ABSTRACT
Time series forecasting has become a very intensive field of research, which is even increasing in recent years. Deep neural networks have proved to be powerful and are achieving high accuracy in many application fields. For these reasons, they are one of the most widely used methods of machine learning to solve problems dealing with big data nowadays. In this work, the time series forecasting problem is initially formulated along with its mathematical fundamentals. Then, the most common deep learning architectures that are currently being successfully applied to predict time series are described, highlighting their advantages and limitations. Particular attention is given to feed forward networks, recurrent neural networks (including Elman, long-short term memory, gated recurrent units, and bidirectional networks), and convolutional neural networks. Practical aspects, such as the setting of values for hyper-parameters and the choice of the most suitable frameworks, for the successful application of deep learning to time series are also provided and discussed. Several fruitful research fields in which the architectures analyzed have obtained a good performance are reviewed. As a result, research gaps have been identified in the literature for several domains of application, thus expecting to inspire new and better forms of knowledge.
Subject(s)
Deep Learning , Big Data , Forecasting , Machine Learning , Neural Networks, ComputerABSTRACT
Serum transferrin (sTf) plays a pivotal role in regulating iron biodistribution and homeostasis within the body. The molecular details of sTf Fe(III) binding blood transport, and cellular delivery through transferrin receptor-mediated endocytosis are generally well-understood. Emerging interest exists in exploring sTf complexation of nonferric metals as it facilitates the therapeutic potential and toxicity of several of them. This review explores recent X-ray structural and physiologically relevant metal speciation studies to understand how sTf partakes in the bioactivity of key non-redox active hard Lewis acidic metals. It challenges preconceived notions of sTf structure function correlations that were based exclusively on the Fe(III) model by revealing distinct coordination modalities that nonferric metal ions can adopt and different modes of binding to metal-free and Fe(III)-bound sTf that can directly influence how they enter into cells and, ultimately, how they may impact human health. This knowledge informs on biomedical strategies to engineer sTf as a delivery vehicle for metal-based diagnostic and therapeutic agents in the cancer field. It is the intention of this work to open new avenues for characterizing the functionality and medical utility of nonferric-bound sTf and to expand the significance of this protein in the context of bioinorganic chemistry.
ABSTRACT
Titanium is one of the most abundant elements in the earth's crust and while there are many examples of its bioactive properties and use by living organisms, there are few studies that have probed its biochemical reactivity in physiological environments. In the cosmetic industry, TiO2 nanoparticles are widely used. They are often incorporated in sunscreens as inorganic physical sun blockers, taking advantage of their semiconducting property, which facilitates absorbing ultraviolet (UV) radiation. Sunscreens are formulated to protect human skin from the redox activity of the TiO2 nanoparticles (NPs) and are mass-marketed as safe for people and the environment. By closely examining the biological use of TiO2 and the influence of biomolecules on its stability and solubility, we reassess the reactivity of the material in the presence and absence of UV energy. We also consider the alarming impact that TiO2 NP seepage into bodies of water can cause to the environment and aquatic life, and the effect that it can have on human skin and health, in general, especially if it penetrates into the human body and the bloodstream.
