Generalized optimal spatial filtering using a kernel approach with application to EEG classification

Common spatial patterns (CSP) has been widely used for finding the linear spatial filters which are able to extract the discriminative brain activities between two different mental tasks. However, the CSP is difficult to capture the nonlinearly clustered structure from the non-stationary EEG signals. To relax the presumption of strictly linear patterns in the CSP, in this paper, a generalized CSP (GCSP) based on generalized singular value decomposition (GSVD) and kernel method is proposed. Our method is able to find the nonlinear spatial filters which are formulated in the feature space defined by a nonlinear mapping through kernel functions. Furthermore, in order to overcome the overfitting problem, the regularized GCSP is developed by adding the regularized parameters. The experimental results demonstrate that our method is an effective nonlinear spatial filtering method.     

Robustness, evolvability, and optimality of evolutionary neural networks

In a typical optimization problem, the main goal is to search for the appropriate values of the variables that provide the optimal solution of the given function. In artificial neural networks (ANN), this translates to the minimization of the error surface during training such that misclassification is minimized during generalization. However, since optimal training performance does not necessarily imply optimal generalization due to the possibility of overfitting or underfitting, we developed SEPA (Structure Evolution and Parameter Adaptation) which addressed these issues by simultaneously evolving ANN structure and weights. Since SEPA primarily relies on the perturbation function to bring variation in its population, this follow-up study aims to find out SEPAs evolvability, optimality, and robustness in other perturbation functions. Our findings indicate that SEPAs optimal generalization performances are stable and robust from the effect of the different perturbation functions. This is due to the feedback loop between its architecture evolution and weight adaptation such that any shortcoming of the former is compensated by the latter, and vice versa. Our results strongly suggest that proper ANN design requires simultaneous adaptation of ANN structure and weights to avoid one-sided or bias convergence to either the weight or architecture space.     

Research on Plant Species Identification Based on Improved Convolutional Neural Network

Plant species recognition is an important research area in image recognition in recent years. However, the existing plant species recognition methods have low recognition accuracy and do not meet professional requirements in terms of recognition accuracy. Therefore, ShuffleNetV2 was improved by combining the current hot concern mechanism, convolution kernel size adjustment, convolution tailoring, and CSP technology to improve the accuracy and reduce the amount of computation in this study. Six convolutional neural network models with sufficient trainable parameters were designed for differentiation learning. The SGD algorithm is used to optimize the training process to avoid overfitting or falling into the local optimum. In this paper, a conventional plant image dataset TJAU10 collected by cell phones in a natural context was constructed, containing 3000 images of 10 plant species on the campus of Tianjin Agricultural University. Finally, the improved model is compared with the baseline version of the model, which achieves better results in terms of improving accuracy and reducing the computational effort. The recognition accuracy tested on the TJAU10 dataset reaches up to 98.3%, and the recognition precision reaches up to 93.6%, which is 5.1% better than the original model and reduces the computational effort by about 31% compared with the original model. In addition, the experimental results were evaluated using metrics such as the confusion matrix, which can meet the requirements of professionals for the accurate identification of plant species.     

人工神经网络模型在水稻群体分蘖动态模拟中外推能力的测试

研究利用人工神经网络模型 ,以水稻群体分蘖动态为例 ,采用交互验证和独立验证的方式 ,对水稻生长 BP网络模型进行了训练与模拟 ,其结果与水稻群体分蘖的积温统计模型、基本动力学模型和复合分蘖模型进行了比较。研究结果表明 ,神经网络模型具有一定的外推能力 ,但其外推能力依赖于大量的训练样本。神经网络模型具有较好的拟合能力 ,是因为有较多的模型参数 ,因此对神经网络模型的训练需要大量的参数来保证其参数不致过度吻合。具有外推能力神经网络模型的最少训练样本数应大于 6 .75倍于神经网络参数数目 ,小于 13.5倍于神经网络参数数目。因此在应用神经网络模型时 ,如果神经网络模型包括较多的输入变量时 ,可考虑采用主成分分析、对应分析等技术对输入变量进行信息综合 ,相应地减少网络模型的参数。另一方面 ,当训练样本不足时 ,最好只用神经网络模型对同一系统的情况进行模拟 ,应谨慎使用神经网络模型进行外推。神经网络模型给作物模拟研究的科学工作者提供了一个“傻瓜”式工具 ,对数学建模不熟悉的农业研究人员 ,人工神经网络可以替代数学建模进行仿真实验 ;对于精通数学建模的研究人员来说 ,它至少是一种补充和可作为比较的非线性数据处理方法     

Avoiding overfitting in multilayer perceptrons with feeling-of-knowing using self-organizing maps

Overfitting in multilayer perceptron (MLP) training is a serious problem. The purpose of this study is to avoid overfitting in on-line learning. To overcome the overfitting problem, we have investigated feeling-of-knowing (FOK) using self-organizing maps (SOMs). We propose MLPs with FOK using the SOMs method to overcome the overfitting problem. In this method, the learning process advances according to the degree of FOK calculated using SOMs. The mean square error obtained for the test set using the proposed method is significantly less than that in a conventional MLP method. Consequently, the proposed method avoids overfitting.     

Assessment of survival prediction models based on microarray data

MOTIVATION: In the process of developing risk prediction models, various steps of model building and model selection are involved. If this process is not adequately controlled, overfitting may result in serious overoptimism leading to potentially erroneous conclusions. METHODS: For right censored time-to-event data, we estimate the prediction error for assessing the performance of a risk prediction model (Gerds and Schumacher, 2006; Graf et al., 1999). Furthermore, resampling methods are used to detect overfitting and resulting overoptimism and to adjust the estimates of prediction error (Gerds and Schumacher, 2007). RESULTS: We show how and to what extent the methodology can be used in situations characterized by a large number of potential predictor variables where overfitting may be expected to be overwhelming. This is illustrated by estimating the prediction error of some recently proposed techniques for fitting a multivariate Cox regression model applied to the data of a prognostic study in patients with diffuse large-B-cell lymphoma (DLBCL). AVAILABILITY: Resampling-based estimation of prediction error curves is implemented in an R package called pec available from the authors.     

Intermolecular electrostatics via molecular fragmentation: an application to crystal structure prediction

Molecular fragmentation is an attractive approach to the simulation of large molecules, in which calculations are carried out on small segments of the molecule, achieving linear scaling but reduced accuracy. Its application to crystal structure prediction (CSP) is challenged by high accuracy requirements. In this study, the applicability of a fragmentation scheme is tested for distributed multipoles, which are used in CSP to model intermolecular electrostatic interactions. Four test systems are investigated: a molecular salt, the highly conjugated molecule retinal, a model pharmaceutical molecule and the nonlinear molecule nitrotriacetanilide. It is demonstrated that fragment-based electrostatics reproduce, to an acceptable degree, a set of crystal structures generated using whole-molecule electrostatics. Inclusion of the molecular environment of each fragment out to four bonds separation is found to provide a sufficiently accurate set of distributed multipoles for the purposes of CSP.     

Characterization of the G alpha(s) regulator cysteine string protein

Cysteine string protein (CSP) is an abundant regulated secretory vesicle protein that is composed of a string of cysteine residues, a linker domain, and an N-terminal J domain characteristic of the DnaJ/Hsp40 co-chaperone family. We have shown previously that CSP associates with heterotrimeric GTP-binding proteins (G proteins) and promotes G protein inhibition of N-type Ca2+ channels. To elucidate the mechanisms by which CSP modulates G protein signaling, we examined the effects of CSP(1-198) (full-length), CSP(1-112), and CSP(1-82) on the kinetics of guanine nucleotide exchange and GTP hydrolysis. In this report, we demonstrate that CSP selectively interacts with G alpha(s) and increases steady-state GTP hydrolysis. CSP(1-198) modulation of G alpha(s) was dependent on Hsc70 (70-kDa heat shock cognate protein) and SGT (small glutamine-rich tetratricopeptide repeat domain protein), whereas modulation by CSP(1-112) was Hsc70-SGT-independent. CSP(1-112) preferentially associated with the inactive GDP-bound conformation of G alpha(s). Consistent with the stimulation of GTP hydrolysis, CSP(1-112) increased guanine nucleotide exchange of G alpha(s). The interaction of native G alpha(s) and CSP was confirmed by coimmunoprecipitation and showed that G alpha(s) associates with CSP. Furthermore, transient expression of CSP in HEK cells increased cellular cAMP levels in the presence of the beta2 adrenergic agonist isoproterenol. Together, these results demonstrate that CSP modulates G protein function by preferentially targeting the inactive GDP-bound form of G alpha(s) and promoting GDP/GTP exchange. Our results show that the guanine nucleotide exchange activity of full-length CSP is, in turn, regulated by Hsc70-SGT.     

Spending degrees of freedom in a poor economy: A case study of building a sightability model for moose in northeastern Minnesota

Sightability models are binary logistic-regression models used to estimate and adjust for visibility bias in wildlife-population surveys. Like many models in wildlife and ecology, sightability models are typically developed from small observational datasets with many candidate predictors. Aggressive model-selection methods are often employed to choose a best model for prediction and effect estimation, despite evidence that such methods can lead to overfitting (i.e., selected models may describe random error or noise rather than true predictor–response curves) and poor predictive ability. We used moose (Alces alces) sightability data from northeastern Minnesota (2005–2007) as a case study to illustrate an alternative approach, which we refer to as degrees-of-freedom (df) spending: sample-size guidelines are used to determine an acceptable level of model complexity and then a pre-specified model is fit to the data and used for inference. For comparison, we also constructed sightability models using Akaike's Information Criterion (AIC) step-down procedures and model averaging (based on a small set of models developed using df-spending guidelines). We used bootstrap procedures to mimic the process of model fitting and prediction, and to compute an index of overfitting, expected predictive accuracy, and model-selection uncertainty. The index of overfitting increased 13% when the number of candidate predictors was increased from three to eight and a best model was selected using step-down procedures. Likewise, model-selection uncertainty increased when the number of candidate predictors increased. Model averaging (based on R = 30 models with 1–3 predictors) effectively shrunk regression coefficients toward zero and produced similar estimates of precision to our 3-df pre-specified model. As such, model averaging may help to guard against overfitting when too many predictors are considered (relative to available sample size). The set of candidate models will influence the extent to which coefficients are shrunk toward zero, which has implications for how one might apply model averaging to problems traditionally approached using variable-selection methods. We often recommend the df-spending approach in our consulting work because it is easy to implement and it naturally forces investigators to think carefully about their models and predictors. Nonetheless, similar concepts should apply whether one is fitting 1 model or using multi-model inference. For example, model-building decisions should consider the effective sample size, and potential predictors should be screened (without looking at their relationship to the response) for missing data, narrow distributions, collinearity, potentially overly influential observations, and measurement errors (e.g., via logical error checks). © 2011 The Wildlife Society.     

A network pharmacology‐based approach to explore the effects of Chaihu Shugan powder on a non‐alcoholic fatty liver rat model through nuclear receptors

The pathogenesis of non‐alcoholic fatty liver disease (NAFLD) is still not fully understood, and currently, no effective pharmacotherapy is available. Nuclear receptors (NRs) are important biological participants in NAFLD that exhibit great therapeutic potential. Chaihu Shugan powder (CSP) is a traditional Chinese medicine (TCM) formula that has a wide therapeutic spectrum including NAFLD, but the effective components and functional mechanisms of CSP are unclear. We adopted a network pharmacology approach using multiple databases for Gene Ontology (GO) enrichment analysis and the molecular complex detection (MCODE) method for a protein‐protein interaction (PPI) analysis, and we used molecular docking method to screen the NR targets and determine the corresponding CSP components. The screening results were validated through a NAFLD rat model that was used to explain the possible relationship between CSP and NAFLD. Finally, we screened PPARγ, FXR, PPARα, RARα and PPARδ as target genes and quercetin, kaempferol, naringenin, isorhamnetin and nobiletin as target compounds. The five components were detected through high‐performance liquid chromatography‐mass spectrometry (HPLC‐MS), the results of which aligned with the docking experiments of PPARγ, PPARα and PPARδ. After CSP intervention, the NAFLD rat model showed ameliorated effects in terms of bodyweight, hepatic histopathology, and serum and liver lipids, and the mRNA levels of PPARγ, FXR, PPARα and RARα were significantly changed. The results from this study indicate that CSP exhibits healing effects in an NAFLD model and that the network pharmacology approach to screening NR targets and determining the corresponding CSP components is a practical strategy for explaining the mechanism by which CSP ameliorates NAFLD.     

A surface molecularly imprinted polymer as chiral stationary phase for chiral separation of 1,1′‐binaphthalene‐2‐naphthol racemates

Acrylamide (AM) was copolymerized with ethylene glycol dimethacrylate (EGDMA) in the presence of (R )‐1,1′‐binaphthalene‐2‐naphthol (BINOL) as the template molecules on the surface of silica gel by a free radical polymerization to produce a chiral stationary phase based on the surface molecularly imprinted polymer (SMIP‐CSP). The SMIP‐CSP showed a much better separation factor (α = 4.28) than the CSP based on the molecularly imprinted polymer (MIP‐CSP) without coating on the silica gel (α = 1.96) during the chiral separation of BINOL enantiomers by high‐performance liquid chromatography. The influence of the pretreatment temperature and the content of the template molecule ((R )‐BINOL) of the SMIP‐CSP, and the mobile phase composition on the separation of the racemic BINOL were systematically investigated.     

Cysteine string protein (CSP) inhibition of N-type calcium channels is blocked by mutant huntingtin

Cysteine string protein (CSP), a 34-kDa molecular chaperone, is expressed on synaptic vesicles in neurons and on secretory vesicles in endocrine, neuroendocrine, and exocrine cells. CSP can be found in a complex with two other chaperones, the heat shock cognate protein Hsc70, and small glutamine-rich tetratricopeptide repeat domain protein (SGT). CSP function is vital in synaptic transmission; however, the precise nature of its role remains controversial. We have previously reported interactions of CSP with both heterotrimeric GTP-binding proteins (G proteins) and N-type calcium channels. These associations give rise to a tonic G protein inhibition of the channels. Here we have examined the effects of huntingtin fragments (exon 1) with (huntingtin(exon1/exp)) and without (huntingtin(exon1/nonexp)) expanded polyglutamine (polyQ) tracts on the CSP chaperone system. In vitro huntingtin(exon1/exp) sequestered CSP and blocked the association of CSP with G proteins. In contrast, huntingtin(exon1/nonexp) did not interact with CSP and did not alter the CSP/G protein association. Similarly, co-expression of huntingtin(exon1/exp) with CSP and N-type calcium channels eliminated CSP's tonic G protein inhibition of the channels, while coexpression of huntingtin(exon1/nonexp) did not alter the robust inhibition promoted by CSP. These results indicate that CSP's modulation of G protein inhibition of calcium channel activity is blocked in the presence of a huntingtin fragment with expanded polyglutamine tracts.     

Regularized common spatial patterns with subject-to-subject transfer of EEG signals

In the context of brain-computer interface (BCI) system, the common spatial patterns (CSP) method has been used to extract discriminative spatial filters for the classification of electroencephalogram (EEG) signals. However, the classification performance of CSP typically deteriorates when a few training samples are collected from a new BCI user. In this paper, we propose an approach that maintains or improves the recognition accuracy of the system with only a small size of training data set. The proposed approach is formulated by regularizing the classical CSP technique with the strategy of transfer learning. Specifically, we incorporate into the CSP analysis inter-subject information involving the same task, by minimizing the difference between the inter-subject features. Experimental results on two data sets from BCI competitions show that the proposed approach greatly improves the classification performance over that of the conventional CSP method; the transformed variant proved to be successful in almost every case, based on a small number of available training samples.     

Phosphorylation of cysteine string protein by protein kinase A. Implications for the modulation of exocytosis

Cyclic AMP-dependent protein kinase (PKA) enhances regulated exocytosis in neurons and most other secretory cells. To explore the molecular basis of this effect, known exocytotic proteins were screened for PKA substrates. Both cysteine string protein (CSP) and soluble NSF attachment protein-alpha (alpha-SNAP) were phosphorylated by PKA in vitro, but immunoprecipitation of cellular alpha-SNAP failed to detect (32)P incorporation. In contrast, endogenous CSP was phosphorylated in synaptosomes, PC12 cells, and chromaffin cells. In-gel kinase assays confirmed PKA to be a cellular CSP kinase, with phosphorylation occurring on Ser(10). PKA phosphorylation of CSP reduced its binding to syntaxin by 10-fold but had little effect on its interaction with HSC70 or G-protein subunits. Furthermore, an in vivo role for Ser(10) phosphorylation at a late stage of exocytosis is suggested by analysis of chromaffin cells transfected with wild type or non-phosphorylatable mutant CSP. We propose that PKA phosphorylation of CSP could modulate the exocytotic machinery, by selectively altering its availability for protein-protein interactions.     

The Synergy between Complex Channel-Specific FIR Filter and Spatial Filter for Single-Trial EEG Classification

The common spatial pattern analysis (CSP), a frequently utilized feature extraction method in brain-computer-interface applications, is believed to be time-invariant and sensitive to noises, mainly due to an inherent shortcoming of purely relying on spatial filtering. Therefore, temporal/spectral filtering which can be very effective to counteract the unfavorable influence of noises is usually used as a supplement. This work integrates the CSP spatial filters with complex channel-specific finite impulse response (FIR) filters in a natural and intuitive manner. Each hybrid spatial-FIR filter is of high-order, data-driven and is unique to its corresponding channel. They are derived by introducing multiple time delays and regularization into conventional CSP. The general framework of the method follows that of CSP but performs better, as proven in single-trial classification tasks like event-related potential detection and motor imagery.     

Head-to-Head Comparison of Soluble vs. Qβ VLP Circumsporozoite Protein Vaccines Reveals Selective Enhancement of NANP Repeat Responses

Circumsporozoite protein (CSP) of Plasmodium falciparum is a promising malaria vaccine target. RTS,S, the most advanced malaria vaccine candidate consists of the central NANP repeat and carboxy-terminal region of CSP displayed on a hepatitis B virus-like particle (VLP). To build upon the success of RTS,S, we produced a near full-length Plasmodium falciparum CSP that also includes the conserved amino-terminal region of CSP. We recently showed that this soluble CSP, combined with a synthetic Toll-like-receptor-4 (TLR4) agonist in stable oil-in-water emulsion (GLA/SE), induces a potent and protective immune response in mice against transgenic parasite challenge. Here we have investigated whether the immunogenicity of soluble CSP could be further augmented by presentation on a VLP. Bacteriophage Qβ VLPs can be readily produced in E.coli, they have a diameter of 25 nm and contain packaged E. coli RNA which serves as a built in adjuvant through the activation of TLR7/8. CSP was chemically conjugated to Qβ and the CSP-Qβ vaccine immunogenicity and efficacy were compared to adjuvanted soluble CSP in the C57Bl/6 mouse model. When formulated with adjuvants lacking a TLR4 agonist (Alum, SE and Montanide) the Qβ-CSP induced higher anti-NANP repeat titers, higher levels of cytophilic IgG2b/c antibodies and a trend towards higher protection against transgenic parasite challenge as compared to soluble CSP formulated in the same adjuvant. The VLP and soluble CSP immunogenicity difference was most pronounced at low antigen dose, and within the CSP molecule, the titers against the NANP repeats were preferentially enhanced by Qβ presentation. While a TLR4 agonist enhanced the immunogenicity of soluble CSP to levels comparable to the VLP vaccine, the TLR4 agonist did not further improve the immunogenicity of the Qβ-CSP vaccine. The data presented here pave the way for further improvement in the Qβ conjugation chemistry and evaluation of both the Qβ-CSP and soluble CSP vaccines in the non-human primate model.     

Asymptotic analysis of a TMDD model: when a reaction contributes to the destruction of its product

The multi-scale dynamics of a two-compartment with first order absorption Target-Mediated Drug Disposition (TMDD) pharmacokinetics model is analysed, using the Computational Singular Perturbation (CSP) algorithm. It is shown that the process evolves along two Slow Invariant Manifolds (SIMs), on which the most intense components of the model are equilibrated, so that the less intensive are the driving ones. The CSP tools allow for the identification of the components of the TMDD model that (i) constrain the evolution of the process on the SIMs, (ii) drive the system along the SIMs and (iii) generate the fast time scales. Among others, such diagnostics identify (i) the factors that determine the start and the duration of the period in which the ligand-receptor complex acts and (ii) the processes that determine its degradation rate. The counterintuitive influence of the process that transfers the ligand from the tissue to the main compartment, as it is manifested during the final stage of the process, is studied in detail.

     

Divergence of evolutionary ways among common sym genes: CASTOR and CCaMK show functional conservation between two symbiosis systems and constitute the root of a common signaling pathway

In recent years a number of legume genes involved in root nodule (RN) symbiosis have been identified in the model legumes, Lotus japonicus (Lotus) and Medicago truncatula. Among them, a distinct set of genes has been categorized as a common symbiosis pathway (CSP), because they are also essential for another mutual interaction, the arbuscular mycorrhiza (AM) symbiosis, which is evolutionarily older than the RN symbiosis and is widely distributed in the plant kingdom. Based on the concept that the legume RN symbiosis has evolved from the ancient AM symbiosis, one issue is whether the CSP is functionally conserved between non-nodulating plants, such as rice, and nodulating legumes. We identified three rice CSP gene orthologs, OsCASTOR, OsPOLLUX and OsCCaMK, and demonstrated the indispensable roles of OsPOLLUX and OsCCaMK in rice AM symbiosis. Interestingly, molecular transfection of either OsCASTOR or OsCCaMK could fully complement symbiosis defects in the corresponding Lotus mutant lines for both the AM and RN symbioses. Our results not only provide a conserved genetic basis for the AM symbiosis between rice and Lotus, but also indicate that the core of the CSP has been well conserved during the evolution of RN symbiosis. Through evolution, CASTOR and CCaMK have remained as the molecular basis for the maintenance of CSP functions in the two symbiosis systems.     

Protein kinase B/Akt is a novel cysteine string protein kinase that regulates exocytosis release kinetics and quantal size

Protein kinase B/Akt has been implicated in the insulin-dependent exocytosis of GLUT4-containing vesicles, and, more recently, insulin secretion. To determine if Akt also regulates insulin-independent exocytosis, we used adrenal chromaffin cells, a popular neuronal model. Akt1 was the predominant isoform expressed in chromaffin cells, although lower levels of Akt2 and Akt3 were also found. Secretory stimuli in both intact and permeabilized cells induced Akt phosphorylation on serine 473, and the time course of Ca2+-induced Akt phosphorylation was similar to that of exocytosis in permeabilized cells. To determine if Akt modulated exocytosis, we transfected chromaffin cells with Akt constructs and monitored catecholamine release by amperometry. Wild-type Akt had no effect on the overall number of exocytotic events, but slowed the kinetics of catecholamine release from individual vesicles, resulting in an increased quantal size. This effect was due to phosphorylation by Akt, because it was not seen in cells transfected with kinase-dead mutant Akt. As overexpression of cysteine string protein (CSP) results in a similar alteration in release kinetics and quantal size, we determined if CSP was an Akt substrate. In vitro 32P-phosphorylation studies revealed that Akt phosphorylates CSP on serine 10. Using phospho-Ser10-specific antisera, we found that both transfected and endogenous cellular CSP is phosphorylated by Akt on this residue. Taken together, these findings reveal a novel role for Akt phosphorylation in regulating the late stages of exocytosis and suggest that this is achieved via the phosphorylation of CSP on serine 10.