Highlights

Every year, a committee of experts sits down with a tough job to do: from among all ICREA publications, they must find a handful that stand out from all the others. This is indeed a challenge. The debates are sometimes heated and always difficult but, in the end, a shortlist of  the most outstanding publications of the year is produced. No prize is awarded, and the only additional acknowledge is the honour of being chosen and highlighted by ICREA. Each piece has something unique about it, whether it be a particularly elegant solution, the huge impact it has in the media or the sheer fascination it generates as a truly new idea. For whatever the reason, these are the best of the best and, as such, we are proud to share them here.

LIST OF SCIENTIFIC HIGHLIGHTS

Format: yyyy
  • Dark Galaxies: finally found exactly where we predicted (2020)

    Jiménez Tellado, Raúl (UB)

    view details
    CLOSE

    Dark Galaxies: finally found exactly where we predicted

    The existence of dark galaxies, those that should have none or negligible starlight, was predicted by us almost 25 years ago. At that time, none of these dark galaxies had been found and its existence reminded very doubtful as it seemed very difficult to stop the processes that trigger star formation. However, blind surveys in the radio, at rest wavelengths of 21cm, that search for the forbidden spin flip transtion of the hidrogen atom, have recently demonstrated that these galaxies do exist.They contain only hydrogen and almost no visible stars. In fact, their abundance is in perfect agreement with our theoretical predictions. Not only that, we have shown that their physical origin is the one proposed by us: the high spin of their host dark matter halo makes the baryons settle into a disk that is too difusse as to allow star formation to proceed as it does in our Milky Way. This origin is intimately related to the nature of dark matter, as primordial tidal forces in the hierarchical cold dark matter model (LCDM) do predict the right abundance for these galaxies. In this respect, dark galaxies are a confirmation of the current cosmological model.

  • Evidence of spectacular four-top-quark production at the LHC (2020)

    Juste, Aurelio (IFAE)

    view details
    CLOSE

    Evidence of spectacular four-top-quark production at the LHC

    During 2015-2018, the Large Hadron Collider (LHC) at CERN collided protons at a center-of-mass energy of 13 TeV, the highest energy ever reached by a particle accelerator.  One of the main goals of the ATLAS experiment at the LHC is to challenge the predictions of the Standard Model (SM), our most successful theory of elementary particles. To this end, a promising direction is the study of the top quark, the heaviest elementary particle known, with a mass close to that of a gold atom.

    The production of two top quarks and two antitop quarks (“four-top-quark” production) is a very rare process in the SM, happening only once every 1012 collisions. However, new particles beyond the SM can significantly enhance this rate. Once produced, each top quark decays into a W boson and a bottom quark, with the W boson decaying into a charged lepton (electron, muon, or tau) and a neutrino, or a quark-antiquark pair. This results in some of the most spectacular signatures ever produced at the LHC.

    The ATLAS Collaboration has recently reported strong evidence for the production of four top quarks a milestone reached by studying events with two same-charge leptons or three leptons, plus additional jets originating from the bottom quarks. The signal significance amounts to 4.3 standard deviations (s.d.), for an expected significance of 2.4 s.d. in the SM. This means that the measured rate is somewhat above the SM prediction, although still consistent with it within 1.7 s.d.

    Since 2015, researchers at IFAE, under A. Juste’s leadership, are playing a major role in the search for four-top-quark production in ATLAS. The team has not only contributed to the recent result, but is also completing a search for this process in a complementary channel featuring only one lepton or two opposite-charge leptons. The combination of both searches is expected to yield the observation of this process. Additional data from the next LHC run, to start in 2022, along with further developments in the analysis techniques, will improve the precision of this challenging measurement, and hopefully allow drawing definite conclusions on whether the breakdown of the SM is finally in sight.

  • Novel approach of using Unsupervised Machine Learning in Physics (2020)

    Lewenstein, Maciej Andrzej (ICFO)
    Acín Dal Maschio, Antonio (ICFO)

    view details
    CLOSE

    Novel approach of using Unsupervised Machine Learning in Physics

    A team of ICFO researchers reports in PRL an entirely new anomaly detection method capable of training a system in very few iterations. Machine Learning (ML) has the main goal of analyzing and interpreting data structures and patterns in order to learn from them, reason and carry out a decision-making task that is completely independent from human reasoning and engagement. Even though this field of study started in the mid 1900s, recent developments in the area have revolutionized the way on how we can process and find correlations in complex data.

    Contrary to supervised learning, unsupervised learning seeks to discover patterns or classify information in large data sets into categories without prior knowledge. That is, it does not have labeled outputs, which means that it basically infers the natural structure that a dataset may have and extracts categorized information from it. This learning has proved to be very efficient for identifying phases and phase transitions of many-body systems. In a study recently published in Physical Review Letters, ICFO researchers Korbinian Kottmann and Patrick Huembeli, led by ICREA Professors at ICFO Antonio Acín and Maciej Lewenstein report on a method that uses an unsupervised machine learning technique based on anomaly detection to automatically map out the phase diagram of a quantum many body system given unlabeled data.

    The following example is very illustrative of what they have achieved. In machine learning the most common and known classification task example is to discriminate, for instance, images of cats and dogs. In the study, the method the researchers use anomaly detection, which handles the classification task of discriminating dogs and everything that is not a dog, approaching the system is an entirely different perspective. The idea is to train a special neural network called an autoencoder to efficiently compress and reproduce images of dogs. If the network is later fed with images of cats, the network does not know how to efficiently compress the features of the cat image and it is possible to tell from the higher reconstruction loss that it is not a dog.

     

  • A compendium of mutational cancer driver genes (2020)

    López-Bigas, Núria (IRB Barcelona)

    view details
    CLOSE

    A compendium of mutational cancer driver genes

    Cancer is a group of diseases characterised by uncontrolled cell growth caused by mutations, and other alterations in the genome of cells. A tumour can present from hundreds to thousands of mutations, but only a few are vital for its tumorigenic capacity. These key mutations affect the function of cancer driver genes. Finding the genes that harbour this cancer driver mutations is one of the main goals in cancer research.

    Since cancer driver genes are under positive selection in tumorigenesis, identifying signals of positive selection in the patterns of somatic mutations across tumors is an effective way to identify cancer genes. We have implemented a systematic approach combining several of these signals to generate a compendium of mutational cancer genes. Its application to somatic mutations of more than 28,000 tumours of 66 cancer types revealed 568 cancer genes and points towards their mechanisms of tumorigenesis. The application of this approach to the ever-growing datasets of somatic tumour mutations will support the continuous refinement of our knowledge of the genetic basis of cancer.

    All the results are available at http://www.intogen.org

     

  • Fusion plasma experiments with controlled variations of the fast ion distribution on ASDEX Upgrade tokamak (2020)

    Mantsinen, Mervi Johanna (BSC-CNS)

    view details
    CLOSE

    Fusion plasma experiments with controlled variations of the fast ion distribution on ASDEX Upgrade tokamak

    Ion Cyclotron Resonance Frequency (ICRF) heating plays an important role in many present day fusion experiments and it is one of the auxiliary heating methods that will be used in the next step device ITER. On the ASDEX Upgrade tokamak, novel applications of ICRF waves for plasma heating have become recently possible through the improved operating space of ICRF system and, in particular, its extended frequency range [1]. They have been instrumental for the experiments using third harmonic ICRF heating of neutral beam injected deuterons for fast ion studies and for further development of fast ion and neutron diagnostics. Figure shows a typical discharge with a more than two-fold increase of the neutron rate due to deuterium-deuterium fusion reactions involving ICRF-accelerated deuterons that has been achieved with this scheme in AUG. These developments have provided for the first time a means for simultaneous controlled variations and measurements of both the confined and the non-confined parts of ICRF-driven fast deuterium distribution on AUG.

  • New genome mapper is like “upgrading from dial-up to fibre-optic” (2020)

    Martí-Renom, Marc (CRG)

    view details
    CLOSE

    New genome mapper is like “upgrading from dial-up to fibre-optic”

    Researchers from Harvard university, the Centro Nacional de Análisis Genómico (CNAG) and the Centre for Genomic Regulation (CRG), describe the first technology able to visualize hundreds to potentially thousands of genes at the same time under the microscope. The tech images genomes more cheaply, more quickly and increases range of visibility compared to currently available methods. The technique was described in Nature Methods.

    Each human cell has two metres of genome condensed down into 10 microns within the cell nucleus. This blueprint of life folds to help genes make physical contact with other genes that may be located quite a distance away along the chromosome. This three-dimensional organisation is crucial for cell function, but its complexity and constant dynamism make it incredibly difficult to visualize. Imaging more than a handful of genes at the same time has been impossible, limiting researchers’ ability to characterize how genomes function. We developed OligoFISSEQ, a technology using new computational methods that overcomes these current limitations.