Can artificial neural replicators be useful for studying RNA replicators?
- PDF / 2,933,863 Bytes
- 17 Pages / 595.276 x 790.866 pts Page_size
- 61 Downloads / 201 Views
ORIGINAL ARTICLE
Can artificial neural replicators be useful for studying RNA replicators? Alexandr A. Ezhov1 Received: 7 May 2020 / Accepted: 16 July 2020 © Springer-Verlag GmbH Austria, part of Springer Nature 2020
Abstract Here, I discuss the usefulness of the application of special artificial neural systems – neural replicators – to study viroids – small pathogens that are short replicating RNA sequences. Using special representations of nucleotide sequences in the form of two sequences with binary components – these two sequences are incomplete representations of the same nucleotide sequence – I show that these neural systems of different sizes are replicated in a special way on them. This allows us to extract some useful information about viroids and their structure, motifs, and relationships. This study is only the first attempt to use neural replicators to analyze genetic data.
Introduction The application of artificial neural networks in genetics began in the late 1980s. The most widely used architectures include both unsupervised systems such as the Hopfield content-addressable memory model [1–4], the Boltzmann machine [3], and Kohonen self-organizing maps (SOMs) [5–7] as well as supervised systems, such as multilayer perceptrons (MLPs) [8, 9], radial basis function neural networks (RBFs) [10], support vector machines (SVMs) [11, 12], and the most powerful and successful architecture such as deep neural networks (for a review of the application of these systems in genetics, see, e.g., reference [13]). Among the applications in genetics are protein structure prediction [14–16], promoter recognition [2, 17], DNA sequence analysis [8, 18] including prediction of the functional activity of DNA sequences [19], analysis of the effects of noncoding DNA [20], search for hidden periodicities [21], analysis of mutation effects [22], and prediction of individual phenotypes [23]. Other important areas include RNA sequences analysis [24–28], detection of tumor genes and cancer analysis [12, 29, 30], genome-wide prediction [31] and population genetics [32]. Here, I present a completely different neural model, neural replicators, which arose in ensembles of the Handling Editor: Marc H. V. Van Regenmortel. * Alexandr A. Ezhov [email protected] 1
interacting Hopfield networks with a mechanism for synchronous threshold change [33] and discuss the possibility of their use for analysis of viroid RNA sequences. A model of neural replicators (self-reproducible neural networks) was proposed in the early 1990s [33, 34]. Two decades later, the idea of the existence of replicators in the brain was widely discussed (for a review, see reference [35]), from the point of view of the selectionist approach to brain functioning. There are many arguments for the existence of replicators in the brain. One of them is as follows: since the brain is a complex system, its creative activity indicates that it acts at the edge of chaos. This view has been confirmed by experimental studies [36]. However, it is in this zone of functioning of
Data Loading...
