Special issue on distributed computing and artificial intelligence

4:1! Google’s artificial intelligence (AI) program, AlphaGo, has won Go Master Lee Sedol in a best-of-five competition held in Korean March 9?15, 2016. Seen by many as a landmark moment for AI, the outcome did not come as a surprise, considering the excellent combination of 1920 CPUs with sophisticated AI algorithms, including neural networks and Monte Carlo tree search (Gibney, 2016; Silver et al., 2016). Indeed, research on distributed computing and artificial intelligence (DCAI) has matured during the last decade and many effective applications are now deployed, performing an increasingly important role in modern computer science, including the two most hyped technologies: Internet of Things and Big Data. Indeed, it is fair to say that the application of artificial intelligence in distributed environments is becoming an essential element of high added value and economic potential.     

Dynamic modeling of a dual active bridge DC to DC converter with average current control and load‐current feed‐forward

Bidirectional power flow is needed in many power conversion systems like energy storage systems, regeneration systems, power converters for improvement of the power quality and some DC‐DC applications where bidirectional high power conversion and galvanic isolation are required. The dual active bridge (DAB) is an isolated, high voltage ratio DC‐DC converter suitable for high power density and high power applications, being a key interface between renewable energy sources and energy storage devices. This paper is focused on the modeling and control design of a DC‐DC system with battery storage based on a DAB converter with average current mode control of the output current and output voltage control. The dynamic response of the output voltage to load steps is improved by means of an additional load‐current feed‐forward control loop. An analytical study of the load‐current feed‐forward is presented and validated by means of both simulations and experimental results. Copyright © 2014 John Wiley & Sons, Ltd.     

BETaaS: A Platform for Development and Execution of Machine-to-Machine Applications in the Internet of Things

Wireless Personal Communications - The integration of everyday objects into the Internet represents the foundation of the forthcoming Internet of Things (IoT). Smart objects will be the building...     

Critical Review on Fatty Acid-Based Food and Nutraceuticals as Supporting Therapy in Cancer

Fatty acids have an important place in both biological and nutritional contexts and, from a clinical point of view, they have known consequences for diseases’ onset and development, including cancer. The use of fatty acid-based food and nutraceuticals to support cancer therapy is a multidisciplinary subject, involving molecular and clinical research. Knowledge regarding polyunsaturated fatty acids essentiality/oxidizability and the role of lipogenesis-desaturase pathways for cell growth, as well as oxidative reactivity in cancer cells, are discussed, since they can drive the choice of fatty acids using their multiple roles to support antitumoral drug activity. The central role of membrane fatty acid composition is highlighted for the application of membrane lipid therapy. As fatty acids are also known as biomarkers of cancer onset and progression, the personalization of the fatty acid-based therapy is also possible, taking into account other important factors such as formulation, bioavailability and the distribution of the supplementation. A holistic approach emerges combining nutra- and pharma-strategies in an appropriate manner, to develop further knowledge and applications in cancer therapy.     

Effective Natural Killer Cell Degranulation Is an Essential Key in COVID-19 Evolution

NK degranulation plays an important role in the cytotoxic activity of innate immunity in the clearance of intracellular infections and is an important factor in the outcome of the disease. This work has studied NK degranulation and innate immunological profiles and functionalities in COVID-19 patients and its association with the severity of the disease. A prospective observational study with 99 COVID-19 patients was conducted. Patients were grouped according to hospital requirements and severity. Innate immune cell subpopulations and functionalities were analyzed. The profile and functionality of innate immune cells differ between healthy controls and severe patients; CD56dim NK cells increased and MAIT cells and NK degranulation rates decreased in the COVID-19 subjects. Higher degranulation rates were observed in the non-severe patients and in the healthy controls compared to the severe patients. Benign forms of the disease had a higher granzymeA/granzymeB ratio than complex forms. In a multivariate analysis, the degranulation capacity resulted in a protective factor against severe forms of the disease (OR: 0.86), whereas the permanent expression of NKG2D in NKT cells was an independent risk factor (OR: 3.81; AUC: 0.84). In conclusion, a prompt and efficient degranulation functionality in the early stages of infection could be used as a tool to identify patients who will have a better evolution.     

Profile of TREM2-Derived circRNA and mRNA Variants in the Entorhinal Cortex of Alzheimer’s Disease Patients

Genetic variants in TREM2, a microglia-related gene, are well-known risk factors for Alzheimer’s disease (AD). Here, we report that TREM2 originates from circular RNAs (circRNAs), a novel class of non-coding RNAs characterized by a covalent and stable closed-loop structure. First, divergent primers were designed to amplify circRNAs by RT-PCR, which were further assessed by Sanger sequencing. Then, additional primer sets were used to confirm back-splicing junctions. In addition, HMC3 cells were used to assess the microglial expression of circTREM2s. Three candidate circTREM2s were identified in control and AD human entorhinal samples. One of the circRNAs, circTREM2_1, was consistently amplified by all divergent primer sets in control and AD entorhinal cortex samples as well as in HMC3 cells. In AD cases, a moderate negative correlation (r = −0.434) was found between the global average area of Aβ deposits in the entorhinal cortex and circTREM2_1 expression level. In addition, by bioinformatics tools, a total of 16 miRNAs were predicted to join with circTREM2s. Finally, TREM2 mRNA corresponding to four isoforms was profiled by RT-qPCR. TREM2 mRNA levels were found elevated in entorhinal samples of AD patients with low or intermediate ABC scores compared to controls. To sum up, a novel circRNA derived from the TREM2 gene, circTREM2_1, has been identified in the human entorhinal cortex and TREM2 mRNA expression has been detected to increase in AD compared to controls. Unraveling the molecular genetics of the TREM2 gene may help to better know the innate immune response in AD.     

An adaptive tracking algorithm for direction finding and array shape estimation in a nonstationary environment

This paper presents an adaptive tracking algorithm for the subspace-based direction of arrival estimation of multiple sources in a nonstationary, environment. The nonstationarities are due to moving sources or to timevarying distortions of the sensor array shape. The proposed algorithm relies on the properties of a linear operator, referred to as the Propagator, which only exploits the linear independency of the source steering vectors. The Propagator allows not only the calibration of the array shape, but also the determination of the source and the noise subspaces without any eigendecomposition of the cross-spectral matrix of the received signals. A gradientbased adaptive algorithm is here proposed for the on-line estimation of the Propagator. A theoretical analysis of the behaviour of this algorithm in a nonstationary environment is given. Simulations are carried out in the case of moving sources and in the case of a time-varying array shape. They exhibit the good performances of the proposed algorithm.     

Alchemical Design of Pharmacological Chaperones with Higher Affinity for Phenylalanine Hydroxylase

Phenylketonuria (PKU) is a rare metabolic disease caused by variations in a human gene, PAH, encoding phenylalanine hydroxylase (PAH), and the enzyme converting the essential amino acid phenylalanine into tyrosine. Many PKU-causing variations compromise the conformational stability of the encoded enzyme, decreasing or abolishing its catalytic activity, and leading to an elevated concentration of phenylalanine in the blood, which is neurotoxic. Several therapeutic approaches have been developed to treat the more severe manifestations of the disorder, but they are either not entirely effective or difficult to adhere to throughout life. In a search for novel pharmacological chaperones to treat PKU, a lead compound was discovered (compound IV) that exhibited promising in vitro and in vivo chaperoning activity on PAH. The structure of the PAH-IV complex has been reported. Here, using alchemical free energy calculations (AFEC) on the structure of the PAH-IV complex, we design a new generation of compound IV-analogues with a higher affinity for the enzyme. Seventeen novel analogues were synthesized, and thermal shift and isothermal titration calorimetry (ITC) assays were performed to experimentally evaluate their stabilizing effect and their affinity for the enzyme. Most of the new derivatives bind to PAH tighter than lead compound IV and induce a greater thermostabilization of the enzyme upon binding. Importantly, the correspondence between the calculated alchemical binding free energies and the experimentally determined ΔΔG_b values is excellent, which supports the use of AFEC to design pharmacological chaperones to treat PKU using the X-ray structure of their complexes with the target PAH enzyme.     

Impact of the radio channel modelling on the performance of?VANET communication protocols

The expected traffic safety and efficiency benefits that can be achieved through the development and deployment of vehicular ad-hoc networks has attracted a significant interest from the networking research community that is currently working on novel vehicular communication protocols. The time-critical nature of vehicular applications and their reliability constraints require a careful protocol design and dimensioning. To this aim, adequate and accurate models should be employed in any research study. One of the critical aspects of any wireless communications system is the radio channel propagation. This is particularly the case in vehicular networks due to their low antenna heights, the fast topology changes and the reliability and latency constraints of traffic safety applications. Despite the research efforts to model the vehicle-to-vehicle communications channel, many networking studies are currently simplifying and even neglecting the radio channel effects on the performance and operation of their protocols. As this work demonstrates, it is critical that realistic and accurate channel models are employed to adequately understand, design and optimize novel vehicular communications and networking protocols.     

Analyzing probabilistic pushdown automata

The paper gives a summary of the existing results about algorithmic analysis of probabilistic pushdown automata and their subclasses.