Classification of Highly Divergent Viruses from DNA/RNA Sequence Using Transformer-Based Models.

Viruses infect millions of people worldwide each year, and some can lead to cancer or increase the risk of cancer. As viruses have highly mutable genomes, new viruses may emerge in the future, such as COVID-19 and influenza. Traditional virology relies on predefined rules to identify viruses, but new viruses may be completely or partially divergent from the reference genome, rendering statistical methods and similarity calculations insufficient for all genome sequences. Identifying DNA/RNA-based viral sequences is a crucial step in differentiating different types of lethal pathogens, including their variants and strains. While various tools in bioinformatics can align them, expert biologists are required to interpret the results. Computational virology is a scientific field that studies viruses, their origins, and drug discovery, where machine learning plays a crucial role in extracting domain- and task-specific features to tackle this challenge. This paper proposes a genome analysis system that uses advanced deep learning to identify dozens of viruses. The system uses nucleotide sequences from the NCBI GenBank database and a BERT tokenizer to extract features from the sequences by breaking them down into tokens. We also generated synthetic data for viruses with small sample sizes. The proposed system has two components: a scratch BERT architecture specifically designed for DNA analysis, which is used to learn the next codons unsupervised, and a classifier that identifies important features and understands the relationship between genotype and phenotype. Our system achieved an accuracy of 97.69% in identifying viral sequences.

Evaluating COVID-19 Sequence Data Using Nearest-Neighbors Based Network Model

The SARS-CoV-2 coronavirus is the cause of the COVID-19 disease in humans. Like many coronaviruses, it can adapt to different hosts and evolve into different lineages. It is well-known that the major SARS-CoV-2 lineages are characterized by mutations that happen predominantly in the spike protein. Understanding the spike protein structure and how it can be perturbed is vital for understanding and determining if a lineage is of concern. These are crucial to identifying and controlling current outbreaks and preventing future pandemics. Machine learning (ML) methods are a viable solution to this effort, given the volume of available sequencing data, much of which is unaligned or even unassembled. However, such ML methods require fixed-length numerical feature vectors in Euclidean space to be applicable. Similarly, euclidean space is not considered the best choice when working with the classification and clustering tasks for biological sequences. For this purpose, we design a method that converts the protein (spike) sequences into the sequence similarity network (SSN). We can then use SSN as an input for the classical algorithms from the graph mining domain for the typical tasks such as classification and clustering to understand the data. We show that the proposed alignment-free method is able to outperform the current SOTA method in terms of clustering results. Similarly, we are able to achieve higher classification accuracy using well-known Node2Vec-based embedding compared to other baseline embedding approaches. © 2022 IEEE.

Characterizing SARS-CoV-2 Spike Sequences Based on Geographical Location.

With the rapid spread of COVID-19 worldwide, viral genomic data are available in the order of millions of sequences on public databases such as GISAID. This Big Data creates a unique opportunity for analysis toward the research of effective vaccine development for current pandemics, and avoiding or mitigating future pandemics. One piece of information that comes with every such viral sequence is the geographical location where it was collected-the patterns found between viral variants and geographical location surely being an important part of this analysis. One major challenge that researchers face is processing such huge, highly dimensional data to obtain useful insights as quickly as possible. Most of the existing methods face scalability issues when dealing with the magnitude of such data. In this article, we propose an approach that first computes a numerical representation of the spike protein sequence of SARS-CoV-2 using k-mers (substrings) and then uses several machine learning models to classify the sequences based on geographical location. We show that our proposed model significantly outperforms the baselines. We also show the importance of different amino acids in the spike sequences by computing the information gain corresponding to the true class labels.

Efficient Reconciliation of Genomic Datasets of High Similarity

We apply Invertible Bloom Lookup Tables (IBLTs) to the comparison of k-mer sets originated from large DNA sequence datasets. We show that for similar datasets, IBLTs provide a more space-efficient and, at the same time, more accurate method for estimating Jaccard similarity of underlying k-mer sets, compared to MinHash which is a go-to sketching technique for efficient pairwise similarity estimation. This is achieved by combining IBLTs with k-mer sampling based on syncmers, which constitute a context-independent alternative to minimizers and provide an unbiased estimator of Jaccard similarity. A key property of our method is that involved data structures require space proportional to the difference of k-mer sets and are independent of the size of sets themselves. As another application, we show how our ideas can be applied in order to efficiently compute (an approximation of) k-mers that differ between two datasets, still using space only proportional to their number. We experimentally illustrate our results on both simulated and real data (SARS-CoV-2 and Streptococcus Pneumoniae genomes). © Yoshihiro Shibuya, Djamal Belazzougui, and Gregory Kucherov.

Spike2Vec: An Efficient and Scalable Embedding Approach for COVID-19 Spike Sequences

With the rapid global spread of COVID-19, more and more data related to this virus is becoming available, including genomic sequence data. The total number of genomic sequences that are publicly available on platforms such as GISAID is currently several million, and is increasing with every day. The availability of such Big Data creates a new opportunity for researchers to study this virus in detail. This is particularly important with all of the dynamics of the COVID-19 variants which emerge and circulate. This rich data source will give us insights on the best ways to perform genomic surveillance for this and future pandemic threats, with the ultimate goal of mitigating or eliminating such threats. Analyzing and processing the several million genomic sequences is a challenging task. Although traditional methods for sequence classification are proven to be effective, they are not designed to deal with these specific types of genomic sequences. Moreover, most of the existing methods also face the issue of scalability. Previous studies which were tailored to coronavirus genomic data proposed to use spike sequences (corresponding to a subsequence of the genome), rather than using the complete genomic sequence, to perform different machine learning (ML) tasks such as classification and clustering. However, those methods suffer from scalability issues. In this paper, we propose an approach called Spike2Vec, an efficient and scalable feature vector representation for each spike sequence that can be used for downstream ML tasks. Through experiments, we show that Spike2Vec is not only scalable on several million spike sequences, but also outperforms the baseline models in terms of prediction accuracy, F1 score, etc. Since this type of study on such huge numbers of spike sequences has not been done before (to the best of our knowledge), we believe that it will open new doors for researchers to use this data and perform different tasks to unfold new information that was not available before. We also use information gain (IG) to compute the importance of each amino acid in the spike sequence. The amino acids with higher IG values tend to be the same as many reported by the USA based Centers for Disease Control and Prevention (CDC) for different variants.

Prediction of binding miRNAs involved with immune genes to the SARS-CoV-2 by using sequence features extraction and One-class SVM.

The prediction of host human miRNA binding to the SARS-COV-2-CoV-2 RNA sequence is of particular interest. This biological process could lead to virus repression, serve as biomarkers for diagnosis, or as potential treatments for this disease. One source of concern is attempting to uncover the viral regions in which this binding could occur, as well as how these miRNAs binding could affect the SARS-COV-2 virus's processes. Using extracted sequence features from this base pairing, we predicted the relationships between miRNAs that interact with genes involved in immune function and bind to the SARS-COV-2 genome in their 5' UTR region. We compared two supervised models, SVM and Random Forest, with an unsupervised One-Class SVM. When the results of the confusion matrices were inspected, the results of the supervised models were misleading, resulting in a Type II error. However, with the latter model, we achieved an average accuracy of 92%, sensitivity of 96.18%, and specificity of 78%. We hypothesize that studying the bind of miRNAs that affect immunological genes and bind to the SARS-COV-2 virus will lead to potential genetic therapies for fighting the disease or understanding how the immune system is affected when this type of viral infection occurs.

Effective and scalable clustering of SARS-CoV-2 sequences

SARS-CoV-2, like any other virus, continues to mutate as it spreads, according to an evolutionary process. Unlike any other virus, the number of currently available sequences of SARS-CoV-2 in public databases such as GISAID is already several million. This amount of data has the potential to uncover the evolutionary dynamics of a virus like never before. However, a million is already several orders of magnitude beyond what can be processed by the traditional methods designed to reconstruct a virus's evolutionary history, such as those that build a phylogenetic tree. Hence, new and scalable methods will need to be devised in order to make use of the ever increasing number of viral sequences being collected. Since identifying variants is an important part of understanding the evolution of a virus, in this paper, we propose an approach based on clustering sequences to identify the current major SARS-CoV-2 variants. Using a k-mer based feature vector generation and efficient feature selection methods, our approach is effective in identifying variants, as well as being efficient and scalable to millions of sequences. Such a clustering method allows us to show the relative proportion of each variant over time, giving the rate of spread of each variant in different locations - something which is important for vaccine development and distribution. We also compute the importance of each amino acid position of the spike protein in identifying a given variant in terms of information gain. Positions of high variant-specific importance tend to agree with those reported by the USA's Centers for Disease Control and Prevention (CDC), further demonstrating our approach. © 2021 ACM.

PWM2Vec: An Efficient Embedding Approach for Viral Host Specification from Coronavirus Spike Sequences.

The study of host specificity has important connections to the question about the origin of SARS-CoV-2 in humans which led to the COVID-19 pandemic-an important open question. There are speculations that bats are a possible origin. Likewise, there are many closely related (corona)viruses, such as SARS, which was found to be transmitted through civets. The study of the different hosts which can be potential carriers and transmitters of deadly viruses to humans is crucial to understanding, mitigating, and preventing current and future pandemics. In coronaviruses, the surface (S) protein, or spike protein, is important in determining host specificity, since it is the point of contact between the virus and the host cell membrane. In this paper, we classify the hosts of over five thousand coronaviruses from their spike protein sequences, segregating them into clusters of distinct hosts among birds, bats, camels, swine, humans, and weasels, to name a few. We propose a feature embedding based on the well-known position weight matrix (PWM), which we call PWM2Vec, and we use it to generate feature vectors from the spike protein sequences of these coronaviruses. While our embedding is inspired by the success of PWMs in biological applications, such as determining protein function and identifying transcription factor binding sites, we are the first (to the best of our knowledge) to use PWMs from viral sequences to generate fixed-length feature vector representations, and use them in the context of host classification. The results on real world data show that when using PWM2Vec, machine learning classifiers are able to perform comparably to the baseline models in terms of predictive performance and runtime-in some cases, the performance is better. We also measure the importance of different amino acids using information gain to show the amino acids which are important for predicting the host of a given coronavirus. Finally, we perform some statistical analyses on these results to show that our embedding is more compact than the embeddings of the baseline models.

A Computational Framework for Pattern Detection on Unaligned Sequences: An Application on SARS-CoV-2 Data.

The exponential growth of genome sequences available has spurred research on pattern detection with the aim of extracting evolutionary signal. Traditional approaches, such as multiple sequence alignment, rely on positional homology in order to reconstruct the phylogenetic history of taxa. Yet, mining information from the plethora of biological data and delineating species on a genetic basis, still proves to be an extremely difficult problem to consider. Multiple algorithms and techniques have been developed in order to approach the problem multidimensionally. Here, we propose a computational framework for identifying potentially meaningful features based on k-mers retrieved from unaligned sequence data. Specifically, we have developed a process which makes use of unsupervised learning techniques in order to identify characteristic k-mers of the input dataset across a range of different k-values and within a reasonable time frame. We use these k-mers as features for clustering the input sequences and identifying differences between the distributions of k-mers across the dataset. The developed algorithm is part of an innovative and much promising approach both to the problem of grouping sequence data based on their inherent characteristic features, as well as for the study of changes in the distributions of k-mers, as the k-value is fluctuating within a range of values. Our framework is fully developed in Python language as an open source software licensed under the MIT License, and is freely available at https://github.com/BiodataAnalysisGroup/kmerAnalyzer.