Search

• MSGMTL designs a new objective function, which integrates the idea of metric learning and transfer learning, alleviates the distribution difference and obtains a reliable metric matrix. • MSGMTL effectively uses the label information of source domain to establish similar feature pairs and dissimilar feature pairs, so as to promote target domain to learn the discriminative feature representation. • MSGMTL calculate the balance factor through metric matrix and dynamically allocate the weights of marginal distribution and conditional distribution. • MSGMTL proves the feasibility of metric transfer learning and provides a new idea for EEG classification of BCI. In the brain computer interfaces (BCIs), transfer learning (TL) has proven its effectiveness and attracted more attention in recent research. However, traditional TL algorithms mainly use Euclidean metric to calculate distance between features, not fully exploiting the potential relationship between feature representations, which makes the improvement of performance limited. This paper proposes a multi-source geometric metric transfer learning (MSGMTL) algorithm. Firstly, multiple sources are aggregated together through Euclidean alignment (EA) to minimize the marginal distribution. Secondly, the tangent space features are extracted from a manifold to obtain the covariance matrices of EEG samples. Thirdly, three optimization components are introduced into a unified function under Mahalanobis distance metric. Namely, MSGMTL integrates pairwise constraints balanced distribution adaption based metric and structure consistency, aiming to preserve discriminative information and geometric structure to improve the performance of motor imagery (MI) classification. Experiments conducted on three datasets show that, compared with other advanced methods, MSGMTL achieves better performance in classification accuracy and computational cost. Conclusion: It comes to the conclusion that the combination of metric learning and transfer learning has achieved superior performance for EEG classification and can be beneficial to advancing the application of MI-based BCIs. Index Terms — Brain computer interface (BCI), metric learning, multi-source geometric metric transfer learning (MSGMTL), Mahalanobis distance. [ABSTRACT FROM AUTHOR]

Abstract: Activation of the aryl hydrocarbon receptor (AHR) by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) is required for AHR dependent transcriptional activation and TCDD toxicity. We previously reported that aqueous tryptophan exposed to sunlight through window glass (aTRP) contains multiple photoproducts, including the well characterized 6-formylindolo[3,2-b]carbazole (FICZ), capable of activating the AHR and inducing CYP1A and CYP1A-mediated enzyme activities. We report here the isolation from aTRP and chemical characterization and synthesis of 1-(1H-indol-3-yl)-9H-pyrido[3,4-b]indole (IPI), a compound previously identified as a natural product of marine ascidia and now shown to be a TRP photoproduct with AHR-inducing properties. IPI, FICZ and TCDD produced equieffective induction of CYP1A-mediated 7-ethoxyresorufin deethylase (EROD) activity in chick embryo primary hepatocytes and mammalian Hepa1c1c7 cells. EROD induction by IPI was markedly curtailed in AHR-defective c35 cells, supporting the AHR dependence of the IPI response. Although IPI had a higher EC50 for EROD induction than FICZ, the much larger amount of IPI than FICZ in aTRP makes IPI a prominent contributor to EROD induction in aTRP. IPI was detected in TRP-containing culture medium under ambient laboratory conditions but not in TRP-free medium, consistent with its production from TRP. Cotreatment of hepatocytes with submaximal EROD-inducing doses of IPI and FICZ or TCDD produced additive increases in EROD without synergistic or inhibitory interactions. IPI and FICZ were readily metabolized by cultured hepatocytes. In addition to increasing CYP1A4 mRNA and EROD, IPI and FICZ decreased hepatocyte phosphoenolpyruvate carboxykinase mRNA expression and glucose output, biological effects associated with TCDD metabolic dysregulation. The findings underscore a role for sunlight in generating AHR-activating bioactive molecules. [Copyright &y& Elsevier]

We demonstrate the presence of reticuline, an isoquinoline alkaloid that was purified and identified in the rat brain. This was achieved by high-performance liquid chromatography coupled with electrochemical detection. This material was finally identified by nano-electrospray ionization quadrupole time-of-flight tandem mass spectrometry. The expression of this tetrahydroisoquinoline alkaloid in rat brain is at 12.7±5.4 ng/g wet tissue. Furthermore, rat chow, rat small and large intestine and bacteria cultured from these tissues did not contain either morphine or reticuline, eliminating the possibility of contamination or an exogenous source of these compounds. This finding adds information which suggests that morphine biosynthesis may occur in rat neural tissues, and that its biosynthesis pathway may be similar to that reported in the poppy plant. [Copyright &y& Elsevier]

Brain computer interface (BCI) utilizes brain signals to help users interact with external devices directly. EEG is one of the most commonly used techniques for brain signal acquisition in BCI. However, it is notoriously difficult to build a generic EEG recognition model due to significant non-stationarity and subject-to-subject variations, and the requirement for long time training. Transfer learning (TL) is particularly useful because it can alleviate the calibration requirement in EEG-based BCI applications by transferring the calibration information from existing subjects to new subject. To take advantage of geometric properties in Riemann manifold and joint distribution adaptation, a manifold embedded transfer learning (METL) framework was proposed for motor imagery (MI) EEG decoding. First, the covariance matrices of the EEG trials are first aligned on the SPD manifold. Then the features are extracted from both the symmetric positive definite (SPD) manifold and Grassmann manifold. Finally, the classification model is learned by combining the structural risk minimization (SRM) of source domain and joint distribution alignment of source and target domains. Experimental results on two MI EEG datasets verify the effectiveness of the proposed METL. In particular, when there are a small amount of labeled samples in the target domain, METL demonstrated a more accurate and stable classification performance than conventional methods. Compared with several state-of-the-art methods, METL has achieved better classification accuracy, 71.81% and 69.06% in single-to-single (STS), 83.14% and 76.00% in multi-to-single (MTS) transfer tasks, respectively. METL can cope with single source domain or multi-source domains and compared with single-source transfer learning, multi-source transfer learning can improve the performance effectively due to the data expansion. It is effective enough to achieve superior performance for classification of EEG signals. • Not only aligning the marginal distribution, but also the conditional distribution. • Operating the form of vector features in the manifold space to reduce time cost. • Using the structural risk minimization to avoid the over-learning problem. [ABSTRACT FROM AUTHOR]

Muscle-specific kinase (MuSK) is a receptor tyrosine kinase expressed selectively in skeletal muscle. During neuromuscular synapse formation, agrin released from motor neurons stimulates MuSK autophosphorylation in the kinase activation loop and in the juxtamembrane region, leading to clustering of acetylcholine receptors. We have determined the crystal structure of the cytoplasmic domain of unphosphorylated MuSK at 2.05 A˚ resolution. The structure reveals an autoinhibited kinase domain in which the activation loop obstructs ATP and substrate binding. Steady-state kinetic analysis demonstrates that autophosphorylation results in a 200-fold increase in kcat and a 10-fold decrease in the Km for ATP. These studies provide a molecular basis for understanding the regulation of MuSK catalytic activity and suggest that an additional in vivo component may contribute to regulation via the juxtamembrane region. [Copyright &y& Elsevier]