Biology subjects

Vikalo, H.

Publications and source records attributed to Vikalo, H..

7 recordsLinked to original sources

bioRxiv2023

XVir: A Transformer-Based Architecture for Identifying Viral Reads from Cancer Samples

It is estimated that approximately 15% of cancers world-wide can be linked to viral infections. The viruses that can cause or increase the risk of cancer include human papillomavirus, hepatitis B and C viruses, Epstein-Barr virus, and human immunodeficiency virus, to name a few. The computational analysis of the massive amounts of tumor DNA data, whose collection is enabled by the recent advancements in sequencing technologies, have allowed studies of the potential association between cancers and viral pathogens. However, the high diversity of oncoviral families makes reliable detection of viral DNA difficult and thus, renders such analysis challenging. In this paper, we introduce XVir, a data pipeline that relies on a transformer-based deep learning architecture to reliably identify viral DNA present in human tumors. In particular, XVir is trained on genomic sequencing reads from viral and human genomes and may be used with tumor sequence information to find evidence of viral DNA in human cancers. Results on semi-experimental data demonstrate that XVir is capable of achieving high detection accuracy, generally outperforming state-of-the-art competing methods while being more compact and less computationally demanding. CCS CONCEPTS* Computer systems organization [->] Embedded systems; Redundancy; Robotics; * Networks [->] Network reliability. ACM Reference FormatShorya Consul, John Robertson, and Haris Vikalo. 2023. XVir: A Transformer-Based Architecture for Identifying Viral Reads from Cancer Samples. In Proceedings of The Eighth International Workshop on Computational Network Biology: Modeling, Analysis, and Control (CNB-MAC 23). ACM, New York, NY, USA, 8 pages.

bioinformatics↗

bioRxiv2022

Graph-Based Reconstruction and Analysis of Disease Transmission Networks using Viral Genomic Data

Understanding the patterns of viral disease transmissions helps establish public health policies and aids in controlling and ending a disease outbreak. Classical methods for studying disease transmission dynamics that rely on epidemiological data, such as times of sample collection and duration of exposure intervals, struggle to provide desired insight due to limited informativeness of such data. A more precise characterization of disease transmissions may be acquired from sequencing data that reveals genetic distance between viral populations in patient samples. Indeed, genetic distance between viral strains present in hosts contains valuable information about transmission history, thus motivating the design of methods that rely on genomic data to reconstruct a directed disease transmission network, detect transmission clusters, and identify significant network nodes (e.g., super-spreaders). In this paper, we present a novel end-to-end framework for the analysis of viral transmissions utilizing viral genomic (sequencing) data. The proposed framework groups infected hosts into transmission clusters based on reconstructed viral quasispecies; the genetic distance between a pair of hosts is calculated using Earth Movers Distance, and further used to infer transmission direction between the hosts. To quantify the significance of a host in the transmission network, the importance score is calculated by a graph convolutional auto-encoder. The viral transmission network is represented by a directed minimum spanning tree utilizing the Edmonds algorithm modified to incorporate constraints on the importance scores of the hosts. Results on realistic synthetic as well as experimental data demonstrate that the proposed framework outperforms state-of-the-art techniques for the analysis of viral transmission dynamics. CCS CONCEPTSO_LIApplied computing [->] Bioinformatics. C_LI ACM Reference FormatZiqi Ke and Haris Vikalo. 2022. Graph-Based Reconstruction and Analysis of Disease Transmission Networks using Viral Genomic Data. In Proceedings of The Seventh International Workshop on Computational Network Biology (CNB-MAC 2022). ACM, New York, NY, USA, 10 pages. https://doi.org/XXXXXXX.XXXXXXX

genomics↗

bioRxiv2020

A Convolutional Auto-Encoder for Haplotype Assembly and Viral Quasispecies Reconstruction

Haplotype assembly and viral quasispecies reconstruction are challenging tasks concerned with analysis of genomic mixtures using sequencing data. High-throughput sequencing technologies generate enormous amounts of short fragments (reads) which essentially oversample components of a mixture; the representation redundancy enables reconstruction of the components (haplotypes, viral strains). The reconstruction problem, known to be NP-hard, boils down to grouping together reads originating from the same component in a mixture. Existing methods struggle to solve this problem with required level of accuracy and low runtimes; the problem is becoming increasingly more challenging as the number and length of the components increase. This paper proposes a read clustering method based on a convolutional auto-encoder designed to first project sequenced fragments to a low-dimensional space and then estimate the probability of the read origin using learned embedded features. The components are reconstructed by finding consensus sequences that agglomerate reads from the same origin. Mini-batch stochastic gradient descent and dimension reduction of reads allow the proposed method to efficiently deal with massive numbers of long reads. Experiments on simulated, semi-experimental and experimental data demonstrate the ability of the proposed method to accurately reconstruct haplotypes and viral quasispecies, often demonstrating superior performance compared to state-of-the-art methods.

bioinformatics↗

bioRxiv2019

A Graph Auto-Encoder for Haplotype Assembly and Viral Quasispecies Reconstruction

Reconstructing components of a genomic mixture from data obtained by means of DNA sequencing is a challenging problem encountered in a variety of applications including single individual haplotyping and studies of viral communities. High-throughput DNA sequencing platforms oversample mixture components to provide massive amounts of reads whose relative positions can be determined by mapping the reads to a known reference genome; assembly of the components, however, requires discovery of the reads origin - an NP-hard problem that the existing methods struggle to solve with the required level of accuracy. In this paper, we present a learning framework based on a graph auto-encoder designed to exploit structural properties of sequencing data. The algorithm is a neural network which essentially trains to ignore sequencing errors and infers the posterior probabilities of the origin of sequencing reads. Mixture components are then reconstructed by finding consensus of the reads determined to originate from the same genomic component. Results on realistic synthetic as well as experimental data demonstrate that the proposed framework reliably assembles haplotypes and reconstructs viral communities, often significantly outperforming state-of-the-art techniques. Source codes and datasets are publicly available at https://github.com/WuLoli/GAEseq.

bioinformatics↗

bioRxiv2017

Sparse Tensor Decomposition For Haplotype Assembly Of Diploids And Polyploids

A framework that formulates haplotype assembly as sparse tensor decomposition is proposed. The problem is cast as that of decomposing a tensor having special structural constraints and missing a large fraction of its entries into a product of two factors, U and [Formula]; tensor [Formula] reveals haplotype information while U is a sparse matrix encoding the origin of erroneous sequencing reads. An algorithm, AltHap, which reconstructs haplotypes of either diploid or poly-ploid organisms by solving this decomposition problem is proposed. Starting from a judiciously selected initial point, AltHap alternates between two optimization tasks to recover U and [Formula] by relying on a modified gradient descent search that exploits salient structural properties of U and [Formula]. The performance and convergence properties of AltHap are theoretically analyzed and, in doing so, guarantees on the achievable minimum error correction scores and correct phasing rate are established. AltHap was tested in a number of different scenarios and was shown to compare favorably to state-of-the-art methods in applications to haplotype assembly of diploids, and significantly outperform existing techniques when applied to haplotype assembly of polyploids.

bioinformatics↗

bioRxiv2017

aBayesQR: A Bayesian method for reconstruction of viral populations characterized by low diversity

RNA viruses replicate with high mutation rates, creating closely related viral populations. The heterogeneous virus populations, referred to as viral quasispecies, rapidly adapt to environmental changes thus adversely affecting efficiency of antiviral drugs and vaccines. Therefore, studying the underlying genetic heterogeneity of viral populations plays a significant role in the development of effective therapeutic treatments. Recent high-throughput sequencing technologies have provided invaluable opportunity for uncovering the structure of quasispecies populations (i.e., reconstruction of viral sequences and discovery of their relative frequencies). However, accurate reconstruction of viral quasispecies remains difficult due to limited read-lengths and presence of sequencing errors. The problem is particularly challenging when the strains in a population are highly similar, i.e., the sequences are characterized by low mutual genetic distances, and further exacerbated if some of those strains are relatively rare; this is the setting where state-of-the-art methods struggle. In this paper, we present a novel viral quasispecies reconstruction algorithm, aBayesQR, that employs a maximum-likelihood framework to infer individual sequences in a mixture from high-throughput sequencing data. The search for the most likely quasispecies is conducted on long contigs that our method constructs from the set of short reads via agglomerative hierarchical clustering; operating on contigs rather than short reads enables identification of close strains in a population and provides computational tractability of the Bayesian method. Results on both simulated and real HIV-1 data demonstrate that the proposed algorithm generally outperforms state-of-the-art methods; aBayesQR particularly stands out when reconstructing a set of closely related viral strains (e.g., quasispecies characterized by low diversity).

bioinformatics↗

bioRxiv2016

Viral Quasispecies Reconstruction via Correlation Clustering

RNA viruses are characterized by high mutation rates that give rise to populations of closely related viral genomes, the so-called viral quasispecies. The underlying genetic heterogeneity occurring as a result of natural mutation-selection process enables the virus to adapt and proliferate in face of changing conditions over the course of an infection. Determining genetic diversity (i.e., inferring viral haplotypes and their proportions in the population) of an RNA virus is essential for the understanding of its origin and mutation patterns, and the development of effective drug treatments. In this paper we present QSdpR, a novel correlation clustering formulation of the quasispecies reconstruction problem which relies on semidefinite programming to accurately estimate the sub-species and their frequencies in a mixed population. Extensive comparisons with existing methods are presented on both synthetic and real data, demonstrating efficacy and superior performance of QSdpR.

bioinformatics↗