A neural network for the detection of soccer headers from wearable sensor data.

To investigate the proposed association between soccer heading and deleterious brain changes, an accurate quantification of heading exposure is crucial. While wearable sensors constitute a popular means for this task, available systems typically overestimate the number of headers by poorly discriminating true impacts from spurious recordings. This study investigated the utility of a neural network for automatically detecting soccer headers from kinematic time series data obtained by wearable sensors. During 26 matches, 27 female soccer players wore head impacts sensors to register on-field impact events (> 8 g), which were labelled as valid headers (VH) or non-headers (NH) upon video review. Of these ground truth data, subsets of 49% and 21% each were used to train and validate a Long Short-Term Memory (LSTM) neural network in order to classify sensor recordings as either VH or NH based on their characteristic linear acceleration features. When tested on a balanced dataset comprising 271 VHs and NHs (which corresponds to 30% and 1.4% of ground truth VHs and NHs, respectively), the network showed very good overall classification performance by reaching scores of more than 90% across all metrics. When testing was performed on an unbalanced dataset comprising 271 VHs and 5743 NHs (i.e., 30% of ground truth VHs and NHs, respectively), as typically obtained in real-life settings, the model still achieved over 90% sensitivity and specificity, but only 42% precision, which would result in an overestimation of soccer players' true heading exposure. Although classification performance suffered from the considerable class imbalance between actual headers and non-headers, this study demonstrates the general ability of a data-driven deep learning network to automatically classify soccer headers based on their linear acceleration profiles.

Bose-Einstein condensation of quasiparticles by rapid cooling.

The fundamental phenomenon of Bose-Einstein condensation has been observed in different systems of real particles and quasiparticles. The condensation of real particles is achieved through a major reduction in temperature, while for quasiparticles, a mechanism of external injection of bosons by irradiation is required. Here, we present a new and universal approach to enable Bose-Einstein condensation of quasiparticles and to corroborate it experimentally by using magnons as the Bose-particle model system. The critical point to this approach is the introduction of a disequilibrium of magnons with the phonon bath. After heating to an elevated temperature, a sudden decrease in the temperature of the phonons, which is approximately instant on the time scales of the magnon system, results in a large excess of incoherent magnons. The consequent spectral redistribution of these magnons triggers the Bose-Einstein condensation.

Thermal behavior of MOCVD-grown Cu-clusters on ZnO(1010).

Scanning tunnelling microscopy (STM) and X-ray photoelectron spectroscopy (XPS, AES) were used to study MOCVD of Cu-clusters on the mixed terminated ZnO(1010) surface in comparison to MBE Cu-deposition. Both deposition methods result in the same Cu cluster morphology. After annealing to 670 K the amount of Cu visible above the oxide surface is found to decrease substantially, indicating a substantial diffusion of Cu atoms inside the ZnO-bulk. The spectroscopic data do not show any evidence for changes in the Cu oxidation state during thermal treatment up to 770 K.

Adsorption of atomic hydrogen on ZnO(1010): STM study.

The adsorption of atomic hydrogen on a single crystal ZnO(1010) surface has been studied by scanning tunneling microscopy (STM) under ultrahigh vacuum conditions at room temperature and at elevated temperatures. High resolution STM images indicate that a well-ordered (1x1) H adlayer is formed on the ZnO(1010) surface. The STM data strongly indicate that the hydrogen adsorbs on top of the oxygen atoms forming hydroxyl species. Scanning tunneling spectroscopy (STS) studies reveal a H atom induced metallization at room temperature. In contrast to the clean surface for the hydrogen-covered surface distinct defects structures consisting of missing O and Zn atoms could be identified.

Distinct BIR domains of cIAP1 mediate binding to and ubiquitination of tumor necrosis factor receptor-associated factor 2 and second mitochondrial activator of caspases.

Inhibitor of apoptosis proteins (IAPs) regulate apoptosis primarily by inhibiting caspase-family proteases. However, many IAPs also possess E3 ligase (ubiquitin-protein isopeptide ligase) activities implicated in both caspase-dependent and -independent functions of these proteins. Here, we compared the structural features of cIAP1 responsible for its interactions with two known target proteins, TRAF2 and SMAC. The N-terminal (BIR1) and C-terminal (BIR3) BIR domains of cIAP1 were determined to be necessary and sufficient for binding TRAF2 and SMAC, respectively. Mutational analysis of the BIR1 and BIR3 domains identified critical residues required for TRAF2 and SMAC binding. Using these mutants, cIAP1-mediated ubiquitination of TRAF2 and SMAC in vitro was determined to be correspondingly dependent on intact binding sites on BIR1 and BIR3. Because TRAF2 regulates NF-kappaB activation, the effects of cIAP1 on TRAF2-mediated induction of NF-kappaB transcriptional activity were studied using reporter gene assays. Expression of a fragment of cIAP1 encompassing the three BIR domains (but not full-length cIAP1) greatly enhanced TRAF2-induced increases in NF-kappaB activity, providing a convenient assay for monitoring BIR-dependent effects of cIAP1 on TRAF2 in cells. BIR1 mutants of the BIR1-3 fragment of cIAP1 that failed to bind TRAF2 lost the ability to modulate NF-kappaB activity, demonstrating a requirement for BIR1-mediated interactions with TRAF2. Altogether, these findings demonstrate the modularity and diversification of BIR domains, showing that a single cIAP can direct its E3 ligase activity toward different substrates and can alter the cellular functions of different protein targets in accordance with differences in the specificity of individual BIR domains.