SEARCH · bioRxiv Science

Results for “Bioinformatics”

Search indexed bioRxiv preprints in genomics, neuroscience, cell biology and bioinformatics. Read source abstracts and check manuscript versions; preprints are not peer reviewed.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,549 records · Page 86Linked to original sources

bioRxiv2022

Using unsupervised learning algorithms to identify essential genes associated with SARS-CoV-2 as potential therapeutic targets for COVID-19

MotivationSevere acute respiratory syndrome coronavirus 2 (SARS-CoV-2) requires the fast discovery of effective treatments to fight this worldwide concern. Several genes associated with the SARS-CoV-2, which are essential for its functionality, pathogenesis, and survival, have been identified. These genes, which play crucial roles in SARS-CoV-2 infection, are considered potential therapeutic targets. Developing drugs against these essential genes to inhibit their regular functions could be a good approach for COVID-19 treatment. Artificial intelligence and machine learning methods provide powerful infrastructures for interpreting and understanding the available data and can assist in finding fast explanations and cures. ResultsWe propose a method to highlight the essential genes that play crucial roles in SARS-CoV-2 pathogenesis. For this purpose, we define eleven informative topological and biological features for the biological and PPI networks constructed on gene sets that correspond to COVID-19. Then, we use three different unsupervised learning algorithms with different approaches to rank the important genes with respect to our defined informative features. Finally, we present a set of 18 important genes related to COVID-19. AvailabilityMaterials and implementations are available at: https://github.com/MahnazHabibi/Gene_analysis. Contactm_habibi@qiau.ac.ir Supplementary informationSupplementary data are available at Bioinformatics online.

bioinformatics↗

bioRxiv2022

Succinct k-mer Set Representations Using Subset Rank Queries on the Spectral Burrows-Wheeler Transform (SBWT)

The k-spectrum of a string is the set of all distinct substrings of length k occurring in the string. This is a lossy but computationally convenient representation of the information in the string, with many applications in high-throughput bioinformatics. In this work, we define the notion of the Spectral Burrows-Wheeler Transform (SBWT), which is a sequence of subsets of the alphabet of the string encoding the k-spectrum of the string. The SBWT is a distillation of the ideas found in the BOSS and Wheeler graph data structures. We explore multiple different approaches to index the SBWT for membership queries on the underlying k-spectrum. We identify subset rank queries as the essential subproblem, and propose four succinct index structures to solve it. One of the approaches essentially leads to the known BOSS data structure, while the other three offer attractive time-space trade-offs and support simpler query algorithms that rely only on fast rank queries. The most general approach involves a novel data structure we call the subset wavelet tree, which we find to be of independent interest. All of the approaches are also amendable to entropy compression, which leads to good space bounds on the sizes of the data structures. Using entropy compression, we show that the SBWT can support membership queries on the k-spectrum of a single string in O(k) time and (n + k)(log{sigma} + 1/ ln 2) + o((n + k){sigma}) bits of space, where n is the number of distinct substrings of length k in the input and{sigma} is the size of the alphabet. This improves from the time O(k log{sigma} ) achieved by the BOSS data structure. We show, via experiments on a range of genomic data sets, that the simplicity of our new indexes translates into large performance gains in practice over prior art.

bioinformatics↗

bioRxiv2022

DProQ: A Gated-Graph Transformer for Protein Complex Structure Assessment

MotivationProteins interact to form complexes to carry out essential biological functions. Computational methods such as AlphaFold-multimer have been developed to predict the quaternary structures of protein complexes. An important yet largely unsolved challenge in protein complex structure prediction is to accurately estimate the quality of predicted protein complex structures without any knowledge of the corresponding native structures. Such estimations can then be used to select high-quality predicted complex structures to facilitate biomedical research such as protein function analysis and drug discovery. ResultsIn this work, we introduce a new gated neighborhood-modulating graph transformer to predict the quality of 3D protein complex structures. It incorporates node and edge gates within a graph transformer framework to control information flow during graph message passing. We trained, evaluated and tested the method (called DProQA) on newly-curated protein complex datasets before the 15th Critical Assessment of Techniques for Protein Structure Prediction (CASP15) and then blindly tested it in the 2022 CASP15 experiment. The method was ranked 3rd among the single-model quality assessment methods in CASP15 in terms of the ranking loss of TM-score on 36 complex targets. The rigorous internal and external experiments demonstrate that DProQA is effective in ranking protein complex structures. AvailabilityThe source code, data, and pre-trained models are available at https://github.com/jianlin-cheng/DProQA Contactchengji@missouri.edu Supplementary informationSupplementary data are available at Bioinformatics online.

bioinformatics↗

bioRxiv2022

The Cancermuts software package for the prioritization of missense cancer variants: a case study of AMBRA1 in melanoma

Cancer genomics and cancer mutation databases have made a wealth of information about missense mutations found in cancer patient samples. Contextualizing by means of annotation and predicting the effect of amino acid change help identify which ones are more likely to have a pathogenic impact. Those can be validated by means of experimental approaches that assess the impact of protein mutations on the cellular functions or their tumorigenic potential. Here, we propose the integrative bioinformatic approach Cancermuts, implemented as a Python package. Cancermuts is able to gather known missense cancer mutations from databases such as cBioPortal and COSMIC, and annotate them with the pathogenicity score REVEL as well as information on their source. It is also able to add annotations about the protein context these mutations are found in, such as post-translational modification sites, structured/ustructured regions, presence of short linear motifs and more. We applied Cancermuts to the intrinsically disordered protein AMBRA1, a key regulator of many cellular processes tightly (de)regulated in cancer. By these means, we classified mutations of AMBRA1 in melanoma, where AMBRA1 is highly mutated and displays a tumor-suppressive role. Next, based on REVEL score, position along the sequence and their local context, we applied cellular and molecular approaches to validate the predicted pathogenicity of a subset of mutations in an in vitro melanoma model. By doing so, we have identified two AMBRA1 mutations which show enhanced tumorigenic potential and are worth further investigation, highlighting the usefulness of the tool. Cancermuts can be used on any protein targets starting from minimal information, and it is available at https://www.github.com/ELELAB/cancermuts as free software.

bioinformatics↗

bioRxiv2022

Improving and evaluating deep learning models of cellular organization

MotivationCells contain dozens of major organelles and thousands of other structures, many of which vary extensively in their number, size, shape and spatial distribution. This complexity and variation dramatically complicates the use of both traditional and deep learning methods to build accurate models of cell organization. Most cellular organelles are distinct objects with defined boundaries that do not overlap, while the pixel resolution of most imaging methods is not sufficient to resolve these boundaries. Thus while cell organization is conceptually object-based, most current methods are pixel-based. Using extensive image collections in which particular organelles were fluorescently-labeled, deep learning methods can be used to build conditional autoencoder models for particular organelles. A major advance occurred with the use of a U-net approach to make multiple models all conditional upon a common reference, unlabeled image, allowing the relationships between different organelles to be at least partially inferred. ResultsWe have developed improved GAN-based approaches for learning these models and have also developed novel criteria for evaluating how well synthetic cell images reflect the properties of real images. The first set of criteria measure how well models preserve the expected property that organelles do not overlap. We also developed a modified loss function that allows retraining of the models to minimize that overlap. The second set of criteria uses object-based modeling to compare object shape and spatial distribution between synthetic and real images. Our work provides the first demonstration that, at least for some organelles, deep learning models can capture object-level properties of cell images. AvailabilityA Reproducible Research Archive containing all source code, generated images and analysis results will be made available at http://murphylab.cbd.cmu.edu/Software upon publication. Contactmurphy@cmu.edu Supplementary informationSupplementary data are available at Bioinformatics online.

bioinformatics↗

bioRxiv2022

plASgraph - using graph neural networks to detect plasmid contigs from an assembly graph

Identification of plasmids from sequencing data is an important and challenging problem related to antimicrobial resistance spread and other One-Health issues. In our work, we provide a new architecture for identifying plasmid contigs in fragmented genome assemblies built from short-read data. Unlike previous machine-learning approaches for this problem, which classify individual contigs separately, we employ graph neural networks (GNNs) to include information from the assembly graph. Propagation of information from nearby nodes in the graph allows accurate classification of even short contigs that are difficult to classify based on sequence features or database searches alone. Our new species-agnostic software tool plASgraph outperforms recently developed PlasForest, which uses database searches to supplement sequence-based features. Since our tool does not rely on existing plasmid databases, it is more suitable for classification of contigs in novel species and discovery of previously unknown plasmid sequences. Our tool can also be trained on a specific species, and in that scenario it outperforms mlplasmids trained on the same species. On one hand, our work provides a new, accurate, and easy to use tool for plasmid classification; on the other hand, it serves as a motivation for more widespread use of GNNs in bioinformatics, such as in pangenome sequence analysis, where sequence graphs serve as a fundamental data structure. Availabilityhttps://github.com/cchauve/plASgraph

bioinformatics↗

bioRxiv2022

Estimating somatic mutation rates by Duplex Sequencing in non-model organisms: Daphnia magna as a case study

Somatic mutations are evolutionarily important as determinants of individual organismal fitness, as well as being a focus of clinical research on age-related disease, such as cancer. Identifying somatic mutations and quantifying mutation rates, however, is extremely challenging and genome-wide somatic mutation rates have only been reported for a few model organisms. Here, we describe the application of Duplex Sequencing on bottlenecked WGS libraries to quantify genome-wide somatic base substitution rates in Daphnia magna. Daphnia, historically an ecological model system, has more recently been the focus of mutation studies, in part because of its high germline mutation rates. Using our protocol and pipeline, we estimate a somatic mutation rate of 2.14 x 10-7 substitutions per site (in a genotype where the germline rate is 3.60 x 10-9 substitutions per site per generation). To obtain this estimate, we tested multiple dilution levels to maximize sequencing efficiency, and developed bioinformatic filters needed to minimize false positives when a high quality reference genome is not available. In addition to laying the groundwork for estimating genotypic variation in rates of somatic mutations within D. magna, we provide a framework for quantifying somatic mutations in other non-model systems, and also highlight recent innovations to single molecule sequencing that will help to further refine such estimates.

bioinformatics↗

bioRxiv2022

scMuffin: an R package for resolving solid tumor heterogeneity from single-cell expression data

INTRODUCTIONSingle-cell (SC) gene expression analysis is crucial to dissect the complex cellular heterogeneity of solid tumors, which is one of the main obstacles for the development of effective cancer treatments. Such tumors typically contain a mixture of cells with aberrant genomic and transcriptomic profiles affecting specific sub-populations that might have a pivotal role in cancer progression, whose identification eludes bulk RNA-sequencing approaches. We presentscMuffin, an R package that enables the characterization of cell identity in solid tumors on the basis of a various and complementary analyses on SC gene expression data. RESULTSscMuffin provides a series of functions to calculate qualitative and quantitative scores, such as: expression of marker sets for normal and tumor conditions, pathway activity, cell state trajectories, CNVs, transcriptional complexity and proliferation state. Thus, scMuffin facilitates the combination of various evidences that can be used to distinguish normal and tumoral cells, define cell identities, cluster cells in different ways, link genomic aberrations to phenotypes and identify subtle differences between cell subtypes or cell states. We analysed public SC expression datasets of human high-grade gliomas as a proof-of-concept to show the value of scMuffin and illustrate its user interface. Nevertheless, these analyses lead to interesting findings, which suggest that some chromosomal amplifications might underlie the invasive tumor phenotype and the presence of cells that possess tumor initiating cells characteristics. CONCLUSIONSThe analyses offered by scMuffin and the results achieved in the case study show that our tool helps addressing the main challenges in the bioinformatics analysis of SC expression data from solid tumors.

bioinformatics↗

bioRxiv2022

Uncovering the spatial landscape of molecular interactions within the tumor microenvironment through latent spaces

Recent advances in spatial transcriptomics (ST) enable gene expression measurements from a tissue sample while retaining its spatial context. This technology enables unprecedented in situ resolution of the regulatory pathways that underlie the heterogeneity in the tumor and its microenvironment (TME). The direct characterization of cellular co-localization with spatial technologies facilities quantification of the molecular changes resulting from direct cell-cell interaction, as occurs in tumor-immune interactions. We present SpaceMarkers, a novel bioinformatics algorithm to infer molecular changes from cell-cell interaction from latent space analysis of ST data. We apply this approach to infer molecular changes from tumor-immune interactions in Visium spatial transcriptomics data of metastasis, invasive and precursor lesions, and immunotherapy treatment. Further transfer learning in matched scRNA-seq data enabled further quantification of the specific cell types in which SpaceMarkers are enriched. Altogether, SpaceMarkers can identify the location and context-specific molecular interactions within the TME from ST data.

bioinformatics↗

bioRxiv2022

massDatabase: utilities for the operation of the public compound and pathway database

SummaryOne of the major challenges in LC-MS data (metabolome, lipidome, and exposome) is converting many metabolic feature entries to biological function information, such as metabolite annotation and pathway enrichment, which are based on the compound and pathway databases. Multiple online databases have been developed, containing lots of information about compounds and pathways. However, there is still no tool developed for operating all these databases for biological analysis. Therefore, we developed massDatabase, an R package that operates the online public databases and combines with other tools for streamlined compound annotation and pathway enrichment analysis. massDatabase is a flexible, simple, and powerful tool that can be installed on all platforms, allowing the users to leverage all the online public databases for biological function mining. A detailed tutorial and a case study are provided in the Supplementary Materials. Availability and implementationhttps://massdatabase.tidymass.org/. Contactshenxt@stanford.edu and mpsnyder@stanford.edu Supplementary informationSupplementary data are available at Bioinformatics online.

bioinformatics↗

bioRxiv2022

Galaxy Training: A Powerful Framework for Teaching!

There is an ongoing explosion of scientific datasets being generated, brought on by recent technological advances in many areas of the natural sciences. As a result, the life sciences have become increasingly computational in nature, and bioinformatics has taken on a central role in research studies. However, basic computational skills, data analysis and stewardship are still rarely taught in life science educational programs [1], resulting in a skills gap in many of the researchers tasked with analysing these big datasets. In order to address this skills gap and empower researchers to perform their own data analyses, the Galaxy Training Network (GTN) has previously developed the Galaxy Training Platform (https://training.galaxyproject.org); an open access, community-driven framework for the collection of FAIR training materials for data analysis utilizing the user-friendly Galaxy framework as its primary data analysis platform [2]. Since its inception, this training platform has thrived, with the number of tutorials and contributors growing rapidly, and the range of topics extending beyond life sciences to include topics such as climatology, cheminformatics and machine learning. While initially aimed at supporting researchers directly, the GTN framework has proven to be an invaluable resource for educators as well. We have focused our efforts in recent years on adding increased support for this growing community of instructors. New features have been added to facilitate the use of the materials in a classroom setting, simplifying the contribution flow for new materials, and have added a set of train-the-trainer lessons. Here, we present the latest developments in the GTN project, aimed at facilitating the use of the Galaxy Training materials by educators, and its usage in different learning environments.

bioinformatics↗

bioRxiv2022

Assessing combinatorial diversity of aureochrome bZIPs through genome-wide screening

Aureochromes are unique blue light-responsive LOV (Light Oxygen Voltage) photoreceptors cum basic leucine zipper (bZIP) transcription factors (TFs), present exclusively in photosynthetic marine stramenopiles. Considering the availability of the complete genome sequence, this study focuses particularly on aureochromes from Ectocaupus siliculosus. Aureochromes mediate light-regulated developmental responses in this brown photosynthetic algae. Both the LOV sensor and the bZIP effector shows sequence-structure conservation. The LOV+bZIP modules of aureochrome homologs/ paralogs are not only structurally similar but also show an identical oligomeric state -- preferably dimeric. Aureochromes execute diverse cellular responses in different photosynthetic stramenopiles-- though their activities can vary even within a given algal species. Besides a heterogeneous linker connecting the sensor-effector and a flexible N-terminal region, the sequence composition of both the domains is vital. Therefore, it is important to understand whether aureochromes select dimerization partners from the same family or interact with other bZIPs as well. To regulate multifarious bio-logical activities, it is possible that aureochromes activate the global TF interaction network. Following homo/heterodimer modeling, we address the compatibility of dimerization partners by screening through heptad repeats. We evaluate the dimer interface area in terms of gain in solvation energy as well as the number of hydrogen bonds/salt bridge interactions. We further explore the relative stability of these structures from a graph-theoretic perspective through well-studied measures such as the energy of the graph and average participation coefficient. Furthermore, we also conduct an information-theoretic analysis using network information centrality and Kullback-Leibler divergence. We find that all our investigations into the relative stability of these dimers using diverse methods from bioinformatics, network science, and, information theory are in harmonious agreement. Our approach and findings should facilitate the design of experiments.

bioinformatics↗

bioRxiv2022

rGREAT: an R/Bioconductor package for functional enrichment on genomic regions

SummaryGREAT is a widely used tool for functional enrichment on genomic regions. However, as an online tool, it has limitations of outdated annotation data, small numbers of supported organisms and gene set collections, and not being extensible for users. Here we developed a new R/Bioconductor package named rGREAT which implements the GREAT algorithm locally. rGREAT by default supports more than 500 organisms and a large number of gene set collections, as well as self-provided gene sets and organisms from users. Additionally, it implements a general method for dealing with background regions. Availability and implementationThe package rGREAT is freely available from the Bioconductor project: https://bioconductor.org/packages/rGREAT/. The development version is available at https://github.com/jokergoo/rGREAT. Gene Ontology gene sets for 556 organisms are freely available at https://jokergoo.github.io/rGREAT_genesets/. Contactz.gu@dkfz.de or d.huebschmann@dkfz.de Supplementary informationSupplementary data are available at Bioinformatics online.

bioinformatics↗

bioRxiv2022

COVID-MVP: an interactive visualization for tracking SARS-CoV-2 mutations, variants, and prevalence, enabled by curated functional annotations and portable genomics workflow

The SARS-CoV-2 pandemic has reemphasized the importance of genomic epidemiology to track the evolution of the virus, dynamics of epidemics, geographic origins, and the emerging variants. It is vital in understanding the epidemiological spread of the virus on global, national, and local scales. Several analytical (bioinformatics) resources have been developed for molecular surveillance. However, a resource that combines genetic mutations and functional annotations on the impact of these mutations has been lacking in SARS-CoV-2 genomics surveillance. COVID-MVP provides an interactive visualization application that summarizes the mutations and their prevalence in SARS-CoV-2 viral lineages and provides functional annotations from the literature curated in an ongoing effort, Pokay. COVID-MVP is a tool that can be used for routine surveillance including spatio-temporal analyses. We have powered the visualization through a scalable and reproducible genomic analysis workflow nf-ncov-voc wrapped in Nextflow. COVID-MVP allows users to interactively explore data and download summarized surveillance reports. COVID-MVP, Pokay, and nf-ncov-voc are open-source tools available under the Massachusetts Institute of Technology (MIT) and GPL-3.0 licenses. COVID-MVP source code is available at https://github.com/cidgoh/COVID-MVP and an instance is hosted at https://covidmvp.cidgoh.ca.

bioinformatics↗

bioRxiv2022

The phers R package: using phenotype risk scores based on electronic health records to study Mendelian disease and rare genetic variants

Electronic health record (EHR) data linked to DNA biobanks are a valuable resource for understanding the phenotypic effects of human genetic variation. We previously developed the phenotype risk score (PheRS) as an approach to quantify the extent to which a patients clinical features resemble a given Mendelian disease. Using PheRS, we have uncovered novel associations between Mendelian diseaselike phenotypes and rare genetic variants, and identified patients who may have undiagnosed Mendelian disease. Although the PheRS approach is conceptually simple, it involves multiple mapping steps and was previously only available as custom scripts, limiting the approachs usability. Thus, we developed the phers R package, a complete and user-friendly set of functions and maps for performing a PheRS-based analysis on linked clinical and genetic data. The package includes up-to-date maps between EHR-based phenotypes (i.e., ICD codes and phecodes), human phenotype ontology (HPO) terms, and Mendelian diseases. Starting with occurrences of ICD codes, the package enables the user to calculate phenotype risk scores, validate the scores using case-control analyses, and perform genetic association analyses. By increasing PheRSs transparency and usability, the phers R package will help improve our understanding of the relationships between rare genetic variants and clinically meaningful human phenotypes. AvailabilityThe phers R package is free and open-source, and available on CRAN and at https://phers.hugheylab.org. Contactjakejhughey@gmail.com Supplementary informationSupplementary data are available at Bioinformatics online.

bioinformatics↗

bioRxiv2022

HybAVPnet: a novel hybrid network architecture for antiviral peptides identification

MotivationThe virus poses a great threat to human production and life, thus the research and development of antiviral drugs is urgently needed. Antiviral peptides play an important role in drug design and development. Compared with the time-consuming and laborious wet chemical experiment methods, accurate and rapid identification of antiviral peptides using computational methods is critical. However, it is still challenging to extract effective feature representations from the sequences for the accurate identification of antiviral peptides. ResultsThis study introduces a novel two-step approach, named HybAVPnet, with a hybrid network architecture to identify antiviral peptides based on neural networks and traditional machine learning methods. Firstly, eighteen kinds of features are extracted to predict labels and probabilities by the neural network classifier and LightGBM classifier, respectively. Secondly, the support vector machine classifier is carried out using the predicted probability of the first step to make the final prediction. The experimental result shows HybAVPnet can achieve better and more robust performance compared with the state-of-the-art methods, especially on independent datasets, which makes it useful for the research and development of antiviral drugs. Meanwhile, it can also be extended to other peptide recognition problems because of its generalization ability. Availability and implementationThe predicted model could be downloaded from: https://github.com/greyspring/HybAVPnet Contactgespring@hdu.edu.cn; yp.cai@siat.ac.cn Supplementary informationSupplementary data are available at Bioinformatics online.

bioinformatics↗

bioRxiv2022

GAMBIT (Genomic Approximation Method for Bacterial Identification and Tracking): A methodology to rapidly leverage whole genome sequencing of bacterial isolates for clinical identification

Whole genome sequencing of clinical bacterial isolates has the potential to transform the fields of medicine and public health, with particular impact on molecular epidemiology, infection control and assessing the spread of antibiotic resistance. To realize this potential, bioinformatic software needs to be developed that meets the quality standards of a diagnostic test to allow the reporting of identification results. Our research group has developed a methodology (GAMBIT: Genomic Approximation Method for Bacterial Identification and Tracking) using k-mer based strategies for identification of bacteria based on whole genome sequence reads. GAMBIT incorporates this algorithm with a highly curated searchable database of genomes, which is a subset of the NCBI RefSeq assembly database. In this manuscript, we describe the validation of the scoring methodology, robustness to chosen parameters, establishment of confidence thresholds and the curation of the reference database. We discuss a validation set with GAMBIT deployed as a laboratory-developed test at the Alameda County Public Health Laboratory in Oakland, California. Three advancements were required to build upon existing k-mer based strategies to allow GAMBIT to possess the quality control parameters desired for its use as a diagnostic laboratory-developed test. Firstly, we innovated the data structure used to store the database of known bacterial genomes. This allowed us to store 48,224 bacterial genomes in a k-mer based database--a majority of the bacterial genomes in the NCBI RefSeq database at the time of development. Secondly, curation of the NCBI RefSeq database was required to remove ambiguous or potentially incorrectly labeled bacterial genomes to greatly increase confidence in positive matches. Lastly, we used this curated version of the NCBI RefSeq database and our scoring method to generate confidence thresholds for identification. Thus, the end-user does not rely simply on the closest match, but is informed whether that closest match exceeds a threshold for highly confident identification. This method greatly reduces or eliminates false identifications which are often detrimental in a clinical setting.

bioinformatics↗

bioRxiv2022

Hybrid Autoencoder with Orthogonal Latent Space for Robust Population Structure Inference

BackgroundAnalysis of population structure and genomic ancestry remains an important topic in human genetics and bioinformatics. Commonly used methods require high-quality genotype data to ensure accurate inference. However, in practice, laboratory artifacts and outliers are often present in the data. Moreover, existing methods are typically affected by the presence of related individuals in the dataset. ResultsIn this work, we propose a novel hybrid method, called SAE-IBS, which combines the strengths of traditional matrix decomposition-based (e.g., principal component analysis) and more recent neural network-based (e.g., autoencoders) solutions. I.e., it yields an orthogonal latent space enhancing dimensionality selection while learning non-linear transformations. The proposed approach achieves higher accuracy than existing methods for projecting poor quality target samples (genotyping errors and missing data) onto a reference ancestry space and generates a robust ancestry space in the presence of relatedness. ConclusionWe introduce a new approach and an accompanying open-source program for robust ancestry inference in the presence of missing data, genotyping errors, and relatedness. The obtained ancestry space allows for non-linear projections and exhibits orthogonality with clearly separable population groups.

bioinformatics↗