BACKGROUND: The advent of low cost, high throughput DNA sequencing has led to the availability of thousands of complete genome sequences for a wide variety of bacterial species. Examining and interpreting genetic variation on this scale represents a significant challenge to existing methods of data analysis and visualisation. RESULTS: Starting with the output of standard pangenome analysis tools, we describe the generation and analysis of interactive, 3D network graphs to explore the structure of bacterial populations, the distribution of genes across a population, and the syntenic order in which those genes occur, in the new open-source network analysis platform, Graphia. Both the analysis and the visualisation are scalable to datasets of thousands of genome sequences. CONCLUSIONS: We anticipate that the approaches presented here will be of great utility to the microbial research community, allowing faster, more intuitive, and flexible interaction with pangenome datasets, thereby enhancing interpretation of these complex data.

Erratum in BMC Bioinformatics. 2022 Nov 18;23(1):494.

BACKGROUND: Transcriptional regulation is a fundamental mechanism underlying biological functions. In recent years, a broad array of RNA-Seq tools have been used to measure transcription levels in biological experiments, in whole organisms, tissues, and at the single cell level. Collectively, this is a vast comparative dataset on transcriptional processes across organisms. Yet, due to technical differences between the studies (sequencing, experimental design, and analysis) extracting usable comparative information and conducting meta-analyses remains challenging. RESULTS: We introduce Comparative RNA-Seq Metadata Analysis Pipeline (CoRMAP), a meta-analysis tool to retrieve comparative gene expression data from any RNA-Seq dataset using de novo assembly, standardized gene expression tools and the implementation of OrthoMCL, a gene orthology search algorithm. It employs the use of orthogroup assignments to ensure the accurate comparison of gene expression levels between experiments and species. Here we demonstrate the use of CoRMAP on two mouse brain transcriptomes with similar scope, that were collected several years from each other using different sequencing technologies and analysis methods. We also compare the performance of CoRMAP with a functional mapping tool, previously published. CONCLUSION: CoRMAP provides a framework for the meta-analysis of RNA-Seq data from divergent taxonomic groups. This method facilitates the retrieval and comparison of gene expression levels from published data sets using standardized assembly and analysis. CoRMAP does not rely on reference genomes and consequently facilitates direct comparison between diverse studies on a range of organisms.

BACKGROUND: Identifying splice site regions is an important step in the genomic DNA sequencing pipelines of biomedical and pharmaceutical research. Within this research purview, efficient and accurate splice site detection is highly desirable, and a variety of computational models have been developed toward this end. Neural network architectures have recently been shown to outperform classical machine learning approaches for the task of splice site prediction. Despite these advances, there is still considerable potential for improvement, especially regarding model prediction accuracy, and error rate. RESULTS: Given these deficits, we propose EnsembleSplice, an ensemble learning architecture made up of four (4) distinct convolutional neural networks (CNN) model architecture combination that outperform existing splice site detection methods in the experimental evaluation metrics considered including the accuracies and error rates. We trained and tested a variety of ensembles made up of CNNs and DNNs using the five-fold cross-validation method to identify the model that performed the best across the evaluation and diversity metrics. As a result, we developed our diverse and highly effective splice site (SS) detection model, which we evaluated using two (2) genomic Homo sapiens datasets and the Arabidopsis thaliana dataset. The results showed that for of the Homo sapiens EnsembleSplice achieved accuracies of 94.16% for one of the acceptor splice sites and 95.97% for donor splice sites, with an error rate for the same Homo sapiens dataset, 4.03% for the donor splice sites and 5.84% for the acceptor splice sites datasets. CONCLUSIONS: Our five-fold cross validation ensured the prediction accuracy of our models are consistent. For reproducibility, all the datasets used, models generated, and results in our work are publicly available in our GitHub repository here: https://github.com/OluwadareLab/EnsembleSplice.

BACKGROUND: In the last few years, multi-omics data, that is, datasets containing different types of high-dimensional molecular variables for the same samples, have become increasingly available. To date, several comparison studies focused on feature selection methods for omics data, but to our knowledge, none compared these methods for the special case of multi-omics data. Given that these data have specific structures that differentiate them from single-omics data, it is unclear whether different feature selection strategies may be optimal for such data. In this paper, using 15 cancer multi-omics datasets we compared four filter methods, two embedded methods, and two wrapper methods with respect to their performance in the prediction of a binary outcome in several situations that may affect the prediction results. As classifiers, we used support vector machines and random forests. The methods were compared using repeated fivefold cross-validation. The accuracy, the AUC, and the Brier score served as performance metrics. RESULTS: The results suggested that, first, the chosen number of selected features affects the predictive performance for many feature selection methods but not all. Second, whether the features were selected by data type or from all data types concurrently did not considerably affect the predictive performance, but for some methods, concurrent selection took more time. Third, regardless of which performance measure was considered, the feature selection methods mRMR, the permutation importance of random forests, and the Lasso tended to outperform the other considered methods. Here, mRMR and the permutation importance of random forests already delivered strong predictive performance when considering only a few selected features. Finally, the wrapper methods were computationally much more expensive than the filter and embedded methods. CONCLUSIONS: We recommend the permutation importance of random forests and the filter method mRMR for feature selection using multi-omics data, where, however, mRMR is considerably more computationally costly.

BACKGROUND: Hot spots play an important role in protein binding analysis. The residue interaction network is a key point in hot spot prediction, and several graph theory-based methods have been proposed to detect hot spots. Although the existing methods can yield some interesting residues by network analysis, low recall has limited their abilities in finding more potential hot spots. RESULT: In this study, we develop three graph theory-based methods to predict hot spots from only a single residue interaction network. We detect the important residues by finding subgraphs with high densities, i.e., high average degrees. Generally, a high degree implies a high binding possibility between protein chains, and thus a subgraph with high density usually relates to binding sites that have a high rate of hot spots. By evaluating the results on 67 complexes from the SKEMPI database, our methods clearly outperform existing graph theory-based methods on recall and F-score. In particular, our main method, Min-SDS, has an average recall of over 0.665 and an f2-score of over 0.364, while the recall and f2-score of the existing methods are less than 0.400 and 0.224, respectively. CONCLUSION: The Min-SDS method performs best among all tested methods on the hot spot prediction problem, and all three of our methods provide useful approaches for analyzing bionetworks. In addition, the densest subgraph-based methods predict hot spots with only one residue interaction network, which is constructed from spatial atomic coordinate data to mitigate the shortage of data from wet-lab experiments.

BACKGROUND: In modern sequencing experiments, quickly and accurately identifying the sources of the reads is a crucial need. In metagenomics, where each read comes from one of potentially many members of a community, it can be important to identify the exact species the read is from. In other settings, it is important to distinguish which reads are from the targeted sample and which are from potential contaminants. In both cases, identification of the correct source of a read enables further investigation of relevant reads, while minimizing wasted work. This task is particularly challenging for long reads, which can have a substantial error rate that obscures the origins of each read. RESULTS: Existing tools for the read classification problem are often alignment or index-based, but such methods can have large time and/or space overheads. In this work, we investigate the effectiveness of several sampling and sketching-based approaches for read classification. In these approaches, a chosen sampling or sketching algorithm is used to generate a reduced representation (a "screen") of potential source genomes for a query readset before reads are streamed in and compared against this screen. Using a query read's similarity to the elements of the screen, the methods predict the source of the read. Such an approach requires limited pre-processing, stores and works with only a subset of the input data, and is able to perform classification with a high degree of accuracy. CONCLUSIONS: The sampling and sketching approaches investigated include uniform sampling, methods based on MinHash and its weighted and order variants, a minimizer-based technique, and a novel clustering-based sketching approach. We demonstrate the effectiveness of these techniques both in identifying the source microbial genomes for reads from a metagenomic long read sequencing experiment, and in distinguishing between long reads from organisms of interest and potential contaminant reads. We then compare these approaches to existing alignment, index and sketching-based tools for read classification, and demonstrate how such a method is a viable alternative for determining the source of query reads. Finally, we present a reference implementation of these approaches at https://github.com/arun96/sketching .

BACKGROUND: Nanoscale connectomics, which aims to map the fine connections between neurons with synaptic-level detail, has attracted increasing attention in recent years. Currently, the automated reconstruction algorithms in electron microscope volumes are in great demand. Most existing reconstruction methodologies for cellular and subcellular structures are independent, and exploring the inter-relationships between structures will contribute to image analysis. The primary goal of this research is to construct a joint optimization framework to improve the accuracy and efficiency of neural structure reconstruction algorithms. RESULTS: In this investigation, we introduce the concept of connectivity consensus between cellular and subcellular structures based on biological domain knowledge for neural structure agglomeration problems. We propose a joint graph partitioning model for solving ultrastructural and neuronal connections to overcome the limitations of connectivity cues at different levels. The advantage of the optimization model is the simultaneous reconstruction of multiple structures in one optimization step. The experimental results on several public datasets demonstrate that the joint optimization model outperforms existing hierarchical agglomeration algorithms. CONCLUSIONS: We present a joint optimization model by connectivity consensus to solve the neural structure agglomeration problem and demonstrate its superiority to existing methods. The intention of introducing connectivity consensus between different structures is to build a suitable optimization model that makes the reconstruction goals more consistent with biological plausible and domain knowledge. This idea can inspire other researchers to optimize existing reconstruction algorithms and other areas of biological data analysis.

BACKGROUND: Lysine succinylation is a newly discovered protein post-translational modifications. Predicting succinylation sites helps investigate the metabolic disease treatments. However, the biological experimental approaches are costly and inefficient, it is necessary to develop efficient computational approaches. RESULTS: In this paper, we proposed a novel predictor based on ensemble dense blocks and an attention module, called as pSuc-EDBAM, which adopted one hot encoding to derive the feature maps of protein sequences, and generated the low-level feature maps through 1-D CNN. Afterward, the ensemble dense blocks were used to capture feature information at different levels in the process of feature learning. We also introduced an attention module to evaluate the importance degrees of different features. The experimental results show that Acc reaches 74.25%, and MCC reaches 0.2927 on the testing dataset, which suggest that the pSuc-EDBAM outperforms the existing predictors. CONCLUSIONS: The experimental results of ten-fold cross-validation on the training dataset and independent test on the testing dataset showed that pSuc-EDBAM outperforms the existing succinylation site predictors and can predict potential succinylation sites effectively. The pSuc-EDBAM is feasible and obtains the credible predictive results, which may also provide valuable references for other related research. To make the convenience of the experimental scientists, a user-friendly web server has been established ( http://bioinfo.wugenqiang.top/pSuc-EDBAM/ ), by which the desired results can be easily obtained.

BACKGROUND: Biological age (BA) has been recognized as a more accurate indicator of aging than chronological age (CA). However, the current limitations include: insufficient attention to the incompleteness of medical data for constructing BA; Lack of machine learning-based BA (ML-BA) on the Chinese population; Neglect of the influence of model overfitting degree on the stability of the association results. METHODS AND RESULTS: Based on the medical examination data of the Chinese population (45-90 years), we first evaluated the most suitable missing interpolation method, then constructed 14 ML-BAs based on biomarkers, and finally explored the associations between ML-BAs and health statuses (healthy risk indicators and disease). We found that round-robin linear regression interpolation performed best, while AutoEncoder showed the highest interpolation stability. We further illustrated the potential overfitting problem in ML-BAs, which affected the stability of ML-Bas' associations with health statuses. We then proposed a composite ML-BA based on the Stacking method with a simple meta-model (STK-BA), which overcame the overfitting problem, and associated more strongly with CA (r = 0.66, P < 0.001), healthy risk indicators, disease counts, and six types of disease. CONCLUSION: We provided an improved aging measurement method for middle-aged and elderly groups in China, which can more stably capture aging characteristics other than CA, supporting the emerging application potential of machine learning in aging research.

BACKGROUND: Compositional systems, represented as parts of some whole, are ubiquitous. They encompass the abundances of proteins in a cell, the distribution of organisms in nature, and the stoichiometry of the most basic chemical reactions. Thus, a central goal is to understand how such processes emerge from the behaviors of their components and their pairwise interactions. Such a study, however, is challenging for two key reasons. Firstly, such systems are complex and depend, often stochastically, on their constituent parts. Secondly, the data lie on a simplex which influences their correlations. RESULTS: To resolve both of these issues, we provide a general and data-driven modeling tool for compositional systems called Compositional Maximum Entropy (CME). By integrating the prior geometric structure of compositions with sample-specific information, CME infers the underlying multivariate relationships between the constituent components. We provide two proofs of principle. First, we measure the relative abundances of different bacteria and infer how they interact. Second, we show that our method outperforms a common alternative for the extraction of gene-gene interactions in triple-negative breast cancer. CONCLUSIONS: CME provides novel and biologically-intuitive insights and is promising as a comprehensive quantitative framework for compositional data.

Erratum in BMC Bioinformatics. 2022 Dec 9;23(1):533.

BACKGROUND: The site information of substrates that can be cleaved by human immunodeficiency virus 1 proteases (HIV-1 PRs) is of great significance for designing effective inhibitors against HIV-1 viruses. A variety of machine learning-based algorithms have been developed to predict HIV-1 PR cleavage sites by extracting relevant features from substrate sequences. However, only relying on the sequence information is not sufficient to ensure a promising performance due to the uncertainty in the way of separating the datasets used for training and testing. Moreover, the existence of noisy data, i.e., false positive and false negative cleavage sites, could negatively influence the accuracy performance. RESULTS: In this work, an ensemble learning algorithm for predicting HIV-1 PR cleavage sites, namely EM-HIV, is proposed by training a set of weak learners, i.e., biased support vector machine classifiers, with the asymmetric bagging strategy. By doing so, the impact of data imbalance and noisy data can thus be alleviated. Besides, in order to make full use of substrate sequences, the features used by EM-HIV are collected from three different coding schemes, including amino acid identities, chemical properties and variable-length coevolutionary patterns, for the purpose of constructing more relevant feature vectors of octamers. Experiment results on three independent benchmark datasets demonstrate that EM-HIV outperforms state-of-the-art prediction algorithm in terms of several evaluation metrics. Hence, EM-HIV can be regarded as a useful tool to accurately predict HIV-1 PR cleavage sites.

BACKGROUND: In the CRISPR-Cas9 system, the efficiency of genetic modifications has been found to vary depending on the single guide RNA (sgRNA) used. A variety of sgRNA properties have been found to be predictive of CRISPR cleavage efficiency, including the position-specific sequence composition of sgRNAs, global sgRNA sequence properties, and thermodynamic features. While prevalent existing deep learning-based approaches provide competitive prediction accuracy, a more interpretable model is desirable to help understand how different features may contribute to CRISPR-Cas9 cleavage efficiency. RESULTS: We propose a gradient boosting approach, utilizing LightGBM to develop an integrated tool, BoostMEC (Boosting Model for Efficient CRISPR), for the prediction of wild-type CRISPR-Cas9 editing efficiency. We benchmark BoostMEC against 10 popular models on 13 external datasets and show its competitive performance. CONCLUSIONS: BoostMEC can provide state-of-the-art predictions of CRISPR-Cas9 cleavage efficiency for sgRNA design and selection. Relying on direct and derived sequence features of sgRNA sequences and based on conventional machine learning, BoostMEC maintains an advantage over other state-of-the-art CRISPR efficiency prediction models that are based on deep learning through its ability to produce more interpretable feature insights and predictions.

BACKGROUND: Anti-CRISPR proteins are potent modulators that inhibit the CRISPR-Cas immunity system and have huge potential in gene editing and gene therapy as a genome-editing tool. Extensive studies have shown that anti-CRISPR proteins are essential for modifying endogenous genes, promoting the RNA-guided binding and cleavage of DNA or RNA substrates. In recent years, identifying and characterizing anti-CRISPR proteins has become a hot and significant research topic in bioinformatics. However, as most anti-CRISPR proteins fall short in sharing similarities to those currently known, traditional screening methods are time-consuming and inefficient. Machine learning methods could fill this gap with powerful predictive capability and provide a new perspective for anti-CRISPR protein identification. RESULTS: Here, we present a novel machine learning ensemble predictor, called PreAcrs, to identify anti-CRISPR proteins from protein sequences directly. Three features and eight different machine learning algorithms were used to train PreAcrs. PreAcrs outperformed other existing methods and significantly improved the prediction accuracy for identifying anti-CRISPR proteins. CONCLUSIONS: In summary, the PreAcrs predictor achieved a competitive performance for predicting new anti-CRISPR proteins in terms of accuracy and robustness. We anticipate PreAcrs will be a valuable tool for researchers to speed up the research process. The source code is available at: https://github.com/Lyn-666/anti_CRISPR.git .

BACKGROUND: Generating polygenic risk scores for diseases and complex traits requires high quality GWAS summary statistic files. Often, these files can be difficult to acquire either as a result of unshared or incomplete data. To date, bioinformatics tools which focus on restoring missing columns containing identification and association data are limited, which has the potential to increase the number of usable GWAS summary statistics files. RESULTS: SumStatsRehab was able to restore rsID, effect/other alleles, chromosome, base pair position, effect allele frequencies, beta, standard error, and p-values to a better extent than any other currently available tool, with minimal loss. CONCLUSIONS: SumStatsRehab offers a unique tool utilizing both functional programming and pipeline-like architecture, allowing users to generate accurate data restorations for incomplete summary statistics files. This in turn, increases the number of usable GWAS summary statistics files, which may be invaluable for less researched health traits.

BACKGROUND: Sophisticated methods to properly pre-process and analyze the increasing collection of single-cell RNA sequencing (scRNA-seq) data are increasingly being developed. On the contrary, the best practices to integrate these data into metabolic networks, aiming at describing metabolic phenotypes within a heterogeneous cell population, have been poorly investigated. In this regard, a critical factor is the presence of false zero values in reactions essential for a fundamental metabolic function, such as biomass or energy production. Here, we investigate the role of denoising strategies in mitigating this problem. METHODS: We applied state-of-the-art denoising strategies - namely MAGIC, ENHANCE, and SAVER - on three public scRNA-seq datasets. We then associated a metabolic flux distribution with every single cell by embedding its noise-free transcriptomics profile in the constraints of the optimization of a core metabolic model. Finally, we used the obtained single-cell optimal metabolic fluxes as features for cluster analysis. We compared the results obtained with different techniques, and with or without the use of denoising. We also investigated the possibility of applying denoising directly on the Reaction Activity Scores, which are metabolic features extracted from the read counts, rather than on the read counts. RESULTS: We show that denoising of transcriptomics data improves the clustering of single cells. We also illustrate that denoising restores important metabolic properties, such as the correlation between cell cycle phase and biomass accumulation, and between the RAS scores of reactions belonging to the same metabolic pathway. We show that MAGIC performs better than ENHANCE and SAVER, and that, denoising applied directly on the RAS matrix could be an effective alternative in removing false zero values from essential metabolic reactions. CONCLUSIONS: Our results indicate that including denoising as a pre-processing operation represents a milestone to integrate scRNA-seq data into Flux Balance Analysis simulations and to perform single-cell cluster analysis with a focus on metabolic phenotypes.

BACKGROUND: For genetic analyses, multi-allelic markers have an advantage over bi-allelic markers like SNPs (single nucleotide polymorphisms) in that they carry more information about the genetic constitution of individuals. This is especially the case in polyploids, where individuals carry more than two alleles at each locus. Haploblocks are multi-allelic markers that can be derived by phasing sets of closely-linked SNP markers. Phased haploblocks, similarly to other multi-allelic markers, will therefore be advantageous in genetic tasks like linkage mapping, QTL mapping and genome-wide association studies. RESULTS: We present a new method to reconstruct haplotypes from SNP dosages derived from genotyping arrays, which is applicable to polyploids. This method is implemented in the software package PolyHaplotyper. In contrast to existing packages for polyploids it makes use of full-sib families among the samples to guide the haplotyping process. We show that in this situation it is much more accurate than other available software, using experimental hexaploid data and simulated tetraploid data. CONCLUSIONS: Our method and the software package PolyHaplotyper in which it is implemented extend the available tools for haplotyping in polyploids. They perform especially well in situations where one or more full-sib families are present.

BACKGROUND: Availability of next generation sequencing data, allows low-frequency and rare variants to be studied through strategies other than the commonly used genome-wide association studies (GWAS). Rare variants are important keys towards explaining the heritability for complex diseases that remains to be explained by common variants due to their low effect sizes. However, analysis strategies struggle to keep up with the huge amount of data at disposal therefore creating a bottleneck. This study describes CLIN_SKAT, an R package, that provides users with an easily implemented analysis pipeline with the goal of (i) extracting clinically relevant variants (both rare and common), followed by (ii) gene-based association analysis by grouping the selected variants. RESULTS: CLIN_SKAT offers four simple functions that can be used to obtain clinically relevant variants, map them to genes or gene sets, calculate weights from global healthy populations and conduct weighted case-control analysis. CLIN_SKAT introduces improvements by adding certain pre-analysis steps and customizable features to make the SKAT results clinically more meaningful. Moreover, it offers several plot functions that can be availed towards obtaining visualizations for interpretation of the analyses results. CLIN_SKAT is available on Windows/Linux/MacOS and is operative for R version 4.0.4 or later. It can be freely downloaded from https://github.com/ShihChingYu/CLIN_SKAT , installed through devtools::install_github("ShihChingYu/CLIN_SKAT", force=T) and executed by loading the package into R using library(CLIN_SKAT). All outputs (tabular and graphical) can be downloaded in simple, publishable formats. CONCLUSIONS: Statistical association analysis is often underpowered due to low sample sizes and high numbers of variants to be tested, limiting detection of causal ones. Therefore, retaining a subset of variants that are biologically meaningful seems to be a more effective strategy for identifying explainable associations while reducing the degrees of freedom. CLIN_SKAT offers users a one-stop R package that identifies disease risk variants with improved power via a series of tailor-made procedures that allows dimension reduction, by retaining functionally relevant variants, and incorporating ethnicity based priors. Furthermore, it also eliminates the requirement for high computational resources and bioinformatics expertise.

BACKGROUND: In fluorescence microscopy, co-localization refers to the spatial overlap between different fluorescent labels in cells. The degree of overlap between two or more channels in a microscope may reveal a physical interaction or topological functional interconnection between molecules. Recent advances in the imaging field require the development of specialized computational analysis software for the unbiased assessment of fluorescently labelled microscopy images. RESULTS: Here we present SpotitPy, a semi-automated image analysis tool for 2D object-based co-localization. SpotitPy allows the user to select fluorescent labels and perform a semi-automated and robust segmentation of the region of interest in distinct cell types. The workflow integrates advanced pre-processing manipulations for de-noising and in-depth semi-automated quantification of the co-localized fluorescent labels in two different channels. We validated SpotitPy by quantitatively assessing the presence of cytoplasmic ribonucleoprotein granules, e.g. processing (P) bodies, under conditions that challenge mRNA translation, thus highlighting SpotitPy benefits for semi-automatic, accurate analysis of large image datasets in eukaryotic cells. SpotitPy comes in a command line interface or a simple graphical user interphase and can be used as a standalone application. CONCLUSIONS: Overall, we present a novel and user-friendly tool that performs a semi-automated image analysis for 2D object-based co-localization. SpotitPy can provide reproducible and robust quantifications for large datasets within a limited timeframe. The software is open-source and can be found in the GitHub project repository: ( https://github.com/alexiaales/SpotitPy ).

BACKGROUND: High-content screening (HCS) experiments generate complex data from multiple object features for each cell within a treated population. Usually, these data are analyzed by using population-averaged values of the features of interest, increasing the amount of false positives and the need for intensive follow-up validation. Therefore, there is a strong need for novel approaches with reproducible hit prediction by identifying significantly altered cell populations. RESULTS: Here we describe SOPRA, a workflow for analyzing image-based HCS data based on regression analysis of non-averaged object features from cell populations, which can be run on hundreds of samples using different cell features. Following plate-wise normalization, the values are counted within predetermined binning intervals, generating unique frequency distribution profiles (histograms) for each population, which are then normalized to control populations (control-based normalization). These control-normalized frequency distribution profiles are analyzed using the Bioconductor R-package maSigPro, originally developed to analyze time profiles. However, statistically significant altered frequency distributions are also identified by maSigPro when integrating it into the SOPRA workflow. Finally, significantly changed profiles can be used to generate a heatmap from which altered cell populations with similar phenotypes can be identified, enabling the detection of siRNAs and compounds with the same 'on-target' profile and reducing the number of false positive hits. CONCLUSIONS: SOPRA is a novel analysis workflow for the detection of statistically significant normalized frequency distribution profiles of cellular features generated in high-throughput RNAi screens. For the validation of the SOPRA software workflow, a screen for cell cycle progression was used. We were able to identify such profiles for siRNA-mediated gene perturbations and chemical inhibitors of different cell cycle stages. The SOPRA software is freely available from Github.

Recently, Deep Learning based automatic generation of treatment recommendation has been attracting much attention. However, medical datasets are usually small, which may lead to over-fitting and inferior performances of deep learning models. In this paper, we propose multi-objective data enhancement method to indirectly scale up the medical data to avoid over-fitting and generate high quantity treatment recommendations. Specifically, we define a main and several auxiliary tasks on the same dataset and train a specific model for each of these tasks to learn different aspects of knowledge in limited data scale. Meanwhile, a Soft Parameter Sharing method is exploited to share learned knowledge among models. By sharing the knowledge learned by auxiliary tasks to the main task, the proposed method can take different semantic distributions into account during the training process of the main task. We collected an ultrasound dataset of thyroid nodules that contains Findings, Impressions and Treatment Recommendations labeled by professional doctors. We conducted various experiments on the dataset to validate the proposed method and justified its better performance than existing methods.

BACKGROUND: In the context of a binary classification problem, the optimal linear combination of continuous predictors can be estimated by maximizing the area under the receiver operating characteristic curve. For ordinal responses, the optimal predictor combination can similarly be obtained by maximization of the hypervolume under the manifold (HUM). Since the empirical HUM is discontinuous, non-differentiable, and possibly multi-modal, solving this maximization problem requires a global optimization technique. Estimation of the optimal coefficient vector using existing global optimization techniques is computationally expensive, becoming prohibitive as the number of predictors and the number of outcome categories increases. RESULTS: We propose an efficient derivative-free black-box optimization technique based on pattern search to solve this problem, which we refer to as Spherically Constrained Optimization Routine (SCOR). Through extensive simulation studies, we demonstrate that the proposed method achieves better performance than existing methods including the step-down algorithm. Finally, we illustrate the proposed method to predict the severity of swallowing difficulty after radiation therapy for oropharyngeal cancer based on radiation dose to various structures in the head and neck. CONCLUSIONS: Our proposed method addresses an important challenge in combining multiple biomarkers to predict an ordinal outcome. This problem is particularly relevant to medical research, where it may be of interest to diagnose a disease with various stages of progression or a toxicity with multiple grades of severity. We provide the implementation of our proposed SCOR method as an R package, available online at https://CRAN.R-project.org/package=SCOR .

BACKGROUND: Few studies have demonstrated that the relationship between m6A-related genes and the prognosis, tumor microenvironment and drug resistance of LC. METHODS: The main results were analyzed with bioinformatics methods. RESULTS: Hence, we found 10 m6A-related genes expressed less in tumor samples in comparison with normal ones. Using consensus clustering, all LC patients were grouped into 2 subgroups according to the overall expression of 10 differential expressed m6A-related genes. In two clusters, the OS and immune characteristics were different. We analyzed the predictive potential of 10 m6A-related genes in the prognosis of LC, and obtained a risk prognosis model on the strength of ZC3H13, CBLL1, ELAVL1 and YTHDF1 as the hub candidate genes through LASSO cox. The expression of 4 hub m6A-related genes was validated by IHC in the HPA database. The infiltration level of dendritic cell, CD4+ T cell and neutrophil that were affected by CNV level of m6A-related genes in LUAD and LUSC patients. Moreover, based on GSCALite database, we found that LUSC patients with hypermethylation tended to have a better overall survival. In terms of drug sensitivity, etoposide correlated negatively with ELAVL1, HNRNPC, RBM15B, YTHDF2 and CBLL1. ZC3H13 had positively association with afatinib, while HNRNPC was positively associated with dasatinib, erlotinib, lapatinib and TGX221. Crizotinib had a negative correlation with ELAVL1, CBLL1, HNRNPC and RBM15B. CONCLUSION: In conclusion, m6A-related genes are important participants in LC and the expression levels of ZC3H13, CBLL1, ELAVL1 and YTHDF1 are significant for prediction and treatment of LC. Researches of drug resistance based on m6A-related genes need to pay more attention for producing new therapeutic strategies of LC and CBLL1 may contribute to target treatment for further research.

PURPOSE: The aim of this study was to identify and screen long non-coding RNA (lncRNA) associated with immune genes in colon cancer, construct immune-related lncRNA pairs, establish a prognostic risk assessment model for colon adenocarcinoma (COAD), and explore prognostic factors and drug sensitivity. METHOD: Our method was based on data from The Cancer Genome Atlas (TCGA). To begin, we obtained all pertinent demographic and clinical information on 385 patients with COAD. All lncRNAs significantly related to immune genes and with differential expression were identified to construct immune lncRNA pairs. Subsequently, least absolute shrinkage and selection operator and Cox models were used to screen out prognostic-related immune lncRNAs for the establishment of a prognostic risk scoring formula. Finally, We analysed the functional differences between subgroups and screened the drugs, and establish an individual prediction nomogram model. RESULTS: Our final analysis confirmed eight lncRNA pairs to construct prognostic risk assessment model. Results showed that the high-risk and low-risk groups had significant differences (training (n = 249): p < 0.001, validation (n = 114): p = 0.022). The prognostic model was certified as an independent prognosis model. Compared with the common clinicopathological indicators, the prognostic model had better predictive efficiency (area under the curve (AUC) = 0.805). Finally, We have analysed highly differentiated cellular pathways such as mucosal immune response, identified 9 differential immune cells, 10 sensitive drugs, and establish an individual prediction nomogram model (C-index = 0.820). CONCLUSION: Our study verified that the eight lncRNA pairs mentioned can be used as biomarkers to predict the prognosis of COAD patients. Identified cells, drugs may have an positive effect on colon cancer prognosis.

BACKGROUND: Breast cancer is one of the most common cancers in women. It is necessary to classify breast cancer subtypes because different subtypes need specific treatment. Identifying biomarkers and classifying breast cancer subtypes is essential for developing appropriate treatment methods for patients. MiRNAs can be easily detected in tumor biopsy and play an inhibitory or promoting role in breast cancer, which are considered promising biomarkers for distinguishing subtypes. RESULTS: A new method combing ensemble regularized multinomial logistic regression and Cox regression was proposed for identifying miRNA biomarkers in breast cancer. After adopting stratified sampling and bootstrap sampling, the most suitable sample subset for miRNA feature screening was determined via ensemble 100 regularized multinomial logistic regression models. 124 miRNAs that participated in the classification of at least 3 subtypes and appeared at least 50 times in 100 integrations were screened as features. 22 miRNAs from the proposed feature set were further identified as the biomarkers for breast cancer by using Cox regression based on survival analysis. The accuracy of 5 methods on the proposed feature set was significantly higher than on the other two feature sets. The results of 7 biological analyses illustrated the rationality of the identified biomarkers. CONCLUSIONS: The screened features can better distinguish breast cancer subtypes. Notably, the genes and proteins related to the proposed 22 miRNAs were considered oncogenes or inhibitors of breast cancer. 9 of the 22 miRNAs have been proved to be markers of breast cancer. Therefore, our results can be considered in future related research.

BACKGROUND: Gene expression is regulated at different molecular levels, including chromatin accessibility, transcription, RNA maturation, and transport. These regulatory mechanisms have strong connections with cellular metabolism. In order to study the cellular system and its functioning, omics data at each molecular level can be generated and efficiently integrated. Here, we propose BRANENET, a novel multi-omics integration framework for multilayer heterogeneous networks. BRANENET is an expressive, scalable, and versatile method to learn node embeddings, leveraging random walk information within a matrix factorization framework. Our goal is to efficiently integrate multi-omics data to study different regulatory aspects of multilayered processes that occur in organisms. We evaluate our framework using multi-omics data of Saccharomyces cerevisiae, a well-studied yeast model organism. RESULTS: We test BRANENET on transcriptomics (RNA-seq) and targeted metabolomics (NMR) data for wild-type yeast strain during a heat-shock time course of 0, 20, and 120 min. Our framework learns features for differentially expressed bio-molecules showing heat stress response. We demonstrate the applicability of the learned features for targeted omics inference tasks: transcription factor (TF)-target prediction, integrated omics network (ION) inference, and module identification. The performance of BRANENET is compared to existing network integration methods. Our model outperforms baseline methods by achieving high prediction scores for a variety of downstream tasks.

BACKGROUND: Increasing biomedical studies have shown that the dysfunction of miRNAs is closely related with many human diseases. Identifying disease-associated miRNAs would contribute to the understanding of pathological mechanisms of diseases. Supervised learning-based computational methods have continuously been developed for miRNA-disease association predictions. Negative samples of experimentally-validated uncorrelated miRNA-disease pairs are required for these approaches, while they are not available due to lack of biomedical research interest. Existing methods mainly choose negative samples from the unlabelled ones randomly. Therefore, the selection of more reliable negative samples is of great importance for these methods to achieve satisfactory prediction results. RESULTS: In this study, we propose a computational method termed as KR-NSSM which integrates two semi-supervised algorithms to select more reliable negative samples for miRNA-disease association predictions. Our method uses a refined K-means algorithm for preliminary screening of likely negative and positive miRNA-disease samples. A Rocchio classification-based method is applied for further screening to receive more reliable negative and positive samples. We implement ablation tests in KR-NSSM and find that the combination of the two selection procedures would obtain more reliable negative samples for miRNA-disease association predictions. Comprehensive experiments based on fivefold cross-validations demonstrate improvements in prediction accuracy on six classic classifiers and five known miRNA-disease association prediction models when using negative samples chose by our method than by previous negative sample selection strategies. Moreover, 469 out of 1123 selected positive miRNA-disease associations by our method are confirmed by existing databases. CONCLUSIONS: Our experiments show that KR-NSSM can screen out more reliable negative samples from the unlabelled ones, which greatly improves the performance of supervised machine learning methods in miRNA-disease association predictions. We expect that KR-NSSM would be a useful tool in negative sample selection in biomedical research.

MOTIVATION: Studies have shown that classifying cancer subtypes can provide valuable information for a range of cancer research, from aetiology and tumour biology to prognosis and personalized treatment. Current methods usually adopt gene expression data to perform cancer subtype classification. However, cancer samples are scarce, and the high-dimensional features of their gene expression data are too sparse to allow most methods to achieve desirable classification results. RESULTS: In this paper, we propose a deep learning approach by combining a convolutional neural network (CNN) and bidirectional gated recurrent unit (BiGRU): our approach, DCGN, aims to achieve nonlinear dimensionality reduction and learn features to eliminate irrelevant factors in gene expression data. Specifically, DCGN first uses the synthetic minority oversampling technique algorithm to equalize data. The CNN can handle high-dimensional data without stress and extract important local features, and the BiGRU can analyse deep features and retain their important information; the DCGN captures key features by combining both neural networks to overcome the challenges of small sample sizes and sparse, high-dimensional features. In the experiments, we compared the DCGN to seven other cancer subtype classification methods using breast and bladder cancer gene expression datasets. The experimental results show that the DCGN performs better than the other seven methods and can provide more satisfactory classification results.

BACKGROUND: Predicting morphological changes to anatomical structures from 3D shapes such as blood vessels or appearance of the face is a growing interest to clinicians. Machine learning (ML) has had great success driving predictions in 2D, however, methods suitable for 3D shapes are unclear and the use cases unknown. OBJECTIVE AND METHODS: This systematic review aims to identify the clinical implementation of 3D shape prediction and ML workflows. Ovid-MEDLINE, Embase, Scopus and Web of Science were searched until 28th March 2022. RESULTS: 13,754 articles were identified, with 12 studies meeting final inclusion criteria. These studies involved prediction of the face, head, aorta, forearm, and breast, with most aiming to visualize shape changes after surgical interventions. ML algorithms identified were regressions (67%), artificial neural networks (25%), and principal component analysis (8%). Meta-analysis was not feasible due to the heterogeneity of the outcomes. CONCLUSION: 3D shape prediction is a nascent but growing area of research in medicine. This review revealed the feasibility of predicting 3D shapes using ML clinically, which could play an important role for clinician-patient visualization and communication. However, all studies were early phase and there were inconsistent language and reporting. Future work could develop guidelines for publication and promote open sharing of source code.

BACKGROUND: Cardiovascular disease (CVD) is a serious disease that endangers human health and is one of the main causes of death. Therefore, using the patient's electronic medical record (EMR) to predict CVD automatically has important application value in intelligent assisted diagnosis and treatment, and is a hot issue in intelligent medical research. However, existing methods based on natural language processing can only predict CVD according to the whole or part of the context information of EMR. RESULTS: Given the deficiencies of the existing research on CVD prediction based on EMRs, this paper proposes a risk factor attention-based model (RFAB) to predict CVD by utilizing CVD risk factors and general EMRs text, which adopts the attention mechanism of a deep neural network to fuse the character sequence and CVD risk factors contained in EMRs text. The experimental results show that the proposed method can significantly improve the prediction performance of CVD, and the F-score reaches 0.9586, which outperforms the existing related methods. CONCLUSIONS: RFAB focuses on the key information in EMR that leads to CVD, that is, 12 risk factors. In the stage of risk factor identification and extraction, risk factors are labeled with category information and time attribute information by BiLSTM-CRF model. In the stage of CVD prediction, the information contained in risk factors and their labels is fused with the information of character sequence in EMR to predict CVD. RFAB makes well use of the fine-grained information contained in EMR, and also provides a reliable idea for predicting CVD.

BACKGROUND: RNA deleterious point mutation prediction was previously addressed with programs such as RNAmute and MultiRNAmute. The purpose of these programs is to predict a global conformational rearrangement of the secondary structure of a functional RNA molecule, thereby disrupting its function. RNAmute was designed to deal with only single point mutations in a brute force manner, while in MultiRNAmute an efficient approach to deal with multiple point mutations was developed. The approach used in MultiRNAmute is based on the stabilization of the suboptimal RNA folding prediction solutions and/or destabilization of the optimal folding prediction solution of the wild type RNA molecule. The MultiRNAmute algorithm is significantly more efficient than the brute force approach in RNAmute, but in the case of long sequences and large m-point mutation sets the MultiRNAmute becomes exponential in examining all possible stabilizing and destabilizing mutations. RESULTS: An inherent limitation in the RNAmute and MultiRNAmute programs is their ability to predict only substitution mutations, as these programs were not designed to work with deletion or insertion mutations. To address this limitation we herein develop a very fast algorithm, based on suboptimal folding solutions, to predict a predefined number of multiple point deleterious mutations as specified by the user. Depending on the user's choice, each such set of mutations may contain combinations of deletions, insertions and substitution mutations. Additionally, we prove the hardness of predicting the most deleterious set of point mutations in structural RNAs. CONCLUSIONS: We developed a method that extends our previous MultiRNAmute method to predict insertion and deletion mutations in addition to substitutions. The additional advantage of the new method is its efficiency to find a predefined number of deleterious mutations. Our new method may be exploited by biologists and virologists prior to site-directed mutagenesis experiments, which involve indel mutations along with substitutions. For example, our method may help to investigate the change of function in an RNA virus via mutations that disrupt important motifs in its secondary structure.

BACKGROUND: Accurate precision approaches have far not been developed for modeling mortality risk in intensive care unit (ICU) patients. Conventional mortality risk prediction methods can hardly extract the information in longitudinal electronic medical records (EHRs) effectively, since they simply aggregate the heterogeneous variables in EHRs, ignoring the complex relationship and interactions between variables and the time dependence in longitudinal records. Recently deep learning approaches have been widely used in modeling longitudinal EHR data. However, most existing deep learning-based risk prediction approaches only use the information of a single disease, neglecting the interactions between multiple diseases and different conditions. RESULTS: In this paper, we address this unmet need by leveraging disease and treatment information in EHRs to develop a mortality risk prediction model based on deep learning (DeepMPM). DeepMPM utilizes a two-level attention mechanism, i.e. visit-level and variable-level attention, to derive the representation of patient risk status from patient's multiple longitudinal medical records. Benefiting from using EHR of patients with multiple diseases and different conditions, DeepMPM can achieve state-of-the-art performances in mortality risk prediction. CONCLUSIONS: Experiment results on MIMIC III database demonstrates that with the disease and treatment information DeepMPM can achieve a good performance in terms of Area Under ROC Curve (0.85). Moreover, DeepMPM can successfully model the complex interactions between diseases to achieve better representation learning of disease and treatment than other deep learning approaches, so as to improve the accuracy of mortality prediction. A case study also shows that DeepMPM offers the potential to provide users with insights into feature correlation in data as well as model behavior for each prediction.

BACKGROUND: Increasing evidence shows that circRNA plays an essential regulatory role in diseases through interactions with disease-related miRNAs. Identifying circRNA-disease associations is of great significance to precise diagnosis and treatment of diseases. However, the traditional biological experiment is usually time-consuming and expensive. Hence, it is necessary to develop a computational framework to infer unknown associations between circRNA and disease. RESULTS: In this work, we propose an efficient framework called MSPCD to infer unknown circRNA-disease associations. To obtain circRNA similarity and disease similarity accurately, MSPCD first integrates more biological information such as circRNA-miRNA associations, circRNA-gene ontology associations, then extracts circRNA and disease high-order features by the neural network. Finally, MSPCD employs DNN to predict unknown circRNA-disease associations. CONCLUSIONS: Experiment results show that MSPCD achieves a significantly more accurate performance compared with previous state-of-the-art methods on the circFunBase dataset. The case study also demonstrates that MSPCD is a promising tool that can effectively infer unknown circRNA-disease associations.

Erratum in BMC Bioinformatics. 2022 Dec 22;23(1):557.

BACKGROUND: CRISPR-Cas based diagnostic assays provide a portable solution which bridges the benefits of qRT-PCR and serological assays in terms of portability, specificity and ease of use. CRISPR-Cas assays are rapidly fieldable, specific and have been rigorously validated against a number of targets, including HIV and vector-borne pathogens. Recently, CRISPR-Cas12 and CRISPR-Cas13 diagnostic assays have been granted FDA approval for the detection of SARS-CoV-2. A critical step in utilizing this technology requires the design of highly-specific and efficient CRISPR RNAs (crRNAs) and isothermal primers. This process involves intensive manual curation and stringent parameters for design in order to minimize off-target detection while also preserving detection across divergent strains. As such, a single, streamlined bioinformatics platform for rapidly designing crRNAs for use with the CRISPR-Cas12 platform is needed. Here we offer PrimedSherlock, an automated, computer guided process for selecting highly-specific crRNAs and primers for targets of interest. RESULTS: Utilizing PrimedSherlock and publicly available databases, crRNAs were designed against a selection of Flavivirus genomes, including West Nile, Zika and all four serotypes of Dengue. Using outputs from PrimedSherlock in concert with both wildtype A.s Cas12a and Alt-R Cas12a Ultra nucleases, we demonstrated sensitive detection of nucleic acids of each respective arbovirus in in-vitro fluorescence assays. Moreover, primer and crRNA combinations facilitated the detection of their intended targets with minimal off-target background noise. CONCLUSIONS: PrimedSherlock is a novel crRNA design tool, specific for CRISPR-Cas12 diagnostic platforms. It allows for the rapid identification of highly conserved crRNA targets from user-provided primer pairs or PrimedRPA output files. Initial testing of crRNAs against arboviruses of medical importance demonstrated a robust ability to distinguish multiple strains by exploiting polymorphisms within otherwise highly conserved genomic regions. As a freely-accessible software package, PrimedSherlock could significantly increase the efficiency of CRISPR-Cas12 diagnostics. Conceptually, the portability of detection kits could also be enhanced when coupled with isothermal amplification technologies.

BACKGROUND: Selecting and prioritizing candidate disease genes is necessary before conducting laboratory studies as identifying disease genes from a large number of candidate genes using laboratory methods, is a very costly and time-consuming task. There are many machine learning-based gene prioritization methods. These methods differ in various aspects including the feature vectors of genes, the used datasets with different structures, and the learning model. Creating a suitable feature vector for genes and an appropriate learning model on a variety of data with different and non-Euclidean structures, including graphs, as well as the lack of negative data are very important challenges of these methods. The use of graph neural networks has recently emerged in machine learning and other related fields, and they have demonstrated superior performance for a broad range of problems. METHODS: In this study, a new semi-supervised learning method based on graph convolutional networks is presented using the novel constructing feature vector for each gene. In the proposed method, first, we construct three feature vectors for each gene using terms from the Gene Ontology (GO) database. Then, we train a graph convolution network on these vectors using protein-protein interaction (PPI) network data to identify disease candidate genes. Our model discovers hidden layer representations encoding in both local graph structure as well as features of nodes. This method is characterized by the simultaneous consideration of topological information of the biological network (e.g., PPI) and other sources of evidence. Finally, a validation has been done to demonstrate the efficiency of our method. RESULTS: Several experiments are performed on 16 diseases to evaluate the proposed method's performance. The experiments demonstrate that our proposed method achieves the best results, in terms of precision, the area under the ROC curve (AUCs), and F1-score values, when compared with eight state-of-the-art network and machine learning-based disease gene prioritization methods. CONCLUSION: This study shows that the proposed semi-supervised learning method appropriately classifies and ranks candidate disease genes using a graph convolutional network and an innovative method to create three feature vectors for genes based on the molecular function, cellular component, and biological process terms from GO data.

BACKGROUND: Despite continuing technological advances, the cost for large-scale genotyping of a high number of samples can be prohibitive. The purpose of this study is to design a cost-saving strategy for SNP genotyping. We suggest making use of pooling, a group testing technique, to drop the amount of SNP arrays needed. We believe that this will be of the greatest importance for non-model organisms with more limited resources in terms of cost-efficient large-scale chips and high-quality reference genomes, such as application in wildlife monitoring, plant and animal breeding, but it is in essence species-agnostic. The proposed approach consists in grouping and mixing individual DNA samples into pools before testing these pools on bead-chips, such that the number of pools is less than the number of individual samples. We present a statistical estimation algorithm, based on the pooling outcomes, for inferring marker-wise the most likely genotype of every sample in each pool. Finally, we input these estimated genotypes into existing imputation algorithms. We compare the imputation performance from pooled data with the Beagle algorithm, and a local likelihood-aware phasing algorithm closely modeled on MaCH that we implemented. RESULTS: We conduct simulations based on human data from the 1000 Genomes Project, to aid comparison with other imputation studies. Based on the simulated data, we find that pooling impacts the genotype frequencies of the directly identifiable markers, without imputation. We also demonstrate how a combinatorial estimation of the genotype probabilities from the pooling design can improve the prediction performance of imputation models. Our algorithm achieves 93% concordance in predicting unassayed markers from pooled data, thus it outperforms the Beagle imputation model which reaches 80% concordance. We observe that the pooling design gives higher concordance for the rare variants than traditional low-density to high-density imputation commonly used for cost-effective genotyping of large cohorts. CONCLUSIONS: We present promising results for combining a pooling scheme for SNP genotyping with computational genotype imputation on human data. These results could find potential applications in any context where the genotyping costs form a limiting factor on the study size, such as in marker-assisted selection in plant breeding.

BACKGROUND: Observational studies and Mendelian randomization experiments have been used to identify many causal factors for complex traits in humans. Given a set of causal factors, it is important to understand the extent to which these causal factors explain some, all, or none of the genetic heritability, as measured by single-nucleotide polymorphisms (SNPs) that are associated with the trait. Using the mediation model framework with SNPs as the exposure, a trait of interest as the outcome, and the known causal factors as the mediators, we hypothesize that any unexplained association between the SNPs and the outcome trait is mediated by an additional unobserved, hidden causal factor. RESULTS: We propose a method to infer the effect size of this hidden mediating causal factor on the outcome trait by utilizing the estimated associations between a continuous outcome trait, the known causal factors, and the SNPs. The proposed method consists of three steps and, in the end, implements Markov chain Monte Carlo to obtain a posterior distribution for the effect size of the hidden mediator. We evaluate our proposed method via extensive simulations and show that when model assumptions hold, our method estimates the effect size of the hidden mediator well and controls type I error rate if the hidden mediator does not exist. In addition, we apply the method to the UK Biobank data and estimate parameters for a potential hidden mediator for waist-hip ratio beyond body mass index (BMI), and find that the hidden mediator has a large effect size relatively to the effect size of the known mediator BMI. CONCLUSIONS: We develop a framework to infer the effect of potential, hidden mediators influencing complex traits. This framework can be used to place boundaries on unexplained risk factors contributing to complex traits.

BACKGROUND: With the advent of metagenomics, the importance of microorganisms and how their interactions are relevant to ecosystem resilience, sustainability, and human health has become evident. Cataloging and preserving biodiversity is paramount not only for the Earth's natural systems but also for discovering solutions to challenges that we face as a growing civilization. Metagenomics pertains to the in silico study of all microorganisms within an ecological community in situ, however, many software suites recover only prokaryotes and have limited to no support for viruses and eukaryotes. RESULTS: In this study, we introduce the Viral Eukaryotic Bacterial Archaeal (VEBA) open-source software suite developed to recover genomes from all domains. To our knowledge, VEBA is the first end-to-end metagenomics suite that can directly recover, quality assess, and classify prokaryotic, eukaryotic, and viral genomes from metagenomes. VEBA implements a novel iterative binning procedure and hybrid sample-specific/multi-sample framework that yields more genomes than any existing methodology alone. VEBA includes a consensus microeukaryotic database containing proteins from existing databases to optimize microeukaryotic gene modeling and taxonomic classification. VEBA also provides a unique clustering-based dereplication strategy allowing for sample-specific genomes and genes to be directly compared across non-overlapping biological samples. Finally, VEBA is the only pipeline that automates the detection of candidate phyla radiation bacteria and implements the appropriate genome quality assessments. VEBA's capabilities are demonstrated by reanalyzing 3 existing public datasets which recovered a total of 948 MAGs (458 prokaryotic, 8 eukaryotic, and 482 viral) including several uncharacterized organisms and organisms with no public genome representatives. CONCLUSIONS: The VEBA software suite allows for the in silico recovery of microorganisms from all domains of life by integrating cutting edge algorithms in novel ways. VEBA fully integrates both end-to-end and task-specific metagenomic analysis in a modular architecture that minimizes dependencies and maximizes productivity. The contributions of VEBA to the metagenomics community includes seamless end-to-end metagenomics analysis but also provides users with the flexibility to perform specific analytical tasks. VEBA allows for the automation of several metagenomics steps and shows that new information can be recovered from existing datasets.

BACKGROUND: Colon adenocarcinoma (COAD) is one of the most common gastrointestinal tumors worldwide, and immunotherapy is one of the most promising treatments for it. Identifying immune genes involved in the development and maintenance of cancer is key to the use of tumor immunotherapy. This study aimed to determine the prognostic value of immune genes in patients with COAD and to establish an immune-related gene signature. Differentially expressed genes, immune-related genes (DEIGs), and transcription factors (DETFs) were screened using the following databases: Cistrome, The Cancer Genome Atlas (TCGA), the Immunology Database and Analysis Portal, and InnateDB. We constructed a network showing the regulation of DEIGs by DETFs. Using weighted gene co-expression network analysis, we prepared 5 co-expressed gene modules; 6 hub genes (CD1A, CD1B, FGF9, GRP, SERPINE1, and F2RL2) obtained using univariate and multivariate regression analysis were used to construct a risk model. Patients from TCGA database were divided into high- and low-risk groups based on whether their risk score was greater or less than the mean; the public dataset GSE40967, which contains gene expression profiles of 566 colon cancer patients, was used for validation. RESULTS: Survival analysis, somatic gene mutations, and tumor-infiltrating immune cells differed significantly between the high- and low-risk groups. CONCLUSIONS: This immune-related gene signature could play an important role in guiding treatment, making prognoses, and potentially developing future clinical applications.

BACKGROUND: Gliomas are highly complex and heterogeneous tumors, rendering prognosis prediction challenging. The advent of deep learning algorithms and the accessibility of multi-omic data represent a new approach for the identification of survival-sensitive subtypes. Herein, an autoencoder-based approach was used to identify two survival-sensitive subtypes using RNA sequencing (RNA-seq) and DNA methylation (DNAm) data from The Cancer Genome Atlas (TCGA) dataset. The subtypes were used as labels to build a support vector machine model with cross-validation. We validated the robustness of the model on Chinese Glioma Genome Atlas (CGGA) dataset. DNAm-driven genes were identified by integrating DNAm and gene expression profiling analyses using the R MethylMix package and carried out for further enrichment analysis. RESULTS: For TCGA dataset, the model produced a high C-index (0.92 ± 0.02), low brier score (0.16 ± 0.02), and significant log-rank p value (p < 0.0001). The model also had a decent performance for CGGA dataset (CGGA DNAm: C-index of 0.70, brier score of 0.21; CGGA RNA-seq: C-index of 0.79, brier score of 0.18). Moreover, we identified 389 DNAm-driven genes of survival-sensitive subtypes, which were significantly enriched in the glutathione metabolism pathway. CONCLUSIONS: Our study identified two survival-sensitive subtypes of glioma and provided insights into the molecular mechanisms underlying glioma development; thus, potentially providing a new target for the prognostic prediction of gliomas and supporting personalized treatment strategies.

BACKGROUND: Sequencing of thousands of samples provides genetic variants with allele frequencies spanning a very large spectrum and gives invaluable insight into genetic determinants of diseases. Protecting the genetic privacy of participants is challenging as only a few rare variants can easily re-identify an individual among millions. In certain cases, there are policy barriers against sharing genetic data from indigenous populations and stigmatizing conditions. RESULTS: We present SVAT, a method for secure outsourcing of variant annotation and aggregation, which are two basic steps in variant interpretation and detection of causal variants. SVAT uses homomorphic encryption to encrypt the data at the client-side. The data always stays encrypted while it is stored, in-transit, and most importantly while it is analyzed. SVAT makes use of a vectorized data representation to convert annotation and aggregation into efficient vectorized operations in a single framework. Also, SVAT utilizes a secure re-encryption approach so that multiple disparate genotype datasets can be combined for federated aggregation and secure computation of allele frequencies on the aggregated dataset. CONCLUSIONS: Overall, SVAT provides a secure, flexible, and practical framework for privacy-aware outsourcing of annotation, filtering, and aggregation of genetic variants. SVAT is publicly available for download from https://github.com/harmancilab/SVAT .

BACKGROUND: We used a hybrid machine learning systems (HMLS) strategy that includes the extensive search for the discovery of the most optimal HMLSs, including feature selection algorithms, a feature extraction algorithm, and classifiers for diagnosing breast cancer. Hence, this study aims to obtain a high-importance transcriptome profile linked with classification procedures that can facilitate the early detection of breast cancer. METHODS: In the present study, 762 breast cancer patients and 138 solid tissue normal subjects were included. Three groups of machine learning (ML) algorithms were employed: (i) four feature selection procedures are employed and compared to select the most valuable feature: (1) ANOVA; (2) Mutual Information; (3) Extra Trees Classifier; and (4) Logistic Regression (LGR), (ii) a feature extraction algorithm (Principal Component Analysis), iii) we utilized 13 classification algorithms accompanied with automated ML hyperparameter tuning, including (1) LGR; (2) Support Vector Machine; (3) Bagging; (4) Gaussian Naive Bayes; (5) Decision Tree; (6) Gradient Boosting Decision Tree; (7) K Nearest Neighborhood; (8) Bernoulli Naive Bayes; (9) Random Forest; (10) AdaBoost, (11) ExtraTrees; (12) Linear Discriminant Analysis; and (13) Multilayer Perceptron (MLP). For evaluating the proposed models' performance, balance accuracy and area under the curve (AUC) were used. RESULTS: Feature selection procedure LGR + MLP classifier achieved the highest prediction accuracy and AUC (balanced accuracy: 0.86, AUC = 0.94), followed by an LGR + LGR classifier (balanced accuracy: 0.84, AUC = 0.94). The results showed that achieved AUC for the LGR + LGR classifier belonged to the 20 biomarkers as follows: TMEM212, SNORD115-13, ATP1A4, FRG2, CFHR4, ZCCHC13, FLJ46361, LY6G6E, ZNF323, KRT28, KRT25, LPPR5, C10orf99, PRKACG, SULT2A1, GRIN2C, EN2, GBA2, CUX2, and SNORA66. CONCLUSIONS: The best performance was achieved using the LGR feature selection procedure and MLP classifier. Results show that the 20 biomarkers had the highest score or ranking in breast cancer detection.