest a symmetrical evolutionary trade-off between learning ability and longevity in Drosophila.

Functional avidity of tumor antigen-specific CTL recognition directly correlates with the stability of MHC/peptide multimer binding to TCR.

Avidity of Ag recognition by tumor-specific T cells is one of the main parameters that determines the potency of a tumor rejection Ag. In this study we show that the relative efficiency of staining of tumor Ag-specific T lymphocytes with the corresponding fluorescent MHC class I/peptide multimeric complexes can considerably vary with staining conditions and does not necessarily correlate with avidity of Ag recognition. Instead, we found a clear correlation between avidity of Ag recognition and the stability of MHC class I/peptide multimeric complexes interaction with TCR as measured in dissociation kinetic experiments. These findings are relevant for both identification and isolation of tumor-reactive CTL.

Morphometric shape analysis using learning vector quantization neural networks - an example distinguishing two microtine vole species

Closely related species may be very difficult to distinguish morphologically, yet sometimes morphology is the only reasonable possibility for taxonomic classification. Here we present learning-vector-quantization artificial neural networks as a powerful tool to classify specimens on the basis of geometric morphometric shape measurements. As an example, we trained a neural network to distinguish between field and root voles from Procrustes transformed landmark coordinates on the dorsal side of the skull, which is so similar in these two species that the human eye cannot make this distinction. Properly trained neural networks misclassified only 3% of specimens. Therefore, we conclude that the capacity of learning vector quantization neural networks to analyse spatial coordinates is a powerful tool among the range of pattern recognition procedures that is available to employ the information content of geometric morphometrics.

The serine-rich N-terminal region of Arabidopsis phytochrome A is required for protein stability.

Deletion or substitution of the serine-rich N-terminal stretch of grass phytochrome A (phyA) has repeatedly been shown to yield a hyperactive photoreceptor when expressed under the control of a constitutive promoter in transgenic tobacco or Arabidopsis seedlings retaining their native phyA. These observations have lead to the proposal that the serine-rich region is involved in negative regulation of phyA signaling. To re-evaluate this conclusion in a more physiological context we produced transgenic Arabidopsis seedlings of the phyA-null background expressing Arabidopsis PHYA deleted in the sequence corresponding to amino acids 6-12, under the control of the native PHYA promoter. Compared to the transgenic seedlings expressing wild-type phyA, the seedlings bearing the mutated phyA showed normal responses to pulses of far-red (FR) light and impaired responses to continuous FR light. In yeast two-hybrid experiments, deleted phyA interacted normally with FHY1 and FHL, which are required for phyA accumulation in the nucleus. Immunoblot analysis showed reduced stability of deleted phyA under continuous red or FR light. The reduced physiological activity can therefore be accounted for by the enhanced destruction of the mutated phyA. These findings do not support the involvement of the serine-rich region in negative regulation but they are consistent with a recent report suggesting that phyA turnover is regulated by phosphorylation.

Fatty acids regulate Thy-1 antigen mRNA stability in T lymphocyte precursors.

In this study, we report the effect of fatty acids on the Thy-1 antigen mRNA decay. Low serum and synthetic medium culture conditions were used to demonstrate that fatty acids, which are important metabolites involved as second messengers in signal transduction, also influence the steady-state mRNA level. Detailed analysis demonstrated that polyunsaturated lipids attached to bovine serum albumin, such as linoleic, linolenic, and arachidonic acids, modulate gene expression specifically in the S1A T lymphoma cell line by inducing a 3-5-fold increase in the steady-state Thy-1 mRNA level, concomitant with a twofold increase in cell surface expression. A similar modulation was observed in the immature CD4-CD8- T cell precursors but not in mature thymocytes. Nuclear run-on and transfection experiments indicated that the observed Thy-1 mRNA level is post-transcriptionally regulated and that the presence of the coding region is sufficient for this adaptive response. A mechanism without a requirement for protein kinase C activation, but involving Ca2+ entry, could account for this difference in Thy-1 mRNA stability.

Engineered aprotinin for improved stability of fibrin biomaterials.

Fibrin has been long used clinically for hemostasis and sealing, yet extension of use in other applications has been limited due to its relatively rapid resorption in vivo, even with addition of aprotinin or other protease inhibitors. We report an engineered aprotinin variant that can be immobilized within fibrin and thus provide extended longevity. When recombinantly fused to a transglutaminase substrate domain from α(2)-plasmin inhibitor (α(2)PI(1-8)), the resulting variant, aprotinin-α(2)PI(1-8), was covalently crosslinked into fibrin matrices during normal thrombin/factor XIIIa-mediated polymerization. Challenge with physiological plasmin concentrations revealed that aprotinin-α(2)PI(1-8)-containing matrices retained 78% of their mass after 3 wk, whereas matrices containing wild type (WT) aprotinin degraded completely within 1 wk. Plasmin challenge of commercial sealants Omrixil and Tisseel, supplemented with aprotinin-α(2)PI(1-8) or WT aprotinin, showed extended longevity as well. When seeded with human dermal fibroblasts, aprotinin-α(2)PI(1-8)-supplemented matrices supported cell growth for at least 33% longer than those containing WT aprotinin. Subcutaneously implanted matrices containing aprotinin-α(2)PI(1-8) were detectable in mice for more than twice as long as those containing WT aprotinin. We conclude that our engineered recombinant aprotinin variant can confer extended longevity to fibrin matrices more effectively than WT aprotinin in vitro and in vivo.

Intra-individual stability of neuromotor tasks from 6 to 18 years: a longitudinal study.

This study investigates the intra-individual stability of the speed of several motor tasks and the intensity of associated movements in 256 children (131 girls, 125 boys) from the Zurich generational study using the Zurich neuromotor assessment battery (ZNA) over a 12-year period from the age of 6 to 18 years. The stability was assessed by correlograms of standard deviation scores calculated from age- and gender-adjusted normative values and compared with standing height and full scale intelligence quotient (IQ). While motor tasks of hand, finger and foot (HFT) and contralateral associated movements (CAM) exhibited a moderate stability (summary measure as correlation coefficients between two measurements made 4 years apart: .61 and .60), other tasks (dynamic balance, static balance and pegboard) were only weakly stable (.46, .47 and .49). IQ and height were more stable than neuromotor components (.72 and .86). We conclude that the moderately stable HFT and CAM may reflect "motor traits", while the stability of the pegboard and balance tasks is weaker because these skills are more experience related and state-dependent.