Search

This article presents an investigation of isospin equilibration in cross-bombarding $^{40,48}$Ca$+^{40,48}$Ca reactions at 35 MeV/nucleon, by comparing experimental data with filtered transport model calculations. Isospin diffusion is studied using the evolution of the isospin transport ratio with centrality. The asymmetry parameter $\delta=(N-Z)/A$ of the quasiprojectile (QP) residue is used as isospin-sensitive observable, while a recent method for impact parameter reconstruction is used for centrality sorting. A benchmark of global observables is proposed to assess the relevance of the antisymmetrized molecular dynamics (AMD) model, coupled to GEMINI++, in the study of dissipative collisions. Our results demonstrate the importance of considering cluster formation to reproduce observables used for isospin transport and centrality studies. Within the AMD model, we prove the applicability of the impact parameter reconstruction method, enabling a direct comparison to the experimental data for the investigation of isospin diffusion. For both, we evidence a tendency to isospin equilibration with an impact parameter decreasing from 9 to 3 fm, while the full equilibration is not reached. A weak sensitivity to the stiffness of the equation of state employed in the model is also observed, with a better reproduction of the experimental trend for the neutron-rich reactions.

An important part of breast cancer staging is the assessment of the sentinel axillary node for early signs of tumor spreading. However, this assessment by pathologists is not always easy and retrospective surveys often requalify the status of a high proportion of sentinel nodes. Convolutional Neural Networks (CNNs) are a class of deep learning algorithms that have shown excellent performances in the most challenging visual classification tasks, with numerous applications in medical imaging. In this study I compare twelve different CNNs and different hardware acceleration devices for the detection of breast cancer from microscopic images of breast cancer tissue. Convolutional models are trained and tested on two public datasets. The first one is composed of more than 300,000 images of sentinel lymph node tissue from breast cancer patients, while the second one has more than 220,000 images from inductive breast carcinoma tissue, one of the most common forms of breast cancer. Four different hardware acceleration cards were used, with an off-the-shelf deep learning framework. The impact of transfer learning and hyperparameters fine-tuning are tested. Hardware acceleration device performance can improve training time by a factor of five to twelve, depending on the model used. On the other hand, increasing convolutional depth will augment the training time by a factor of four to six times, depending on the acceleration device used. Increasing the depth and the complexity of the model generally improves performance, but the relationship is not linear and also depends on the architecture of the model. The performance of transfer learning is always worse compared to a complete retraining of the model. Fine-tuning the hyperparameters of the model improves the results, with the best model showing a performance comparable to state-of-the-art models., Comment: 13 pages, 4 figures, 4 tables; corrected typos; added an additional breast carcinoma image dataset; added a total of twelve CNN models tested. Additional testing for transfer learning and complexity of the models

The equation of state with light clusters for nuclear and stellar matter is determined using chemical equilibrium constants evaluated from the analysis of the recently published (Xe$+$Sn) heavy ion data, corresponding to three reactions with different isotopic contents of the emission source. The measured multiplicities are used to extract the thermodynamic properties, and an in-medium correction to the ideal gas internal partition function of the clusters is included in the analysis. This in-medium correction and its respective uncertainty are calculated via a Bayesian analysis, with the unique hypothesis that the different nuclear species in a given sample must correspond to a unique common value for the density of the expanding source. Different parameter sets for the correction are tested, and the effect of the radius of the clusters on the thermodynamics and on the chemical equilibrium constants is also addressed. It is shown that the equilibrium constants obtained are almost independent of the isospin content of the analysed systems. Finally, a comparison with a relativistic mean field model proves that data are consistent with a universal in-medium correction of the scalar $\sigma$-meson coupling for nucleons bound in clusters. The obtained value, $g_s/g_s^0 = 0.92 \pm 0.02$, is larger than that obtained in a previous study not including in-medium effects in the data analysis. This result implies a smaller effect on the binding energy of the clusters and, as a consequence, larger melting densities, and an increased cluster contribution in supernova matter., Comment: 29 pages, 16 figures, accepted for publication in J. Phys. G

The modification of the ground state properties of light atomic nuclei in the nuclear and stellar medium is addressed, using chemical equilibrium constants evaluated from a new analysis of the intermediate energy heavy-ion (Xe$+$Sn) collision data measured by the INDRA collaboration. Three different reactions are considered, mainly differing by the isotopic content of the emission source. The thermodynamic conditions of the data samples are extracted from the measured multiplicities allowing for a parametrization of the in-medium modification, determined with the single hypothesis that the different nuclear species in a given sample correspond to a unique common value for the density of the expanding source. We show that this correction, which was not considered in previous analyses of chemical constants from heavy ion collisions, is necessary, since the observables of the analyzed systems show strong deviations from the expected results for an ideal gas of free clusters. This data set is further compared to a relativistic mean-field model, and seen to be reasonably compatible with a universal correction of the attractive $\sigma$-meson coupling., Comment: 5 pages, 2 figures, 1 table, accepted for publication in Phys. Rev. Lett

Module network inference is a statistical method to reconstruct gene regulatory networks, which uses probabilistic graphical models to learn modules of coregulated genes and their upstream regulatory programs from genome-wide gene expression and other omics data. Here we review the basic theory of module network inference, present protocols for common gene regulatory network reconstruction scenarios based on the Lemon-Tree software, and show, using human gene expression data, how the software can also be applied to learn differential module networks across multiple experimental conditions., Comment: Minor revision; 19 pages, 5 figures; chapter for a forthcoming book on gene regulatory network inference

Module network inference is an established statistical method to reconstruct co-expression modules and their upstream regulatory programs from integrated multi-omics datasets measuring the activity levels of various cellular components across different individuals, experimental conditions or time points of a dynamic process. We have developed Lemon-Tree, an open-source, platform-independent, modular, extensible software package implementing state-of-the-art ensemble methods for module network inference. We benchmarked Lemon-Tree using large-scale tumor datasets and showed that Lemon-Tree algorithms compare favorably with state-of-the-art module network inference software. We also analyzed a large dataset of somatic copy-number alterations and gene expression levels measured in glioblastoma samples from The Cancer Genome Atlas and found that Lemon-Tree correctly identifies known glioblastoma oncogenes and tumor suppressors as master regulators in the inferred module network. Novel candidate driver genes predicted by Lemon-Tree were validated using tumor pathway and survival analyses. Lemon-Tree is available from http://lemon-tree.googlecode.com under the GNU General Public License version 2.0., Comment: minor revision; 13 pages text + 4 figures + 4 tables + 4 pages supplementary methods; supplementary tables available from the authors

We present an analysis of multifragmentation events observed in central Xe+Sn reactions at Fermi energies. Performing a comparison between the predictions of the Stochastic Mean Field (SMF) transport model and experimental data, we investigate the impact of the compression-expansion dynamics on the properties of the final reaction products. We show that the amount of radial collective expansion, which characterizes the dynamical stage of the reaction, influences directly the onset of multifragmentation and the kinematic properties of multifragmentation events. For the same set of events we also undertake a shape analysis in momentum space, looking at the degree of stopping reached in the collision, as proposed in recent experimental studies. We show that full stopping is achieved for the most central collisions at Fermi energies. However, considering the same central event selection as in the experimental data, we observe a similar behavior of the stopping power with the beam energy, which can be associated with a change of the fragmentation mechanism, from statistical to prompt fragment emission., Comment: 15 pages

In 129 Xe+ nat Sn central collisions from 8 to 25 MeV/A, the three-fragment exit channel occurs with a significant cross section. We show that these fragments arise from two successive binary splittings of a heavy composite system. The sequence of fragment production is determined. Strong Coulomb proximity effects are observed in the three-fragment final state. A comparison with Coulomb trajec-tory calculations shows that the time scale between the consecutive break-ups decreases with increasing bombarding energy, becoming quasi-simultaneous above excitation energy E * = 4.0$\pm$0.5 MeV/A. This transition from sequential to simultaneous break-up was interpreted as the signature of the onset of multifragmentation for the three-fragment exit channel in this system., Comment: 12 pages; 13 Figures; 4 Table; Accepted for publication in Physical Review C

Peripheral collisions give access to a set of events where hot quasi-projectile (QP) sources are produced exploring a large range of excitation energy. In this range evaporation and multifragmentation are both observed for a similar charge/size of the sources. In this work, this experimental fact will be described in terms of coexistence of two classes of events in the first order phase transition formalism. For this, a detailed study of the experimental correlation between the size/charge of the biggest cluster/fragment (Z1) and the excitation energy (E*) is made. Making the parallel with the same correlation derived from the canonical ensemble description, a first value of the latent heat and boundaries of spinodal and coexistence zone are extracted., Comment: 7 pages

The charge distribution of the heaviest fragment detected in the decay of quasi-projectiles produced in intermediate energy heavy-ion collisions has been observed to be bimodal. This feature is expected as a generic signal of phase transition in non-extensive systems. In this paper we present new analyses of experimental data from Au on Au collisions at 60, 80 and 100 MeV/nucleon showing that bimodality is largely independent of the data selection procedure, and of entrance channel effects. An estimate of the latent heat of the transition is extracted., Comment: 5 pages, submitted to PRL

Fragment properties of hot fragmenting sources of similar sizes produced in central and semi-peripheral collisions are compared in the excitation energy range 5-10 AMeV. For semi-peripheral collisions a method for selecting compact quasi-projectiles sources in velocity space similar to those of fused systems (central collisions) is proposed. The two major results are related to collective energy. The weak radial collective energy observed for quasi-projectile sources is shown to originate from thermal pressure only. The larger fragment multiplicity observed for fused systems and their more symmetric fragmentation are related to the extra radial collective energy due to expansion following a compression phase during central collisions. A first attempt to locate where the different sources break in the phase diagram is proposed., Comment: 23 pages submitted to NPA

In recent years, several authors have used probabilistic graphical models to learn expression modules and their regulatory programs from gene expression data. Here, we demonstrate the use of the synthetic data generator SynTReN for the purpose of testing and comparing module network learning algorithms. We introduce a software package for learning module networks, called LeMoNe, which incorporates a novel strategy for learning regulatory programs. Novelties include the use of a bottom-up Bayesian hierarchical clustering to construct the regulatory programs, and the use of a conditional entropy measure to assign regulators to the regulation program nodes. Using SynTReN data, we test the performance of LeMoNe in a completely controlled situation and assess the effect of the methodological changes we made with respect to an existing software package, namely Genomica. Additionally, we assess the effect of various parameters, such as the size of the data set and the amount of noise, on the inference performance. Overall, application of Genomica and LeMoNe to simulated data sets gave comparable results. However, LeMoNe offers some advantages, one of them being that the learning process is considerably faster for larger data sets. Additionally, we show that the location of the regulators in the LeMoNe regulation programs and their conditional entropy may be used to prioritize regulators for functional validation, and that the combination of the bottom-up clustering strategy with the conditional entropy-based assignment of regulators improves the handling of missing or hidden regulators., Comment: 13 pages, 6 figures + 2 pages, 2 figures supplementary information

In the canonical formalism of statistical physics, a signature of a first order phase transition for finite systems is the bimodal distribution of an order parameter. Previous thermodynamical studies of nuclear sources produced in heavy-ion collisions provide information which support the existence of a phase transition in those finite nuclear systems. Some results suggest that the observable Z1 (charge of the biggest fragment) can be considered as a reliable order parameter of the transition. This talk will show how from peripheral collisions studied with the INDRA detector at GSI we can obtain this bimodal behaviour of Z1. Getting rid of the entrance channel effects and under the constraint of an equiprobable distribution of excitation energy (E*), we use the canonical description of a phase transition to link this bimodal behaviour with the residual convexity of the entropy. Theoretical (with and without phase transition) and experimental Z1-E* correlations are compared. This comparison allows us to rule out the case without transition. Moreover that quantitative comparison provides us with information about the coexistence region in the Z1-E* plane which is in good agreement with that obtained with the signal of abnormal uctuations of configurational energy (microcanonical negative heat capacity)., Comment: 8 pages