Representation Of Sequences Research Articles

RNA transcription and degradation of Alu retrotransposons depends on sequence features and evolutionary history.

Alu elements are one of the most successful groups of RNA retrotransposons and make up 11% of the human genome with over 1 million individual loci. They are linked to genetic defects, increases in sequence diversity, and influence transcriptional activity. Still, their RNA metabolism is poorly understood yet. It is even unclear whether Alu elements are mostly transcribed by RNA Polymerase II or III. We have conducted a transcription shutoff experiment by α-amanitin and metabolic RNA labeling by 4-thiouridine combined with RNA fragmentation (TT-seq) and RNA-seq to shed further light on the origin and life cycle of Alu transcripts. We find that Alu RNAs are more stable than previously thought and seem to originate in part from RNA Polymerase II activity, as previous reports suggest. Their expression however seems to be independent of the transcriptional activity of adjacent genes. Furthermore, we have developed a novel statistical test for detecting the expression of quantitative trait loci in Alu elements that relies on the de Bruijn graph representation of all Alu sequences. It controls for both statistical significance and biological relevance using a tuned k-mer representation, discovering influential sequence features missed by regular motif search. In addition, we discover several point mutations using a generalized linear model, and motifs of interest, which also match transcription factor-binding motifs.

Deep flight track clustering based on spatial–temporal distance and denoising auto-encoding

The rapid development of the aviation industry imposes an urgent need for airspace traffic management. Meaningful clustering of flight tracks is of paramount importance for efficient operation and management of increasingly complex aerial space and traffic. Two key components exist in track clustering: similarity metric and clustering method. Most of the existing studies on track similarity metrics only consider the spatial coordinates of the track points without taking into consideration of the rich information of the track data, such as flight heading and flight speed, on the measurement of track similarity. In addition, temporal properties and the derived features of the flight tracks shall be utilized to reveal the underlying patterns and overcome distortions from noise. In this paper, we propose a track similarity based on the spatial–temporal characteristics of flight tracks and a Deep Temporal Clustering method using a denoising autoencoder. Our proposed method employs the Deep Temporal Denoising Auto-encoding network to extract the latent representations of the track sequences. By extending the idea of k-means clustering, Deep Temporal Clustering groups the flight tracks with a Time Clustering Layer. Experiments are conducted using Automatic Dependent Surveillance-Broadcast track data. In comparison with classical and state-of-the-art methods, among all cases, our Deep Temporal Clustering method achieved a much-improved performance of more than 57.3%. When we introduce noise to the track records and increase its magnitude, the performance of our method degrades but the trend slows down as the noise magnitude increases. The change is less than 7% and, in some cases, is close to zero, which demonstrates the robustness of our method to noise.

Semiotic Thoughts on Biological Sequence Representations.

The deluge of biological sequences ranging from those of proteins, DNA and RNA to genomes has motivated to devise models to represent them, which are further used to contrast those sequences. Here we present a brief bibliometric description of the research area devoted to the representation of biological sequences and highlight the semiotic elements of this process. Finally, we argue that this research area must learn from the evolution of mathematical chemistry and try to avoid its pitfalls.

Numerical Characterization of DNA Sequences for Alignment-free Sequence Comparison - A Review.

Biological macromolecules, namely, DNA, RNA, and protein, have their building blocks organized in a particular sequence and the sequential arrangement encodes the evolutionary history of the organism (species). Hence, biological sequences have been used for studying evolutionary relationships among the species. This is usually carried out by Multiple Sequence Algorithms (MSA). Due to certain limitations of MSA, alignment-free sequence comparison methods were developed. The present review is on alignment-free sequence comparison methods carried out using the numerical characterization of DNA sequences. The graphical representation of DNA sequences by chaos game representation and other 2-dimensional and 3-dimensional methods are discussed. The evolution of numerical characterization from the various graphical representations and the application of the DNA invariants thus computed in phylogenetic analysis are presented. The extension of computing molecular descriptors in chemometrics to the calculation of a new set of DNA invariants and their use in alignment-free sequence comparison in an N-dimensional space and construction of phylogenetic trees are also reviewed. The phylogenetic tress constructed by the alignment-free sequence comparison methods using DNA invariants were found to be better than those constructed using alignment-based tools such as PHLYIP and ClustalW. One of the graphical representation methods is now extended to study viral sequences of infectious diseases for the identification of conserved regions to design peptidebased vaccines by combining numerical characterization and graphical representation.

Mapping Biomolecular Sequences: Graphical Representations - Their Origins, Applications and Future Prospects

The exponential growth in the depositories of biological sequence data has generated an urgent need to store, retrieve and analyse the data efficiently and effectively for which the standard practice of using alignment procedures are not adequate due to high demand on computing resources and time. Graphical representation of sequences has become one of the most popular alignment-free strategies to analyse the biological sequences where each basic unit of the sequences - the bases adenine, cytosine, guanine and thymine for DNA/RNA, and the 20 amino acids for proteins - are plotted on a multi-dimensional grid. The resulting curve in 2D and 3D space and the implied graph in higher dimensions provide a perception of the underlying information of the sequences through visual inspection; numerical analyses, in geometrical or matrix terms, of the plots provide a measure of comparison between sequences and thus enable study of sequence hierarchies. The new approach has also enabled studies of comparisons of DNA sequences over many thousands of bases and provided new insights into the structure of the base compositions of DNA sequences. In this article we review in brief the origins and applications of graphical representations and highlight the future perspectives in this field.

Genetic circuit design automation with Cello 2.0.

Cells interact with their environment, communicate among themselves, track time and make decisions through functions controlled by natural regulatory genetic circuits consisting of interacting biological components. Synthetic programmable circuits used in therapeutics and other applications can be automatically designed by computer-aided tools. The Cello software designs the DNA sequences for programmable circuits based on a high-level software description and a library of characterized DNA parts representing Boolean logic gates. This process allows for design specification reuse, modular DNA part library curation and formalized circuit transformations based on experimental data. This protocol describes Cello 2.0, a freely available cross-platform software written in Java. Cello 2.0 enables flexible descriptions of the logic gates' structure and their mathematical models representing dynamic behavior, new formal rules for describing the placement of gates in a genome, a new graphical user interface, support for Verilog 2005 syntax and a connection to the SynBioHub parts repository software environment. Collectively, these features expand Cello's capabilities beyond Escherichia coli plasmids to new organisms and broader genetic contexts, including the genome. Designing circuits with Cello 2.0 produces an abstract Boolean network from a Verilog file, assigns biological parts to each node in the Boolean network, constructs a DNA sequence and generates highly structured and annotated sequence representations suitable for downstream processing and fabrication, respectively. The result is a sequence implementing the specified Boolean function in the organism and predictions of circuit performance. Depending on the size of the design space and users' expertise, jobs may take minutes or hours to complete.

Determinantal and permanental representations of companion sequences associated to the r-Fibonacci sequence

Recently companion sequences of r-Fibonacci sequence were defined. The aim of this paper is to give some determinantal and permanental representations of these sequences via Hessenberg matrices. Several representations of classical sequences and polynomials are established. We conclude by using our representations to give n consecutive terms of companion sequences simultaneously.

ProtPlat: an efficient pre-training platform for protein classification based on FastText

BackgroundFor the past decades, benefitting from the rapid growth of protein sequence data in public databases, a lot of machine learning methods have been developed to predict physicochemical properties or functions of proteins using amino acid sequence features. However, the prediction performance often suffers from the lack of labeled data. In recent years, pre-training methods have been widely studied to address the small-sample issue in computer vision and natural language processing fields, while specific pre-training techniques for protein sequences are few.ResultsIn this paper, we propose a pre-training platform for representing protein sequences, called ProtPlat, which uses the Pfam database to train a three-layer neural network, and then uses specific training data from downstream tasks to fine-tune the model. ProtPlat can learn good representations for amino acids, and at the same time achieve efficient classification. We conduct experiments on three protein classification tasks, including the identification of type III secreted effectors, the prediction of subcellular localization, and the recognition of signal peptides. The experimental results show that the pre-training can enhance model performance effectively and ProtPlat is competitive to the state-of-the-art predictors, especially for small datasets. We implement the ProtPlat platform as a web service (https://compbio.sjtu.edu.cn/protplat) that is accessible to the public.ConclusionsTo enhance the feature representation of protein amino acid sequences and improve the performance of sequence-based classification tasks, we develop ProtPlat, a general platform for the pre-training of protein sequences, which is featured by a large-scale supervised training based on Pfam database and an efficient learning model, FastText. The experimental results of three downstream classification tasks demonstrate the efficacy of ProtPlat.

Mental Representations of Time in English Monolinguals, Mandarin Monolinguals, and Mandarin-English Bilinguals.

This study recruited English monolinguals, Mandarin monolinguals, and Mandarin–English (ME) bilinguals to examine whether native English and native Mandarin speakers think about time differently and whether the acquisition of L2 English could reshape native Mandarin speakers’ mental representations of temporal sequence. Across two experiments, we used the temporal congruency categorization paradigm which involved two-alternative forced-choice reaction time tasks to contrast experimental conditions that were assumed to be either compatible or incompatible with the internal spatiotemporal associations. Results add to previous studies by confirming that native English and native Mandarin speakers do think about time differently, and the significant crosslinguistic discrepancy primarily lies in the vertical representations of time flow. However, current findings also clarify the existing literature, demonstrating that the acquisition of L2 English does not appear to affect native Mandarin speakers’ temporal cognition. ME bilinguals, irrespective of whether they attained elementary or advanced level of English proficiency, exhibited temporal thinking patterns commensurate with those of Mandarin monolinguals. Some theoretical implications regarding the effect of bilingualism on cognition in general can be drawn from the present study, a crucial one being that it provides evidence against the view that L2 acquisition can reshape habitual modes of thinking established by L1.

Sequence representations after action-imagery practice of one-finger movements are effector-independent

Action-imagery practice (AIP) is often less effective than action-execution practice (AEP). We investigated whether this is due to a different time course of learning of different types of sequence representations in AIP and AEP. Participants learned to sequentially move with one finger to ten targets, which were visible the whole time. All six sessions started with a test. In the first four sessions, participants performed AIP, AEP, or control-practice (CP). Tests involved the practice sequence, a mirror sequence, and a different sequence, which were performed both with the practice hand and the other (transfer) hand. In AIP and AEP, movement times (MTs) in both hands were significantly shorter in the practice sequence than in the other sequences, indicating sequence-specific learning. In the transfer hand, this indicates effector-independent visual-spatial representations. The time course of the acquisition of effector-independent visual-spatial representations did not significantly differ between AEP and AIP. In AEP (but not in AIP), MTs in the practice sequence were significantly shorter in the practice hand than in the transfer hand, indicating effector-dependent representations. In conclusion, effector-dependent representations were not acquired after extensive AIP, which may be due to the lack of actual feedback. Therefore, AIP may replace AEP to acquire effector-independent visual-spatial representations, but not to acquire effector-dependent representations.

Exploring the sequence-structure map: joint embedding of protein sequence & structure

While the success of Google's AlphaFold2 (AF2) is widely recognized as a watershed moment in generative protein modeling, AF2 does not inherently have the ability to generate synthetic proteins with desired properties, nor does AF2's static 3D structure prediction give a characterization of protein dynamics. We seek to leverage the ability of AF2 to accurately predict protein structures to develop novel protein engineering artificial intelligence (AI). By embedding joint representations of protein sequences with their corresponding structures using autoencoder neural networks, we can encode protein sequence-structure relationships into a continuous latent variable space.

Algorithms for the Minimal Rational Fraction Representation of Sequences Revisited

Algorithms for the Minimal Rational Fraction Representation of Sequences Revisited

Motion primitives learning of ship-ship interaction patterns in encounter situations

Understanding the ship encounter situations is a tough task, since the uncertain motions of encountering ships and the long-lasting frequent interactions make the encounter processes dynamic and stochastic. If we can decompose these complex encounter processes into motion primitives, which represent the elementary interaction patterns, an encounter can be more easily understood and identified. In this study, we propose a two-stream Long Short-Term Memory-based autoencoder (LSTM-based AE) approach to extract motion primitives from multi-dimensional encounter data without specific rules and prior knowledge. This approach leverages the sequential representation ability of LSTM to learn the temporal dependencies of each encountering ship motion. Then, we developed a fusion gate named dynamic artificial potential field (DAPF) to fuse the outputs of two LSTMs and generated the high-level representation needed to capture the spatiotemporal relationships of the interactions. After that, through a clustering method, the motion primitives were automatically extracted to describe various interaction patterns. The effectiveness of this approach was validated by naturalistic encounter data. The low reconstruction errors of the LSTM-based AE demonstrated that high-level representations captured the relationships of the interactions in both temporal and spatial dimensions. Case studies demonstrated that the utilization of motion primitives provided a semantically interpretable technique to analyze the interaction patterns and encounter process, which is conducive to situation awareness and decision-making for developing intelligent ships.

A time series representation of protein sequences for similarity comparison

Based on the physicochemical indexes of 20 amino acids and the Hungarian algorithm, each amino acid was mapped into a vector. And, the protein sequence can be represented as time series in eleven-dimensional space. In addition, the DTW algorithm was applied to calculate the distance between two time series to compare the similarities of protein sequences. The validity and accuracy of this method was illustrated by similarity comparison of ND5 proteins of nine species. Furthermore, homology analysis of eleven ACE2 proteins, which included human, Malayan pangolin and six species of bats, confirmed that the human had shorter evolutionary distance from the pangolin than those bats. The phylogenetic tree of spike protein sequences of 36 coronaviruses, which were divided into five groups, Class I, Class II, Class III, SARS-CoVs and COVID-19, was constructed.

Multiple Protein Subcellular Locations Prediction Based on Deep Convolutional Neural Networks with Self-Attention Mechanism.

As an important research field in bioinformatics, protein subcellular location prediction is critical to reveal the protein functions and provide insightful information for disease diagnosis and drug development. Predicting protein subcellular locations remains a challenging task due to the difficulty of finding representative features and robust classifiers. Many feature fusion methods have been widely applied to tackle the above issues. However, they still suffer from accuracy loss due to feature redundancy. Furthermore, multiple protein subcellular locations prediction is more complicated since it is fundamentally a multi-label classification problem. The traditional binary classifiers or even multi-class classifiers cannot achieve satisfactory results. This paper proposes a novel method for protein subcellular location prediction with both single and multiple sites based on deep convolutional neural networks. Specifically, we first obtain the integrated features by simultaneously considering the pseudo amino acid, amino acid index distribution, and physicochemical property. We then adopt deep convolutional neural networks to extract high-dimensional features from the fused feature, removing the redundant preliminary features and gaining better representations of the raw sequences. Moreover, we use the self-attention mechanism and a customized loss function to ensure that the model is more inclined to positive data. In addition, we use randomk-label sets to reduce the number of prediction labels. Meanwhile, we employ a hybrid strategy of over-sampling and under-sampling to tackle the data imbalance problem. We compare our model with three representative classification alternatives. The experiment results show that our model achieves the best performance in terms of accuracy, demonstrating the efficacy of the proposed model.

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.

Recurrent Multinomial Models for Categorical Sequences

This paper presents a model of recurrent multinomial sequences. Though there exists a quite considerable literature on modeling autocorrelation in numerical data and sequences of categorical outcomes, there is currently no systematic method of modeling patterns of recurrence in categorical sequences. This paper develops a means of discovering recurrent patterns by employing a more restrictive Markov assumption. The resulting model, which I call the recurrent multinomial model, provides a parsimonious representation of recurrent sequences, enabling the investigation of recurrences on longer time scales than existing models. The utility of recurrent multinomial models is demonstrated by applying them to the case of conversational turn-taking in meetings of the Federal Open Market Committee (FOMC). Analyses are effectively able to discover norms around turn-reclaiming, participation, and suppression and to evaluate how these norms vary throughout the course of the meeting.

Dual Co-Attention-Based Multi-Feature Fusion Method for Rumor Detection

Social media has become more popular these days due to widely used instant messaging. Nevertheless, rumor propagation on social media has become an increasingly important issue. The purpose of this study is to investigate the impact of various features in social media on rumor detection, propose a dual co-attention-based multi-feature fusion method for rumor detection, and explore the detection capability of the proposed method in early rumor detection tasks. The proposed BERT-based Dual Co-attention Neural Network (BDCoNN) method for rumor detection, which uses BERT for word embedding. It simultaneously integrates features from three sources: publishing user profiles, source tweets, and comments. In the BDCoNN method, user discrete features and identity descriptors in user profiles are extracted using a one-dimensional convolutional neural network (CNN) and TextCNN, respectively. The bidirectional gate recurrent unit network (BiGRU) with a hierarchical attention mechanism is used to learn the hidden layer representation of tweet sequence and comment sequence. A dual collaborative attention mechanism is used to explore the correlation among publishing user profiles, tweet content, and comments. Then the feature vector is fed into classifier to identify the implicit differences between rumor spreaders and non-rumor spreaders. In this study, we conducted several experiments on the Weibo and CED datasets collected from microblog. The results show that the proposed method achieves the state-of-the-art performance compared with baseline methods, which is 5.2% and 5% higher than the dEFEND. The F1 value is increased by 4.4% and 4%, respectively. In addition, this paper conducts research on early rumor detection tasks, which verifies the proposed method detects rumors more quickly and accurately than competitors.

Mixed Information Flow for Cross-Domain Sequential Recommendations

Cross-domain sequential recommendation is the task of predict the next item that the user is most likely to interact with based on past sequential behavior from multiple domains. One of the key challenges in cross-domain sequential recommendation is to grasp and transfer the flow of information from multiple domains so as to promote recommendations in all domains. Previous studies have investigated the flow of behavioral information by exploring the connection between items from different domains. The flow of knowledge (i.e., the connection between knowledge from different domains) has so far been neglected. In this article, we propose a mixed information flow network for cross-domain sequential recommendation to consider both the flow of behavioral information and the flow of knowledge by incorporating a behavior transfer unit and a knowledge transfer unit . The proposed mixed information flow network is able to decide when cross-domain information should be used and, if so, which cross-domain information should be used to enrich the sequence representation according to users’ current preferences. Extensive experiments conducted on four e-commerce datasets demonstrate that the proposed mixed information flow network is able to improve recommendation performance in different domains by modeling mixed information flow. In this article, we focus on the application of mixed information flow network s to a scenario with two domains, but the method can easily be extended to multiple domains.

Session interest model for CTR prediction based on self-attention mechanism

Click-through rate prediction, which aims to predict the probability of the user clicking on an item, is critical to online advertising. How to capture the user evolving interests from the user behavior sequence is an important issue in CTR prediction. However, most existing models ignore the factor that the sequence is composed of sessions, and user behavior can be divided into different sessions according to the occurring time. The user behaviors are highly correlated in each session and are not relevant across sessions. We propose an effective model for CTR prediction, named Session Interest Model via Self-Attention (SISA). First, we divide the user sequential behavior into session layer. A self-attention mechanism with bias coding is used to model each session. Since different session interest may be related to each other or follow a sequential pattern, next, we utilize gated recurrent unit (GRU) to capture the interaction and evolution of user different historical session interests in session interest extractor module. Then, we use the local activation and GRU to aggregate their target ad to form the final representation of the behavior sequence in session interest interacting module. Experimental results show that the SISA model performs better than other models.