Frequency Chaos Research Articles

LHPre: Phage Host Prediction with VAE-based Class Imbalance Correction and Lyase Sequence Embedding.

The escalation of antibiotic resistance underscores the need for innovative approaches to combat bacterial infections. Phage therapy has emerged as a promising solution, wherein host determination plays an important role. Phage lysins, characterized by their specificity in targeting and cleaving corresponding host bacteria, serve as key players in this paradigm. In this study, we present a novel approach by leveraging genes of phage-encoded lytic enzymes for host prediction, culminating in the development of LHPre. Initially, gene fragments of phage-encoded lytic enzymes and their respective hosts were collected from the database. Secondly, DNA sequences were encoded using the Frequency Chaos Game Representation (FCGR) method, and pseudo samples were generated employing the Variational Autoencoder (VAE) model to address class imbalance. Finally, a prediction model was constructed using the Vision Transformer(Vit) model. Five-fold cross-validation results demonstrated that LHPre surpassed other state-of-the-art phage host prediction methods, achieving accuracies of 85.04%, 90.01%, and 93.39% at the species, genus, and family levels, respectively.

Secure WDM communication based on second order differential feedback electro-optical chaotic mutual injection synchronization

An optical phase chaos and mutual injection chaos synchronization scheme based on second order differential feedback electro-optical loop is proposed for chaotic communication. The filtered chaotic waves for mutual injection synchronization and multi-channel signals are transmitted through wavelength division multiplexing (WDM) to realize large-capacity confidential communication. The results show that the bandwidth of the optical chaos is 41.38 GHz, the normalized permutation entropy is 0.999, and the chaos has the performance of time delay suppressing (TDS). This scheme achieves up to 7-channel 25 GBaud 16QAM with the multi-modulus algorithm (MMA) algorithm at OSNR of 25 dB after 100 km transmission. The bit-rate distance of 280,000 Gb/s km is successfully realized by 500 km transmission of 4-channel 35 GBaud 16QAM signals at 25 dB OSNR. The analysis proves that the performance of the second order digital differential electro-optical phase feedback (2nd order DEOP) chaos is better than that of the 1st order DEOP chaos and 3rd order DEOP chaos. The 2nd order DEOP chaos can be generated at a smaller feedback coefficient, the bandwidth of radio frequency chaos is wider, and the TDS suppression is better. This scheme realizes high-speed and large-capacity secure optical chaos communication, which can be widely used in WDM networks and has practical feasibility and application value.

Zero-shot-capable identification of phage-host relationships with whole-genome sequence representation by contrastive learning.

Accurately identifying phage-host relationships from their genome sequences is still challenging, especially for those phages and hosts with less homologous sequences. In this work, focusing on identifying the phage-host relationships at the species and genus level, we propose a contrastive learning based approach to learn whole-genome sequence embeddings that can take account of phage-host interactions (PHIs). Contrastive learning is used to make phages infecting the same hosts close to each other in the new representation space. Specifically, we rephrase whole-genome sequences with frequency chaos game representation (FCGR) and learn latent embeddings that 'encapsulate' phages and host relationships through contrastive learning. The contrastive learning method works well on the imbalanced dataset. Based on the learned embeddings, a proposed pipeline named CL4PHI can predict known hosts and unseen hosts in training. We compare our method with two recently proposed state-of-the-art learning-based methods on their benchmark datasets. The experiment results demonstrate that the proposed method using contrastive learning improves the prediction accuracy on known hosts and demonstrates a zero-shot prediction capability on unseen hosts. In terms of potential applications, the rapid pace of genome sequencing across different species has resulted in a vast amount of whole-genome sequencing data that require efficient computational methods for identifying phage-host interactions. The proposed approach is expected to address this need by efficiently processing whole-genome sequences of phages and prokaryotic hosts and capturing features related to phage-host relationships for genome sequence representation. This approach can be used to accelerate the discovery of phage-host interactions and aid in the development of phage-based therapies for infectious diseases.

Cancer disease multinomial classification using transfer learning and SVM on the genes’ sequences

INTRODUCTION: Early disease detection plays an important role in medical field especially for cancer disease, which helps doctors in diagnosing and identifying the therapeutic process. Aiming to provide assistance, many biological techniques other than machine and deep learning models were proposed. They were applied on a different type of data such as medical images and clinical data. Despite the efficiency of those techniques, they remain costly and need a lot of execution and preparation time, and resources.OBJECTIVES: In this paper, we present a novel method of disease detection analyzing the genes sequences composition.METHODS: We start by extracting k-mer nucleotides as features from gene sequences with the Frequency Chaos Game Representation (FCGR) technique. Since extracted data are huge, we use a DeepInsight model to extract the most representative k-mers.A combination of a transfer learning model, which is Residual neural Network (ResNet), and a support vector machine (SVM) algorithm is then used then to classify samples into 18 cancer disease types.RESULTS: We achieved an accuracy of 0.98 while choosing FCGR6 in feature extraction, and a combination of ResNet50 and SVM in the multinomial classification step, against an accuracy of 0.97 while using ResNet50 with a fully connected layer and FCGR5.CONCLUSION: Defining the gene sequence alterations helps in the disease detection at early stage. Here, we adopt the FCGR method (that gives the frequency of each k-mer) in defining features of the gene sequences. Then, we use deep learning models to deal with the big number of characteristics and predicting different cancer diseases.

A hybrid deep learning approach for COVID-19 detection based on genomic image processing techniques

The coronavirus disease 2019 (COVID-19) pandemic has been spreading quickly, threatening the public health system. Consequently, positive COVID-19 cases must be rapidly detected and treated. Automatic detection systems are essential for controlling the COVID-19 pandemic. Molecular techniques and medical imaging scans are among the most effective approaches for detecting COVID-19. Although these approaches are crucial for controlling the COVID-19 pandemic, they have certain limitations. This study proposes an effective hybrid approach based on genomic image processing (GIP) techniques to rapidly detect COVID-19 while avoiding the limitations of traditional detection techniques, using whole and partial genome sequences of human coronavirus (HCoV) diseases. In this work, the GIP techniques convert the genome sequences of HCoVs into genomic grayscale images using a genomic image mapping technique known as the frequency chaos game representation. Then, the pre-trained convolution neural network, AlexNet, is used to extract deep features from these images using the last convolution (conv5) and second fully-connected (fc7) layers. The most significant features were obtained by removing the redundant ones using the ReliefF and least absolute shrinkage and selection operator (LASSO) algorithms. These features are then passed to two classifiers: decision trees and k-nearest neighbors (KNN). Results showed that extracting deep features from the fc7 layer, selecting the most significant features using the LASSO algorithm, and executing the classification process using the KNN classifier is the best hybrid approach. The proposed hybrid deep learning approach detected COVID-19, among other HCoV diseases, with 99.71% accuracy, 99.78% specificity, and 99.62% sensitivity.

Accurate and fast clade assignment via deep learning and frequency chaos game representation.

Since the beginning of the coronavirus disease 2019 pandemic, there has been an explosion of sequencing of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus, making it the most widely sequenced virus in the history. Several databases and tools have been created to keep track of genome sequences and variants of the virus; most notably, the GISAID platform hosts millions of complete genome sequences, and it is continuously expanding every day. A challenging task is the development of fast and accurate tools that are able to distinguish between the different SARS-CoV-2 variants and assign them to a clade. In this article, we leverage the frequency chaos game representation (FCGR) and convolutional neural networks (CNNs) to develop an original method that learns how to classify genome sequences that we implement into CouGaR-g, a tool for the clade assignment problem on SARS-CoV-2 sequences. On a testing subset of the GISAID, CouGaR-g achieved an $96.29\%$ overall accuracy, while a similar tool, Covidex, obtained a $77,12\%$ overall accuracy. As far as we know, our method is the first using deep learning and FCGR for intraspecies classification. Furthermore, by using some feature importance methods, CouGaR-g allows to identify k-mers that match SARS-CoV-2 marker variants. By combining FCGR and CNNs, we develop a method that achieves a better accuracy than Covidex (which is based on random forest) for clade assignment of SARS-CoV-2 genome sequences, also thanks to our training on a much larger dataset, with comparable running times. Our method implemented in CouGaR-g is able to detect k-mers that capture relevant biological information that distinguishes the clades, known as marker variants. The trained models can be tested online providing a FASTA file (with 1 or multiple sequences) at https://huggingface.co/spaces/BIASLab/sars-cov-2-classification-fcgr. CouGaR-g is also available at https://github.com/AlgoLab/CouGaR-g under the GPL.

In silico identification of multiple conserved motifs within the control region of Culicidae mitogenomes

Mosquitoes are important vectors for human and animal diseases. Genetic markers, like the mitochondrial COI gene, can facilitate the taxonomic classification of disease vectors, vector-borne disease surveillance, and prevention. Within the control region (CR) of the mitochondrial genome, there exists a highly variable and poorly studied non-coding AT-rich area that contains the origin of replication. Although the CR hypervariable region has been used for species differentiation of some animals, few studies have investigated the mosquito CR. In this study, we analyze the mosquito mitogenome CR sequences from 125 species and 17 genera. We discovered four conserved motifs located 80 to 230 bp upstream of the 12S rRNA gene. Two of these motifs were found within all 392 Anopheles (An.) CR sequences while the other two motifs were identified in all 37 Culex (Cx.) CR sequences. However, only 3 of the 304 non-Culicidae Dipteran mitogenome CR sequences contained these motifs. Interestingly, the short motif found in all 37 Culex sequences had poly-A and poly-T stretch of similar length that is predicted to form a stable hairpin. We show that supervised learning using the frequency chaos game representation of the CR can be used to differentiate mosquito genera from their dipteran relatives.

Genomic image representation of human coronavirus sequences for COVID-19 detection

Coronavirus (CoV) disease 2019 (COVID-19) is a severe pandemic affecting millions worldwide. Due to its rapid evolution, researchers have been working on developing diagnostic approaches to suppress its spread. This study presents an effective automated approach based on genomic image processing (GIP) techniques to rapidly detect COVID-19, among other human CoV diseases, with high acceptable accuracy. The GIP technique was applied as follows: first, genomic graphical mapping techniques were used to convert the genome sequences into genomic grayscale images. The frequency chaos game representation (FCGR) and single gray-level representation (SGLR) techniques were used in this investigation. Then, several statistical features were obtained from the images to train and test many classifiers, including the k-nearest neighbors (KNN). This study aimed to determine the efficacy of the FCGR (with different orders) and SGLR images for accurately detecting COVID-19,using a dataset containing both partial and complete genome sequences. The results recommended the fourth-order FCGR image as a proper genomic image for extracting statistical features and achieving accurate classification. Furthermore, the results showed that KNN achieved an overall accuracy of 99.39% in detecting COVID-19, among other human CoV diseases, with 99.48% precision, 99.31% sensitivity, 99.47% specificity, 0.99 F1-score, and 0.99 Matthew's correlation coefficient.

Fractal Analysis of DNA Sequences Using Frequency Chaos Game Representation and Small-Angle Scattering.

The fractal characteristics of DNA sequences are studied using the frequency chaos game representation (FCGR) and small-angle scattering (SAS) technique. The FCGR allows representation of the frequencies of occurrence of k-mers (oligonucleotides of length k) in the form of images. The numerically encoded data are then used in a SAS analysis to enhance hidden features in DNA sequences. It is shown that the simulated SAS intensity allows us to obtain the fractal dimensions and scaling factors at various scales. These structural parameters can be used to distinguish unambiguously between the scaling properties of complex hierarchical DNA sequences. The validity of this approach is illustrated on several sequences from: Escherichia coli, Mouse mitochondrion, Homo sapiens mitochondrion and Human cosmid.

Dissipation Assisted Frequency Comb and Chaos Generation in the Optomechanics

Dissipation Assisted Frequency Comb and Chaos Generation in the Optomechanics

DeLUCS: Deep learning for unsupervised clustering of DNA sequences.

We present a novel Deep Learning method for the Unsupervised Clustering of DNA Sequences (DeLUCS) that does not require sequence alignment, sequence homology, or (taxonomic) identifiers. DeLUCS uses Frequency Chaos Game Representations (FCGR) of primary DNA sequences, and generates “mimic” sequence FCGRs to self-learn data patterns (genomic signatures) through the optimization of multiple neural networks. A majority voting scheme is then used to determine the final cluster assignment for each sequence. The clusters learned by DeLUCS match true taxonomic groups for large and diverse datasets, with accuracies ranging from 77% to 100%: 2,500 complete vertebrate mitochondrial genomes, at taxonomic levels from sub-phylum to genera; 3,200 randomly selected 400 kbp-long bacterial genome segments, into clusters corresponding to bacterial families; three viral genome and gene datasets, averaging 1,300 sequences each, into clusters corresponding to virus subtypes. DeLUCS significantly outperforms two classic clustering methods (K-means++ and Gaussian Mixture Models) for unlabelled data, by as much as 47%. DeLUCS is highly effective, it is able to cluster datasets of unlabelled primary DNA sequences totalling over 1 billion bp of data, and it bypasses common limitations to classification resulting from the lack of sequence homology, variation in sequence length, and the absence or instability of sequence annotations and taxonomic identifiers. Thus, DeLUCS offers fast and accurate DNA sequence clustering for previously intractable datasets.

Bidirectional Long Short-Term Memory Network for Taxonomic Classification

Identifying and classifying Deoxyribonucleic Acid (DNA) sequences and their functions have been considered as the main challenges in bioinformatics. Advances in machine learning and Deep Learning (DL) techniques are expected to improve DNA sequence classification. Since the DNA sequence classification depends on analyzing textual data, Bidirectional Long Short-Term Memory (BLSTM) algorithms are suitable for tackling this task. Generally, classifiers depend on the patterns to be processed and the pre-processing method. This paper is concerned with a new proposed classification framework based on Frequency Chaos Game Representation (FCGR) followed by Discrete Wavelet Transform (DWT) and BLSTM. Firstly, DNA strings are transformed into numerical matrices by FCGR. Then, the DWT is used instead of the pooling layer as a tool of data compression. The benefit of using the DWT is two-fold. It preserves the useful information only that enables the following BLSTM training, effectively. Besides, DWT adds more important details to the encoded sequences due to finding effective features in the DNA fragments. Finally, the BLSTM model is trained to classify the DNA sequences. Evaluation metrics such as F1 score and accuracy show that the proposed framework outperforms the state-of-the-art algorithms. Hence, it can be used in DNA classification applications.

Wavelet-based multifractal analysis of C.elegans sequences based on FCGS signal

The frequency chaos game signal (FCGS) is a new mapping technique of DNA sequences inspired by the Chaos Game theory. It has the particularity of exploiting the statistical properties of the genomic sequences composition which may serve in detecting interesting structures within the DNA sequences. Unlike the classical DNA sequences coding techniques, where the nucleotides are assigned numerical values depending on their chemical and structural characteristics, the advantage of the FCGS coding method is its invariance concerning the assignment of nucleotides to their numerical values. Mapping nucleotide sequences by the FCGS technique produces a multifractal landscape that can be studied quantitatively by applying the so-called wavelet transform modulus maxima method (WTMM). This method provides a natural generalization of the classical box-counting techniques for the multifractal signal analysis. In fact, the wavelets are playing the role of generalized oscillating boxes. In this paper, we use the WTMM method to perform a multifractal analysis of the C.elegans genome using FCGS signal with order two. First, we generate the FCGS signal of particulars C.elegans genome regions like exon, intron, Helitron, CERP3, and CEREP55. Next, we apply the WTMM to calculate the singularity spectrum. Finally, we prove, with the obtained results, the multifractal nature of this genome and the variability of this multifractal characteristic according to the region studied. We also discover that this variability was mainly dependent on differences in the contents of repetitive DNA in each DNA sequence. This approach will be used to characterize sequences to allow their automatic classification. The technique aims to characterize structural and functional regions of chromosomes in genomes and will help later to explore automatically and study unidentified sequences.

Comparative analysis and prediction of nucleosome positioning using integrative feature representation and machine learning algorithms

BackgroundNucleosome plays an important role in the process of genome expression, DNA replication, DNA repair and transcription. Therefore, the research of nucleosome positioning has invariably received extensive attention. Considering the diversity of DNA sequence representation methods, we tried to integrate multiple features to analyze its effect in the process of nucleosome positioning analysis. This process can also deepen our understanding of the theoretical analysis of nucleosome positioning.ResultsHere, we not only used frequency chaos game representation (FCGR) to construct DNA sequence features, but also integrated it with other features and adopted the principal component analysis (PCA) algorithm. Simultaneously, support vector machine (SVM), extreme learning machine (ELM), extreme gradient boosting (XGBoost), multilayer perceptron (MLP) and convolutional neural networks (CNN) are used as predictors for nucleosome positioning prediction analysis, respectively. The integrated feature vector prediction quality is significantly superior to a single feature. After using principal component analysis (PCA) to reduce the feature dimension, the prediction quality of H. sapiens dataset has been significantly improved.ConclusionsComparative analysis and prediction on H. sapiens, C. elegans, D. melanogaster and S. cerevisiae datasets, demonstrate that the application of FCGR to nucleosome positioning is feasible, and we also found that integrative feature representation would be better.

Applying frequency chaos game representation with perceptual image hashing to gene sequence phylogenetic analyses

As a very important research direction in the field of bioinformatics, sequence alignment plays a vital role in the research and development of biology. Converting genome sequence to graph by using frequency chaos game representation (FCGR) is an excellent gene sequence mapping technology, which can store rich genetic information into FCGR graphics. To each FCGR image, we construct its perceptual image hashing (PIH) matrix using the bicubic interpolation zooming. The difference of the perceptual hash matrix of each two images is calculated, and the clustering distance of the corresponding two gene sequences is represented by the differentials of the perceptual hash matrix. In this paper, we aligned and analyzed several typical genome sequence datasets including mammalian mitochondrial genes, human immunodeficiency virus 1 (HIV-1) and hepatitis E virus (HEV) to build their evolutionary trees. Experimental results showed that our PIH combining FCGR method (FCGR-PIH) has similar classification accuracy to the classical Clustal W sequence alignment method. Furthermore, 25 complete mitochondrial DNA sequences of cichlid fishes and 27 Escherichia coli/Shigella full genome sequences were selected from the AFproject test platform for tests. The performance benchmark rankings demonstrate the effectiveness of the FCGR-PIH algorithm and its potential for large-scale genome sequence analysis.

Involving FCGR images in studying fractality and multifractality in human chromosome 22 and bacteria complete genome

The frequency chaos game representation (FCGR) is a simple yet powerful visualization method of DNA sequences. It provides the possibility of representing genomes by images, revealing in such a way different fractal structures. In this paper, we perform a fractal and multifractal analysis of Human chromosome 22 and some complete genomes based on the FGCR image. We used the fractal dimension (FD) and the multifractality degree (ΔDq) to characterize and distinguish genomes. First, we construct the FCGR image with different orders of human chromosome 22. Next, we calculate the fractal dimension, the general dimension spectrum and the multifractal spectrum of each FCGR image using the box-counting method. Then, we examine the FCGR image fractal and multifractal characteristics impact on highlighting the existence of repetitive DNA sequences in human chromosome 22. We also observe the relationship between fractality and multifractality. After that, we apply this study to bacteria completes genomes and C.elegans chromosome I. The obtained results show that the multifractal spectra of all organisms studied are multifractal-like and chromosome 22 strong multifractality proves its richness of repetitive sequences. Also, we observed that with the increasing the FCGR order value, the multifractality grows and the fractal dimension lessens. Finally, by assigning to each sequence a point in two-dimensional space (FD, ΔDq), we obtained three classes of genomes. We can easily distinguish the human chromosome 22 from other genomes and Bacteria are almost close in the spaces (FD, ΔDq).

An analysis of k-mer frequency features with SVM and CNN for viral subtyping classification

Viral subtyping classification is very relevant for the appropriate diagnosis and treatment of illnesses. The most used tools are based on alignment-based methods, nevertheless, they are becoming too slow with the increase of genomic data. For that reason, alignment-free methods have emerged as an alternative. In this work, we analyzed four alignment-free algorithms: two methods use k-mer frequencies (Kameris and Castor-KRFE); the third method used a frequency chaos game representation of a DNA with CNNs; finally the last one, process DNA sequences as a digital signal (ML-DSP). From the comparison, Kameris and Castor-KRFE outperformed the rest, followed by the method based on CNNs.

A deep learning model for plant lncRNA-protein interaction prediction with graph attention.

Long non-coding RNAs (lncRNAs) play a broad spectrum of distinctive regulatory roles through interactions with proteins. However, only a few plant lncRNAs have been experimentally characterized. We propose GPLPI, a graph representation learning method, to predict plant lncRNA-protein interaction (LPI) from sequence and structural information. GPLPI employs a generative model using long short-term memory (LSTM) with graph attention. Evolutionary features are extracted using frequency chaos game representation (FCGR). Manifold regularization and l2-norm are adopted to obtain discriminant feature representations and mitigate overfitting. The model captures locality preserving and reconstruction constraints that lead to better generalization ability. Finally, potential interactions between lncRNAs and proteins are predicted by integrating catboost and regularized Logistic regression based on L-BFGS optimization algorithm. The method is trained and tested on Arabidopsis thaliana and Zea mays datasets. GPLPI achieves accuracies of 85.76% and 91.97% respectively. The results show that our method consistently outperforms other state-of-the-art methods.

New Intraclass Helitrons Classification Using DNA-Image Sequences and Machine Learning Approaches

Helitrons, eukaryotic transposable elements (TEs) transposed by rolling-circle mechanism, have been found in various species with highly variable copy numbers and sometimes with a large portion of their genomes. The impact of helitrons sequences in the genome is to frequently capture host genes during their transposition. Since their discovery, 18 years ago, by computational analysis of whole genome sequences of Arabidopsis thaliana plant and Caenorhabditis elegans (C. elegans) nematode, the identification and classification of these mobile genetic elements remain a challenge due to the fact that the wide majority of their families are non-autonomous. In C. elegans genome, DNA helitrons sequences possess great variability in terms of length that varies between 11 and 8965 base pairs (bps) from one sequence to another. In this work, we develop a new method to predict helitrons DNA-sequences, which is particularly based on Frequency Chaos Game Representation (FCGR) DNA-images. Thus, we introduce an automatic system in order to classify helitrons families in C. elegans genome, based on a combination between machine learning approaches and features extracted from DNA-sequences. Consequently, the new set of helitrons features (the FCGR images and K-mers) are extracted from DNA sequences. These helitrons features consist of the frequency apparition number of K nucleotides pairs (Tandem Repeat) in the DNA sequences. Indeed, three different classifiers are used for the classification of all existing helitrons families. The results have shown potential global score equal to 72.7% due to FCGR images which constitute helitrons features and the pre-trained neural network as a classifier. The two other classifiers demonstrate that their efficiency reaches 68.7% for Support Vector Machine (SVM) and 91.45% for Random Forest (RF) algorithms using the K-mers features corresponding to the genomic sequences.

Alignment-free genomic sequence comparison using FCGR and signal processing

BackgroundAlignment-free methods of genomic comparison offer the possibility of scaling to large data sets of nucleotide sequences comprised of several thousand or more base pairs. Such methods can be used for purposes of deducing “nearby” species in a reference data set, or for constructing phylogenetic trees.ResultsWe describe one such method that gives quite strong results. We use the Frequency Chaos Game Representation (FCGR) to create images from such sequences, We then reduce dimension, first using a Fourier trig transform, followed by a Singular Values Decomposition (SVD). This gives vectors of modest length. These in turn are used for fast sequence lookup, construction of phylogenetic trees, and classification of virus genomic data. We illustrate the accuracy and scalability of this approach on several benchmark test sets.ConclusionsThe tandem of FCGR and dimension reductions using Fourier-type transforms and SVD provides a powerful approach for alignment-free genomic comparison. Results compare favorably and often surpass best results reported in prior literature. Good scalability is also observed.