48
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Deep Artificial Neural Networks and Neuromorphic Chips for Big Data Analysis: Pharmaceutical and Bioinformatics Applications

      review-article

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Over the past decade, Deep Artificial Neural Networks (DNNs) have become the state-of-the-art algorithms in Machine Learning (ML), speech recognition, computer vision, natural language processing and many other tasks. This was made possible by the advancement in Big Data, Deep Learning (DL) and drastically increased chip processing abilities, especially general-purpose graphical processing units (GPGPUs). All this has created a growing interest in making the most of the potential offered by DNNs in almost every field. An overview of the main architectures of DNNs, and their usefulness in Pharmacology and Bioinformatics are presented in this work. The featured applications are: drug design, virtual screening (VS), Quantitative Structure–Activity Relationship (QSAR) research, protein structure prediction and genomics (and other omics) data mining. The future need of neuromorphic hardware for DNNs is also discussed, and the two most advanced chips are reviewed: IBM TrueNorth and SpiNNaker. In addition, this review points out the importance of considering not only neurons, as DNNs and neuromorphic chips should also include glial cells, given the proven importance of astrocytes, a type of glial cell which contributes to information processing in the brain. The Deep Artificial Neuron–Astrocyte Networks (DANAN) could overcome the difficulties in architecture design, learning process and scalability of the current ML methods.

          Related collections

          Most cited references122

          • Record: found
          • Abstract: found
          • Article: found
          Is Open Access

          Deep Learning in Neural Networks: An Overview

          (2014)
          In recent years, deep artificial neural networks (including recurrent ones) have won numerous contests in pattern recognition and machine learning. This historical survey compactly summarises relevant work, much of it from the previous millennium. Shallow and deep learners are distinguished by the depth of their credit assignment paths, which are chains of possibly learnable, causal links between actions and effects. I review deep supervised learning (also recapitulating the history of backpropagation), unsupervised learning, reinforcement learning & evolutionary computation, and indirect search for short programs encoding deep and large networks.
            Bookmark
            • Record: found
            • Abstract: found
            • Article: not found

            Uniquely hominid features of adult human astrocytes.

            Defining the microanatomic differences between the human brain and that of other mammals is key to understanding its unique computational power. Although much effort has been devoted to comparative studies of neurons, astrocytes have received far less attention. We report here that protoplasmic astrocytes in human neocortex are 2.6-fold larger in diameter and extend 10-fold more GFAP (glial fibrillary acidic protein)-positive primary processes than their rodent counterparts. In cortical slices prepared from acutely resected surgical tissue, protoplasmic astrocytes propagate Ca(2+) waves with a speed of 36 microm/s, approximately fourfold faster than rodent. Human astrocytes also transiently increase cystosolic Ca(2+) in response to glutamatergic and purinergic receptor agonists. The human neocortex also harbors several anatomically defined subclasses of astrocytes not represented in rodents. These include a population of astrocytes that reside in layers 5-6 and extend long fibers characterized by regularly spaced varicosities. Another specialized type of astrocyte, the interlaminar astrocyte, abundantly populates the superficial cortical layers and extends long processes without varicosities to cortical layers 3 and 4. Human fibrous astrocytes resemble their rodent counterpart but are larger in diameter. Thus, human cortical astrocytes are both larger, and structurally both more complex and more diverse, than those of rodents. On this basis, we posit that this astrocytic complexity has permitted the increased functional competence of the adult human brain.
              Bookmark
              • Record: found
              • Abstract: found
              • Article: not found

              Tripartite synapses: glia, the unacknowledged partner.

              According to the classical view of the nervous system, the numerically superior glial cells have inferior roles in that they provide an ideal environment for neuronal-cell function. However, there is a wave of new information suggesting that glia are intimately involved in the active control of neuronal activity and synaptic neurotransmission. Recent evidence shows that glia respond to neuronal activity with an elevation of their internal Ca2+ concentration, which triggers the release of chemical transmitters from glia themselves and, in turn, causes feedback regulation of neuronal activity and synaptic strength. In view of these new insights, this article suggests that perisynaptic Schwann cells and synaptically associated astrocytes should be viewed as integral modulatory elements of tripartite synapses.
                Bookmark

                Author and article information

                Contributors
                Role: Academic Editor
                Role: Academic Editor
                Role: Academic Editor
                Role: Academic Editor
                Journal
                Int J Mol Sci
                Int J Mol Sci
                ijms
                International Journal of Molecular Sciences
                MDPI
                1422-0067
                11 August 2016
                August 2016
                : 17
                : 8
                : 1313
                Affiliations
                [1 ]Department of Information and Communications Technologies, University of A Coruña, A Coruña 15071, Spain; pastur90@ 123456gmail.com (L.A.P-R.); flanciskinho@ 123456gmail.com (F.C.); apazos@ 123456udc.es (A.P.)
                [2 ]Instituto de Investigación Biomédica de A Coruña (INIBIC), Complexo Hospitalario Universitario de A Coruña (CHUAC), A Coruña 15006, Spain
                Author notes
                [* ]Correspondence: ana.portop@ 123456udc.es ; Tel.: +34-881-011-380
                Article
                ijms-17-01313
                10.3390/ijms17081313
                5000710
                27529225
                112be88f-794c-4d1a-9410-c262e2f3c8c9
                © 2016 by the authors; licensee MDPI, Basel, Switzerland.

                This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license ( http://creativecommons.org/licenses/by/4.0/).

                History
                : 16 May 2016
                : 25 July 2016
                Categories
                Review

                Molecular biology
                artificial neural networks,artificial neuron–astrocyte networks,tripartite synapses,deep learning,neuromorphic chips,big data,drug design,quantitative structure–activity relationship,genomic medicine,protein structure prediction

                Comments

                Comment on this article