Search

Background Early cancer recurrence after oesophagectomy is a common problem, with an incidence of 20–30 per cent despite the widespread use of neoadjuvant treatment. Quantification of this risk is difficult and existing models perform poorly. This study aimed to develop a predictive model for early recurrence after surgery for oesophageal adenocarcinoma using a large multinational cohort and machine learning approaches. Methods Consecutive patients who underwent oesophagectomy for adenocarcinoma and had neoadjuvant treatment in one Dutch and six UK oesophagogastric units were analysed. Using clinical characteristics and postoperative histopathology, models were generated using elastic net regression (ELR) and the machine learning methods random forest (RF) and extreme gradient boosting (XGB). Finally, a combined (ensemble) model of these was generated. The relative importance of factors to outcome was calculated as a percentage contribution to the model. Results A total of 812 patients were included. The recurrence rate at less than 1 year was 29·1 per cent. All of the models demonstrated good discrimination. Internally validated areas under the receiver operating characteristic (ROC) curve (AUCs) were similar, with the ensemble model performing best (AUC 0·791 for ELR, 0·801 for RF, 0·804 for XGB, 0·805 for ensemble). Performance was similar when internal–external validation was used (validation across sites, AUC 0·804 for ensemble). In the final model, the most important variables were number of positive lymph nodes (25·7 per cent) and lymphovascular invasion (16·9 per cent). Conclusion The model derived using machine learning approaches and an international data set provided excellent performance in quantifying the risk of early recurrence after surgery, and will be useful in prognostication for clinicians and patients., Early recurrence after surgery for adenocarcinoma of the oesophagus is common. A risk prediction model was derived using modern machine learning methods that accurately predicts risk of early recurrence using postoperative pathology. Machine learning may help

The large majority of extragalactic very high energy (VHE; E>100 GeV) sources belongs to the class of active galactic nuclei (AGN), in particular the BL Lac sub-class. AGNs are characterized by an extremely bright and compact emission region, powered by a super-massive black hole (SMBH) and an accretion disk, and relativistic outflows (jets) detected all across the electro-magnetic spectrum. In BL Lac sources the jet axis is oriented close to the line of sight, giving rise to a relativistic boosting of the emission. In radio galaxies, on the other hand, the jet makes a larger angle to the line of sight allowing to resolve the central core and the jet in great details. The giant radio galaxy M 87 with its proximity (1 6Mpc) and its very massive black hole ((3-6) x 10^9 M_solar) provides a unique laboratory to investigate VHE emission in such objects and thereby probe particle acceleration to relativistic energies near SMBH and in jets. M 87 has been established as a VHE emitter since 2005. The VHE emission displays strong variability on time-scales as short as a day. It has been subject of a large joint VHE and multi-wavelength (MWL) monitoring campaign in 2008, where a rise in the 43 GHz VLBA radio emission of the innermost region (core) was found to coincide with a flaring activity at VHE. This had been interpreted as a strong indication that the VHE emission is produced in the direct vicinity of the SMBH black hole. In 2010 again a flare at VHE was detected triggering further MWL observations with the VLBA, Chandra, and other instruments. At the same time M 87 was also observed with the Fermi-LAT telescope at GeV energies and the European VLBI Network (EVN). In this contribution preliminary results from the campaign will be presented., 5 pages, 2 figures, in the proceedings of the "International Workshop on Beamed and Unbeamed Gamma-Rays from Galaxies" 11-15 April 2011, Lapland Hotel Olos, Muonio, Finland, Journal of Physics: Conference Series Volume 355, 2012

One of the key challenges in computational biology is prediction of three-dimensional protein structures from amino-acid sequences. For most proteins, the "native state" lies at the bottom of a free-energy landscape. Protein structure prediction involves varying the degrees of freedom of the protein in a constrained manner until it approaches its native state. In the Rosetta protein structure prediction protocols, a large number of independent folding trajectories are simulated, and several lowest-energy results are likely to be close to the native state. The availability of hundred-teraflop, and shortly, petaflop, computing resources is revolutionizing the approaches available for protein structure prediction. Here, we discuss issues involved in utilizing such machines efficiently with the Rosetta code, including an overview of recent results of the Critical Assessment of Techniques for Protein Structure Prediction 7 (CASP7) in which the computationally demanding structure-refinement process was run on 16 racks of the IBM Blue Gene/L (TM) system at the IBM T. J. Watson Research Center. We highlight recent advances in high-performance computing and discuss,future development paths that make use of the next-generation petascale (> 10(12) floating-point operations per second) machines.

Searches for the pair production of the supersymmetric partner particles of leptons, quarks, and non-minimal Higgs have been made in e + e − annihilations at center-of-mass energies between 50 and 60.8 GeV using the AMY detector at TRISTAN. No evidence for their existence is observed and 95% CL mass limits are presented.

We examine the radio properties of EGRET-detected blazars observed as part of the VLBA Imaging and Polarimetry Survey (VIPS). VIPS has a flux limit roughly an order of magnitude below the MOJAVE survey and most other samples that have been used to study the properties of EGRET blazars. At lower flux levels, radio flux density does not directly correlate with gamma-ray flux density. We do find that the EGRET-detected blazars tend to have higher brightness temperatures, greater core fractions, and possibly larger than average jet opening angles. A weak correlation is also found with jet length and with polarization. All of the well-established trends can be explained by systematically larger Doppler factors in the gamma-ray loud blazars, consistent with the measurements of higher apparent velocities found in monitoring programs carried out at radio frequencies above 10 GHz., 20 pages, 7 figures, accepted to ApJ

We present full polarization Very Long Baseline Array (VLBA) observations at 5 GHz and 15 GHz of 24 compact active galactic nuclei (AGN). These sources were observed as part of a pilot project to demonstrate the feasibility of conducting a large VLBI survey to further our understanding of the physical properties and temporal evolution of AGN jets. The sample is drawn from the Cosmic Lens All-Sky Survey (CLASS) where it overlaps with the Sloan Digital Sky Survey at declinations north of 15 degrees. There are 2100 CLASS sources brighter than 50 mJy at 8.4 GHz, of which we have chosen 24 for this pilot study. All 24 sources were detected and imaged at 5 GHz with a typical dynamic range of 500:1, and 21 of 24 sources were detected and imaged at 15 GHz. Linear polarization was detected in 8 sources at both 5 and 15 GHz, allowing for the creation of Faraday rotation measure (RM) images. The core RMs for the sample were found to have an average absolute value of 390 +/- 100 rad/m^2. We also present the discovery of a new Compact Symmetric Object, J08553+5751. All data were processed automatically using pipelines created or adapted for the survey., 25 pages, 3 figures, 4 tables, accepted to the Astrophysical Journal Supplement Series Full resolution figures and more available from http://www.aoc.nrao.edu/~gtaylor/VIPS

We report on archival data from the Chandra X-ray Observatory for the radio jet of 3C120. We consider the emission process responsible for the X-rays from 4 knots using spectra constructed from radio, optical, and X-ray intensities. While a simple synchrotron model is adequate for three of the knots, the fourth ('k25'), which was previously detected by ROSAT and is now well resolved with Chandra, still represents a problem for the conventional emission processes. If, as we argue, the flat X-ray spectra from two parts of k25 are synchrotron emission, then it appears that either the emission comes from an electron distribution spectrally distinct from that responsible for the radio emission, or at the highest electron energies, there is a significant deviation from the power law describing the electron distribution at lower energies., Accepted for publication in the Astrophysical Journal (tentatively scheduled for the ApJ 1 November 2004, v615 issue). 18 pages including 9 figures. Figs 1, 2, 3, 4, and 8 are jpegs to limit the size of the files. Figs 4 and 8 are in color

Results are presented from long term VLBI monitoring of the parsec-scale radio jet in 3C120, primarily at 5 and 1.7 GHz. Numerous superluminal features are followed, some over large fractional distances from the core. The speeds of the features are not all the same, but vary by less than a factor of 2. The superluminal features extend to core distances of over 0.2 arcseconds and the region of slowing has not been found. The 1.7 GHz images show evidence for stationary features in addition to the superluminal features. The structure of the jet in the vicinity of the most likely stationary feature is suggestive of a helical pattern seen in projection. The deprojected wavelength of the pattern is very long relative to the jet radius, unlike the situation in sources such as M87. If the 3C120 jet does contain a slowly-moving, helical structure, then theory suggests that the jet resides in a relatively cool medium, not in a relativistically hot cocoon or lobe., 35 pages with 8 postscript figures. Accepted for publication in the Astrophysical Journal

We report on measurements of differential $\mu - {\bar b}$ cross sections, where the muon is from a semi-leptonic $b$ decay and the ${\bar b}$ is identified using precision track reconstruction in jets. The semi-differential correlated cross sections, d$\sigma$/d$\Et^{{\bar b}}$, d$\sigma$/d$\pt^{{\bar b}}$, and d$\sigma$/d$\delta\phi(\mu - {\bar b})$ for $\pt^{\mu}>$~9 GeV/c, $|\eta^{\mu}| $~10 GeV, $|\eta^{{\bar b}}

Superluminal motions were first seen in 3C 120 between 1972.5 and 1974.4 (Seielstad et al. 1979 and references therein). Between 1975 and 1980, the source was monitored along with 3C 273 and 3C 345 by the Caltech group. One reasonably clear episode of expansion was seen in 1979 (Walker et al. 1982) but, for most of the time, the source evolution was so rapid that it was difficult to relate the structures seen at successive epochs.

We report results of an electron-beam-dump search for neutral particles with masses in the range 1 to 15 MeV and lifetimes τ between 10^-14 and 10^-10 s. No evidence was found for such an object. We fule out the existence of any 1.8-MeV pseudoscalar boson with τ>8.2×10^-15 s and an absorption cross section in matter less than 1 mb per nucleon, and exclude τ>1×10^-14 s were its cross section to equal 50 mb per nucleon. In conjunction with measurements of the electron’s anomalous magnetic moment, this experiment shows that the narrow positron peaks observed in heavy-ion collisions at the Gessellschaft fur Schwerionenforschung are not due to an elementary pseudoscalar.