Numerical Mapping Techniques Research Articles

The Effect of Numerical Mapping Techniques on Performance in Genomic Research

In genomic signal processing applications, digitization of these signals is needed to process and analyze DNA signals. In the digitization process, the mapping technique to be chosen greatly affects the performance of the system for the genomic domain to be studied. The purpose of this review is to analyze how numerical mapping techniques used in digitizing DNA sequences affect performance in genomic studies. For this purpose, all digital coding techniques presented in the literature in the studies conducted in the last 10 years have been examined, and the numerical representations of these techniques are given in a sample DNA sequence. In addition, the frequency of use of these coding techniques in four popular genomic areas such as exon region identification, exon-intron classification, phylogenetic analysis, gene detection, and the min-max range of the performances obtained by using these techniques in that area are also given. This study is thought to be a guide for researchers who want to work in the field of bioinformatics.

Deep transfer learning for automated liver cancer gene recognition using spectrogram images of digitized DNA sequences

DNA biomarkers are considered to be an important diagnostic factor for early detection of cancer. With the development of technology, many machine learning methods have been introduced to detect cancer biomarkers from DNA sequences in the literature. In this study, a new approach was proposed for the detection of cancer genes, which is an important step for the prediction of cancer. In the proposed approach, the gene sequences were digitized by three different numerical mapping techniques. Following digitization, these DNA sequences were initially examined with two different ways as one-dimensional signal and two-dimensional spectrogram images. Firstly, the digitized sequences were examined with the designed CNN model as a one-dimensional signal. Secondly, DNA signals were converted to 2D spectrogram images and examined with two different 2D CNN models. In the first model, feature vectors were obtained by VGG16 and classified by SVM. In the second model, new layers were added to the final output layers of VGG16, and fine-tuning was applied. An accuracy of 80.36% was obtained in one-dimensional CNN model, an accuracy of 98.86% was achieved in the model where features were extracted with VGG16 and classified with SVM, and an accuracy of 100% was observed in the model where fine tuning was applied to the layers of VGG16. The proposed method indicated that effective features were extracted with CNN models to distinguish liver cancer and normal liver gene sequences. The application results showed that the system is ready to be tested with a larger dataset and different cancer types.

The Issue of Groundwater Salinization in Coastal Areas of the Mediterranean Region: A Review

The Mediterranean area is undergoing intensive demographic, social, cultural, economic, and environmental changes. This generates multiple environmental pressures such as increased demand for water resources, generation of pollution related to wastewater discharge, and land consumption. In the Mediterranean area, recent climate change studies forecast large impacts on the hydrologic cycle. Thus, in the next years, surface and ground-water resources will be gradually more stressed, especially in coastal areas. In this review paper, the historical and geographical distribution of peer-review studies and the main mechanisms that promote aquifer salinization in the Mediterranean area are critically discussed, providing the state of the art on topics such as actual saltwater wedge characterization, paleo-salinities in coastal areas, water-rock interactions, geophysical techniques aimed at delineating the areal and vertical extent of saltwater intrusion, management of groundwater overexploitation using numerical models and GIS mapping techniques for aquifer vulnerability to salinization. Each of the above-mentioned approaches has potential advantages and drawbacks; thus, the best tactic to tackle coastal aquifer management is to employ a combination of approaches. Finally, the number of studies focusing on predictions of climate change effects on coastal aquifers are growing but are still very limited and surely need further research.

A novel genome analysis method with the entropy-based numerical technique using pretrained convolutional neural networks

The identification of DNA sequences as exon and intron is a common problem in genome analysis. The methods used for feature extraction and mapping techniques for the digitization of sequences affect directly the solution of this problem. The existing mapping techniques are not enough to detect coding and noncoding regions in some genomes because the digital representation of each base in a DNA sequence with an integer does not fully reflect the structure of an original DNA sequence. In the entropy-based mapping technique, we could overcome this problem because the technique deepens distinction rates of exon regions, and better reflects the complexity of DNA sequences. Moreover, in the literature, features are extracted by using various statistical techniques. The statistical features to be extracted are chosen by a system designer's experience. The other proposed approach in this study is to carry out the feature extraction using the transfer learning method. Transfer learning and feature extraction are performed automatically by convolutional neural network models as independent of the data set. In this study, we propose a new method to classify DNA sequences as exon and intron using two approaches. In the first approach, the entropy-based numerical technique was used for the numerical representation of DNA sequences. In the second approach, transfer learning was used to extract features. Then, the obtained features were classified by support vector machine and k -nearest neighbors algorithm. As a result of the classification, accurate performance with 97.8% was achieved. The performance of the current method was compared with the other numerical mapping techniques and feature extraction methods. The results showed that the developed method was much more successful than other methods.

A novel numerical mapping method based on entropy for digitizing DNA sequences

Recently, digital signal processing has been widely applied in the study of genomics. One of the genomic studies is identification of protein-coding regions. Where is a protein coded? How much is encoded? Where are growth and development regulated? The answer to these questions is possible by DNA sequences that can be classified as the exon and intron. In signal processing application, numerical signals are used due to symbolic signal nature of DNA sequence; yet, it must be converted from symbolic sequence to numeric sequence prior the analysis in data preprocessing. The bases in a DNA sequence are represented with four letters A, G, C and T. Each letter corresponds to a numeric value. In the literature, several numerical mapping techniques exist. In this paper, a novel numerical mapping approach has been proposed for converting string to numerical values. Each codon is mapped by improved fractional derivative of Shannon equation in this approach. For exon regions prediction, three methods have been used. These methods are singular value decomposition (SVD), discrete Fourier transform (DFT) and short-time Fourier transform (STFT). The performance of the proposed mapping technique has been evaluated based on the above-mentioned three classification methods. The proposed novel technique has showed more success in the identification of protein-coding regions as compared to the predominant existing mapping techniques SVD, DFT and STFT methods.

Integrated Model of DNA Sequence Numerical Representation and Artificial Neural Network for Human Donor and Acceptor Sites Prediction

Human Genome Project has led to a huge inflow of genomic data. After the completion of human genome sequencing, more and more effort is being put into identification of splicing sites of exons and introns (donor and acceptor sites). These invite bioinformatics to analysis the genome sequences and identify the location of exon and intron boundaries or in other words prediction of splicing sites. Prediction of splice sites in genic regions of DNA sequence is one of the most challenging aspects of gene structure recognition. Over the last two decades, artificial neural networks gradually became one of the essential tools in bioinformatics. In this paper artificial neural networks with different numerical mapping techniques have been employed for building integrated model for splice site prediction in genes. An artificial neural network is trained and then used to find splice sites in human genes. A comparison between different mapping methods using trained neural network in terms of their precision in prediction of donor and acceptor sites will be presented in this paper. Training and measuring performance of neural network are carried out using sequences of the human genome (GRch37/hg19- chr21). Simulation results indicate that using Electron-Ion Interaction Potential numerical mapping method with neural network yields to the best performance in prediction.

Genomic Analysis and Classification of Exon and Intron Sequences Using DNA Numerical Mapping Techniques

Using digital signal processing in genomic field is a key of solving most problems in this area such as prediction of gene locations in a genomic sequence and identifying the defect regions in DNA sequence. It is found that, using DSP is possible only if the symbol sequences are mapped into numbers. In literature many techniques have been developed for numerical representation of DNA sequences. They can be classified into two types, Fixed Mapping (FM) and Physico Chemical Property Based Mapping (PCPBM ( . The open question is that, which one of these numerical representation techniques is to be used? The answer to this question needs understanding these numerical representations considering the fact that each mapping depends on a particular application. This paper explains this answer and introduces comparison between these techniques in terms of their precision in exon and intron classification. Simulations are carried out using short sequences of the human genome (GRch37/hg19). The final results indicate that the classification performance is a function of the numerical representation method.

Identification of protein coding regions using antinotch filters

A major area of research in genomic sequence analysis is the identification of protein coding regions using the period-3 property. Previously antinotch filter has been used for this purpose. In this paper, three antinotch filters, namely conjugate suppression antinotch filter, antinotch filter followed by moving average filter and harmonic suppression antinotch filter are proposed to improve the identification accuracy. Conjugate suppression antinotch filter suppresses the conjugate frequency component, antinotch filter followed by moving average filter reduces the background noise and harmonic suppression antinotch filter suppresses the harmonic frequency component. Several existing DNA to numerical mapping techniques are compared for GENSCAN test set and based on the result one mapping technique is recommended so that detailed analysis can be performed using various datasets. The computational complexity of the antinotch filters is evaluated in comparison with the ST-DFT method and it is found that the computational load is reduced to a greater extent in antinotch filter. The identification accuracy of the proposed antinotch filter methods is compared with the existing antinotch filter method at the nucleotide level for benchmark datasets. The results show that proposed methods outperform the existing method, giving improved identification of the protein coding regions.

Numerical Construction of Potential Flows in Multiply Connected Channel Domains

We consider inviscid and incompressible flows in two-dimensional multiply connected domains between two infinitely long parallel straight lines, which are simple models describing channel flows such as rivers, streams and canals. In the present paper, we propose two computational methods to construct the complex potentials for a uniform flow, a point vortex, a source and a sink in the multiply connected channel domains. First, we give analytic formulae of the potential flows in a special channel domain with circular holes, which are described in terms of the Schottky-Klein prime function as in Crowdy and Marshall [5]. Second, we make use of the numerical conformal mapping techniques by Amano [1] and Ogata [11] to approximate the complex potentials in the channel domains that contain boundaries of arbitrary shapes. We also confirm that the numerical techniques can be applied to construct the potential flows in a real river. The two methods provide us with analytic representations of the complex potentials, which enable us to compute the velocity field in the channel domains and the force on the boundaries quickly and accurately.

Numerical solution to the solidification of aluminium in the presence of various fibres

In an attempt to understand the experimentally observed solidification microstructures in metal matrix composites, the influence of SiC, graphite and alumina fibres on the solidification of aluminium has been studied numerically. Irregular geometries of the composite material were mapped into simple rectangles through numerical conformal mapping techniques to analyse the influence of a single fibre or a row of fibres on a unidirectionally advancing planar solid-liquid interface. The fibres were assumed to be circular in cross-section and the direction of the interface movement was perpendicular to the length of the fibres. The study showed that for fibres with lower thermal conductivity than aluminium, the interface first goes through acceleration as it approaches and ascends the fibre and then deceleration as it descends the fibre. The acceleration and deceleration phenomena of the interface increases as the thermal conductivity ratio of fibre to liquid aluminium decreases. With low thermal conductivity ratios (K f/K L≪1), the interface is orthogonal to the fibre surface. When the conductivity of the fibre is lower than that of the melt, the interface becomes convex facing the fibre; this mode would lead to pushing of the fibre ahead if it was free to move, as has been experimentally observed in cast microstructures of metal matrix composites. The temperature versus solidification time plots of two points, one in the fibre and the other in aluminium, show that the fibre with a conductivity lower than the matrix is at a temperature higher than the melt; the temperature difference between the two points increases with increasing solidification rate for all the positions of the interface before it touches the fibre. The three-fibre study shows that as the number of fibres increases, the curvature of the interface increases upon approaching the subsequent fibres. The relationship between these numerical computations and experimental observations has been discussed.

A numerical study of thermal interaction of solidification fronts with spherical particles during solidification of metal-matrix composite materials

The potential influence of the presence of copper, iron, mica, TiB 2, graphite and alumina particles dispersed in melts of several matrix metals including aluminum, copper and lead was studied. Specifically the influence of the particles on the solidification fronts during freezing of various matrix melts was numerically studied in order to understand the experimentally observed solidification microsctructures in metal-matrix composites. Irregular geometries of the composite material were mapped into simple rectangles through numerical conformal mapping techniques to analyse the influence of a single particle or a row of particles on a unidirectionally advancing planar solidification front. The particles were assumed to be spherical and located in the center of the cylindrical mold. The study showed that for particles with lower thermal conductivity than the melt, the interface first goes through acceleration as it approaches and encapsulates the front half of the particle, and then it decelerates as it encapsulates the remaining half of the particle. The acceleration and deceleration phenomenon of the interface increases as the thermal conductivity ratio of particle to melt decreases. With low thermal conductivity ratios ( K p K l ⪡ 1 ), the interface is orthogonal to the particle. When the conductivity of the particle is lower than that of the melt, the interface becomes convex facing the particle; this mode would lead to pushing of the particle ahead if it was free to move as has been experimentally observed during solidification of several metal-matrix composites when K p K l ⪢ 1 , the phenomenon is the opposite to that mentioned above and the interface would become concave facing the particle tending to entrap the particle. The temperature vs solidification time plots of two points, one in the particle and the other in melt, show that the particle with a conductivity lower than the matrix is at a temperature higher than the melt; the temperature difference between the two points decreases with increasing solidification rate for all the positions of the interface before it touches the particle. The influences of these temperature differences between the melt and the particles on solidification of composites are discussed.

Chaotic motion in a primordial comet disk beyond Neptune and comet influx to the Solar System

CHAOTIC dynamics, characterized by extremely sensitive dependence on initial conditions with exponential divergence of neighbouring trajectories and subsequent loss of predictability1, have recently been demonstrated to be quite common in the Solar System. The long-standing problems of the delivery of meteorites to Earth-crossing orbits2 and the existence of the Kirkwood gaps3 have essentially been solved by the discovery of the chaotic zones with divergence timescales of a few million years in the asteroid zone near orbital resonances of Jupiter3–7. It has also been demonstrated8 that the orbit of Pluto is chaotic with divergence timescales of ∼20 Myr. Numerical mapping techniques, which underestimate chaotic behaviour, have determined that orbits lying between the outer planets can exhibit chaotic motions9, and recent numerical studies10 have indicated that even the inner planets exhibit chaotic motion with divergence timescales of ∼5 Myr. Previous calculations11 using a heliocentric frame have shown that a scattered disk of cometesimals beyond Neptune develops chaotic motions with divergence timescales less than 1 Myr. Here we describe improved numerical simulations using a barycentric system and thus including the additional frequencies associated with motion of the Sun. Because the onset of chaotic motion is characterized by the appearance of broad-band noise, the inclusion of these additional perturbations should increase the likelihood of stochastic behaviour and thus the chaotic zone should be larger. Our results indeed indicate that comets in a low-eccentricity primordial disk beyond Neptune can exhibit chaotic motions which lead to interesting consequences for the dynamics of comets.

The treatment of corner and pole-type singularities in numerical conformal mapping techniques

This paper is a report of recent developments concerning the nature and the treatment of singularities that affect certain numerical conformal mapping techniques. The paper also includes some new results on the nature of singularities that the mapping function may have in the complement of the closure of the domain under consideration.

Development of a minicomputer atmospheric noise model

A practical but simplified atmospheric noise model in universal time for 1‐MHz Fam usable in estimating system signal‐to‐noise ratios in microminicomputer‐based HF propagation prediction systems is presented. Numerical mapping techniques were used to represent worldwide atmospheric noise in 24 numerical maps: one for each of six 4‐hour UT time blocks and each of four 3‐month periods. This representation results in a minimum reduction by one fifth in the number of coefficients necessary in computer memory over that of an earlier model in which the map represented an entire 24‐hour day. The simpler expression for the median noise levels (Fam) results in significant savings in computer code and time. Several versions were produced, each with a different number of coefficients (number of harmonics) in the Fourier representation for each time block. The number of coefficients ranged from 192 coefficients per time block for the most accurate versions to 35 coefficients for the least accurate. The accuracies were determined by calculating the rms residual between values of Fam used to develop the model and the corresponding values computed from each model. The most accurate model had an average rms error for all times and seasons of 1.3 dB, whereas the least accurate had a rms error of 4.5 dB. During summer, the worst season, the most accurate model had a rms error of 1.4 dB, whereas the least accurate had a rms error of 5.6 dB. The International Radio Consultative Committee noise data used to generate the models had an average standard deviation of the error at 1 MHz of 5.4 dB and an error of 6.7 dB during summer; the numerical map in use in current HF prediction programs has an average rms error of 1.7 dB and an error of 2.0 dB during summer. The locations of the maximum absolute deviations of the models were confined to a small region in central Africa.