MOTIVATION: Accurate identification of Drug-Target Interactions (DTIs) plays a crucial role in many stages of drug development and drug repurposing. (i) Traditional methods do not consider the use of multi-source data and do not consider the complex relationship between data sources. (ii) How to better mine the hidden features of drug and target space from high-dimensional data, and better solve the accuracy and robustness of the model. RESULTS: To solve the above problems, a novel prediction model named VGAEDTI is proposed in this paper. We constructed a heterogeneous network with multiple sources of information using multiple types of drug and target dataIn order to obtain deeper features of drugs and targets, we use two different autoencoders. One is variational graph autoencoder (VGAE) which is used to infer feature representations from drug and target spaces. The second is graph autoencoder (GAE) propagating labels between known DTIs. Experimental results on two public datasets show that the prediction accuracy of VGAEDTI is better than that of six DTIs prediction methods. These results indicate that model can predict new DTIs and provide an effective tool for accelerating drug development and repurposing.

BACKGROUND: Molecular interaction networks have become an important tool in providing context to the results of various omics experiments. For example, by integrating transcriptomic data and protein-protein interaction (PPI) networks, one can better understand how the altered expression of several genes are related with one another. The challenge then becomes how to determine, in the context of the interaction network, the subset(s) of genes that best captures the main mechanisms underlying the experimental conditions. Different algorithms have been developed to address this challenge, each with specific biological questions in mind. One emerging area of interest is to determine which genes are equivalently or inversely changed between different experiments. The equivalent change index (ECI) is a recently proposed metric that measures the extent to which a gene is equivalently or inversely regulated between two experiments. The goal of this work is to develop an algorithm that makes use of the ECI and powerful network analysis techniques to identify a connected subset of genes that are highly relevant to the experimental conditions. RESULTS: To address the above goal, we developed a method called Active Module identification using Experimental data and Network Diffusion (AMEND). The AMEND algorithm is designed to find a subset of connected genes in a PPI network that have large experimental values. It makes use of random walk with restart to create gene weights, and a heuristic solution to the Maximum-weight Connected Subgraph problem using these weights. This is performed iteratively until an optimal subnetwork (i.e., active module) is found. AMEND was compared to two current methods, NetCore and DOMINO, using two gene expression datasets. CONCLUSION: The AMEND algorithm is an effective, fast, and easy-to-use method for identifying network-based active modules. It returned connected subnetworks with the largest median ECI by magnitude, capturing distinct but related functional groups of genes. Code is freely available at https://github.com/samboyd0/AMEND .

BACKGROUND: In many applications of bioinformatics, data stem from distinct heterogeneous sources. One of the well-known examples is the identification of drug-target interactions (DTIs), which is of significant importance in drug discovery. In this paper, we propose a novel framework, manifold optimization based kernel preserving embedding (MOKPE), to efficiently solve the problem of modeling heterogeneous data. Our model projects heterogeneous drug and target data into a unified embedding space by preserving drug-target interactions and drug-drug, target-target similarities simultaneously. RESULTS: We performed ten replications of ten-fold cross validation on four different drug-target interaction network data sets for predicting DTIs for previously unseen drugs. The classification evaluation metrics showed better or comparable performance compared to previous similarity-based state-of-the-art methods. We also evaluated MOKPE on predicting unknown DTIs of a given network. Our implementation of the proposed algorithm in R together with the scripts that replicate the reported experiments is publicly available at https://github.com/ocbinatli/mokpe .

BACKGROUND: N6-methyladenosine (m6A) and 5-methylcytosine (m5C) are the main RNA methylation modifications involved in the oncogenesis of cancer. However, it remains obscure whether m6A/m5C-related long non-coding RNAs (lncRNAs) affect the development and progression of low grade gliomas (LGG). METHODS: We summarized 926 LGG tumor samples with RNA-seq data and clinical information from The Cancer Genome Atlas and Chinese Glioma Genome Atlas. 105 normal brain samples with RNA-seq data from the Genotype Tissue Expression project were collected for control. We obtained a molecular classification cluster from the expression pattern of sreened lncRNAs. The least absolute shrinkage and selection operator Cox regression was employed to construct a m6A/m5C-related lncRNAs prognostic signature of LGG. In vitro experiments were employed to validate the biological functions of lncRNAs in our risk model. RESULTS: The expression pattern of 14 sreened highly correlated lncRNAs could cluster samples into two groups, in which various clinicopathological features and the tumor immune microenvironment were significantly distinct. The survival time of cluster 1 was significantly reduced compared with cluster 2. This prognostic signature is based on 8 m6A/m5C-related lncRNAs (GDNF-AS1, HOXA-AS3, LINC00346, LINC00664, LINC00665, MIR155HG, NEAT1, RHPN1-AS1). Patients in the high-risk group harbored shorter survival times. Immunity microenvironment analysis showed B cells, CD4 + T cells, macrophages, and myeloid-derived DC cells were significantly increased in the high-risk group. Patients in high-risk group had the worse overall survival time regardless of followed TMZ therapy or radiotherapy. All observed results from the TCGA-LGG cohort could be validated in CGGA cohort. Afterwards, LINC00664 was found to promote cell viability, invasion and migration ability of glioma cells in vitro. CONCLUSION: Our study elucidated a prognostic prediction model of LGG by 8 m6A/m5C methylated lncRNAs and a critical lncRNA regulation function involved in LGG progression. High-risk patients have shorter survival times and a pro-tumor immune microenvironment.

BACKGROUND: P4 medicine (predict, prevent, personalize, and participate) is a new approach to diagnosing and predicting diseases on a patient-by-patient basis. For the prevention and treatment of diseases, prediction plays a fundamental role. One of the intelligent strategies is the design of deep learning models that can predict the state of the disease using gene expression data. RESULTS: We create an autoencoder deep learning model called DeeP4med, including a Classifier and a Transferor that predicts cancer's gene expression (mRNA) matrix from its matched normal sample and vice versa. The range of the F1 score of the model, depending on tissue type in the Classifier, is from 0.935 to 0.999 and in Transferor from 0.944 to 0.999. The accuracy of DeeP4med for tissue and disease classification was 0.986 and 0.992, respectively, which performed better compared to seven classic machine learning models (Support Vector Classifier, Logistic Regression, Linear Discriminant Analysis, Naive Bayes, Decision Tree, Random Forest, K Nearest Neighbors). CONCLUSIONS: Based on the idea of DeeP4med, by having the gene expression matrix of a normal tissue, we can predict its tumor gene expression matrix and, in this way, find effective genes in transforming a normal tissue into a tumor tissue. Results of Differentially Expressed Genes (DEGs) and enrichment analysis on the predicted matrices for 13 types of cancer showed a good correlation with the literature and biological databases. This led that by using the gene expression matrix, to train the model with features of each person in a normal and cancer state, this model could predict diagnosis based on gene expression data from healthy tissue and be used to identify possible therapeutic interventions for those patients.

BACKGROUND: Integrating multi-omics data is fast becoming a powerful approach for predicting disease progression and treatment outcomes. In light of that, we introduce a modified version of the NetRank algorithm, a network-based algorithm for biomarker discovery that incorporates the protein associations, co-expressions, and functions with its phenotypic association to differentiate different types of cancer. NetRank is introduced here as a robust feature selection method for biomarker selection in cancer prediction. We assess the robustness and suitability of the RNA gene expression data through scanning genomic data for 19 cancer types with more than 3000 patients from The Cancer Genome Atlas (TCGA). RESULTS: The results of evaluating different cancer type profiles from the TCGA data demonstrate the strength of our approach to identifying interpretable biomarker signatures for cancer outcome prediction. NetRank's biomarkers segregate most cancer types with an area under the curve (AUC) above 90% using compact signatures. CONCLUSION: In this paper we provide a fast and efficient implementation of NetRank, with a case study from The Cancer Genome Atlas, to assess the performance. We incorporated complete functionality for pre and post-processing for RNA-seq gene expression data with functions for building protein-protein interaction networks. The source code of NetRank is freely available (at github.com/Alfatlawi/Omics-NetRank) with an installable R library. We also deliver a comprehensive practical user manual with examples and data attached to this paper.

BACKGROUND: The growing power and ever decreasing cost of RNA sequencing (RNA-Seq) technologies have resulted in an explosion of RNA-Seq data production. Comparing gene expression values within RNA-Seq datasets is relatively easy for many interdisciplinary biomedical researchers; however, user-friendly software applications increase the ability of biologists to efficiently explore available datasets. RESULTS: Here, we describe ROGUE (RNA-Seq Ontology Graphic User Environment, https://marisshiny. RESEARCH: chop.edu/ROGUE/ ), a user-friendly R Shiny application that allows a biologist to perform differentially expressed gene analysis, gene ontology and pathway enrichment analysis, potential biomarker identification, and advanced statistical analyses. We use ROGUE to identify potential biomarkers and show unique enriched pathways between various immune cells. CONCLUSIONS: User-friendly tools for the analysis of next generation sequencing data, such as ROGUE, will allow biologists to efficiently explore their datasets, discover expression patterns, and advance their research by allowing them to develop and test hypotheses.

BACKGROUND: Single-cell RNA sequencing (scRNA-seq) enables the high-throughput profiling of gene expression at the single-cell level. However, overwhelming dropouts within data may obscure meaningful biological signals. Various imputation methods have recently been developed to address this problem. Therefore, it is important to perform a systematic evaluation of different imputation algorithms. RESULTS: In this study, we evaluated 11 of the most recent imputation methods on 12 real biological datasets from immunological studies and 4 simulated datasets. The performance of these methods was compared, based on numerical recovery, cell clustering and marker gene analysis. Most of the methods brought some benefits on numerical recovery. To some extent, the performance of imputation methods varied among protocols. In the cell clustering analysis, no method performed consistently well across all datasets. Some methods performed poorly on real datasets but excellent on simulated datasets. Surprisingly and importantly, some methods had a negative effect on cell clustering. In marker gene analysis, some methods identified potentially novel cell subsets. However, not all of the marker genes were successfully imputed in gene expression, suggesting that imputation challenges remain. CONCLUSIONS: In summary, different imputation methods showed different effects on different datasets, suggesting that imputation may have dataset specificity. Our study reveals the benefits and limitations of various imputation methods and provides a data-driven guidance for scRNA-seq data analysis.

BACKGROUND: The identification of tumor T cell antigens (TTCAs) is crucial for providing insights into their functional mechanisms and utilizing their potential in anticancer vaccines development. In this context, TTCAs are highly promising. Meanwhile, experimental technologies for discovering and characterizing new TTCAs are expensive and time-consuming. Although many machine learning (ML)-based models have been proposed for identifying new TTCAs, there is still a need to develop a robust model that can achieve higher rates of accuracy and precision. RESULTS: In this study, we propose a new stacking ensemble learning-based framework, termed StackTTCA, for accurate and large-scale identification of TTCAs. Firstly, we constructed 156 different baseline models by using 12 different feature encoding schemes and 13 popular ML algorithms. Secondly, these baseline models were trained and employed to create a new probabilistic feature vector. Finally, the optimal probabilistic feature vector was determined based the feature selection strategy and then used for the construction of our stacked model. Comparative benchmarking experiments indicated that StackTTCA clearly outperformed several ML classifiers and the existing methods in terms of the independent test, with an accuracy of 0.932 and Matthew's correlation coefficient of 0.866. CONCLUSIONS: In summary, the proposed stacking ensemble learning-based framework of StackTTCA could help to precisely and rapidly identify true TTCAs for follow-up experimental verification. In addition, we developed an online web server ( http://2pmlab.camt.cmu.ac.th/StackTTCA ) to maximize user convenience for high-throughput screening of novel TTCAs.

BACKGROUND: Modern genome sequencing leads to an ever-growing collection of genomic annotations. Combining these elements with a set of input regions (e.g. genes) would yield new insights in genomic associations, such as those involved in gene regulation. The required data are scattered across different databases making a manual approach tiresome, unpractical, and prone to error. Semi-automatic approaches require programming skills in data parsing, processing, overlap calculation, and visualization, which most biomedical researchers lack. Our aim was to develop an automated tool providing all necessary algorithms, benefiting both bioinformaticians and researchers without bioinformatic training. RESULTS: We developed overlapping annotated genomic regions (OGRE) as a comprehensive tool to associate and visualize input regions with genomic annotations. It does so by parsing regions of interest, mining publicly available annotations, and calculating possible overlaps between them. The user can thus identify location, type, and number of associated regulatory elements. Results are presented as easy to understand visualizations and result tables. We applied OGRE to recent studies and could show high reproducibility and potential new insights. To demonstrate OGRE's performance in terms of running time and output, we have conducted a benchmark and compared its features with similar tools. CONCLUSIONS: OGRE's functions and built-in annotations can be applied as a downstream overlap association step, which is compatible with most genomic sequencing outputs, and can thus enrich pre-existing analyses pipelines. Compared to similar tools, OGRE shows competitive performance, offers additional features, and has been successfully applied to two recent studies. Overall, OGRE addresses the lack of tools for automatic analysis, local genomic overlap calculation, and visualization by providing an easy to use, end-to-end solution for both biologists and computational scientists.

BACKGROUND: An updated version of the mwtab Python package for programmatic access to the Metabolomics Workbench (MetabolomicsWB) data repository was released at the beginning of 2021. Along with updating the package to match the changes to MetabolomicsWB's 'mwTab' file format specification and enhancing the package's functionality, the included validation facilities were used to detect and catalog file inconsistencies and errors across all publicly available datasets in MetabolomicsWB. RESULTS: The MetabolomicsWB File Status website was developed to provide continuous validation of MetabolomicsWB data files and a useful interface to all found inconsistencies and errors. This list of detectable issues/errors include format parsing errors, format compliance issues, access problems via MetabolomicsWB's REST interface, and other small inconsistencies that can hinder reusability. The website uses the mwtab Python package to pull down and validate each available analysis file and then generates an html report. The website is updated on a weekly basis. Moreover, the Python website design utilizes GitHub and GitHub.io, providing an easy to replicate template for implementing other metadata, virtual, and meta- repositories. CONCLUSIONS: The MetabolomicsWB File Status website provides a metadata repository of validation metadata to promote the FAIR use of existing metabolomics datasets from the MetabolomicsWB data repository.

BACKGROUND: Protein biomarkers of cancer progression and response to therapy are increasingly important for improving personalized medicine. Advanced quantitative pathology platforms enable measurement of protein expression in tissues at the single-cell level. However, this rich quantitative cell-by-cell biomarker information is most often not exploited. Instead, it is reduced to a single mean across the cells of interest or converted into a simple proportion of binary biomarker-positive or -negative cells. RESULTS: We investigated the utility of retaining all quantitative information at the single-cell level by considering the values of the quantile function (inverse of the cumulative distribution function) estimated from a sample of cell signal intensity levels in a tumor tissue. An algorithm was developed for selecting optimal cutoffs for dichotomizing cell signal intensity distribution quantiles as predictors of continuous, categorical or survival outcomes. The proposed algorithm was used to select optimal quantile biomarkers of breast cancer progression based on cancer cells' cell signal intensity levels of nuclear protein Ki-67, Proliferating cell nuclear antigen, Programmed cell death 1 ligand 2, and Progesterone receptor. The performance of the resulting optimal quantile biomarkers was validated and compared to the standard cancer compartment mean signal intensity markers using an independent external validation cohort. For Ki-67, the optimal quantile biomarker was also compared to established biomarkers based on percentages of Ki67-positive cells. For proteins significantly associated with PFS in the external validation cohort, the optimal quantile biomarkers yielded either larger or similar effect size (hazard ratio for progression-free survival) as compared to cancer compartment mean signal intensity biomarkers. CONCLUSION: The optimal quantile protein biomarkers yield generally improved prognostic value as compared to the standard protein expression markers. The proposed methodology has a broad application to single-cell data from genomics, transcriptomics, proteomics, or metabolomics studies at the single cell level.

BACKGROUND: Accurate prediction of molecular property holds significance in contemporary drug discovery and medical research. Recent advances in AI-driven molecular property prediction have shown promising results. Due to the costly annotation of in vitro and in vivo experiments, transfer learning paradigm has been gaining momentum in extracting general self-supervised information to facilitate neural network learning. However, prior pretraining strategies have overlooked the necessity of explicitly incorporating domain knowledge, especially the molecular fragments, into model design, resulting in the under-exploration of the molecular semantic space. RESULTS: We propose an effective model with FRagment-based dual-channEL pretraining (FREL). Equipped with molecular fragments, FREL comprehensively employs masked autoencoder and contrastive learning to learn intra- and inter-molecule agreement, respectively. We further conduct extensive experiments on ten public datasets to demonstrate its superiority over state-of-the-art models. Further investigations and interpretations manifest the underlying relationship between molecular representations and molecular properties. CONCLUSIONS: Our proposed model FREL achieves state-of-the-art performance on the benchmark datasets, emphasizing the importance of incorporating molecular fragments into model design. The expressiveness of learned molecular representations is also investigated by visualization and correlation analysis. Case studies indicate that the learned molecular representations better capture the drug property variation and fragment semantics.

BACKGROUND: Statistical correlation analysis is currently the most typically used approach for investigating the risk factors of type 2 diabetes mellitus (T2DM). However, this approach does not readily reveal the causal relationships between risk factors and rarely describes the causal relationships visually. RESULTS: Considering the superiority of reinforcement learning in prediction, a causal discovery approach with reinforcement learning for T2DM risk factors is proposed herein. First, a reinforcement learning model is constructed for T2DM risk factors. Second, the process involved in the causal discovery method for T2DM risk factors is detailed. Finally, several experiments are designed based on diabetes datasets and used to verify the proposed approach. CONCLUSIONS: The experimental results show that the proposed approach improves the accuracy of causality mining between T2DM risk factors and provides new evidence to researchers engaged in T2DM prevention and treatment research.

BACKGROUND: Protein engineering aims to improve the functional properties of existing proteins to meet people's needs. Current deep learning-based models have captured evolutionary, functional, and biochemical features contained in amino acid sequences. However, the existing generative models need to be improved when capturing the relationship between amino acid sites on longer sequences. At the same time, the distribution of protein sequences in the homologous family has a specific positional relationship in the latent space. We want to use this relationship to search for new variants directly from the vicinity of better-performing varieties. RESULTS: To improve the representation learning ability of the model for longer sequences and the similarity between the generated sequences and the original sequences, we propose a temporal variational autoencoder (T-VAE) model. T-VAE consists of an encoder and a decoder. The encoder expands the receptive field of neurons in the network structure by dilated causal convolution, thereby improving the encoding representation ability of longer sequences. The decoder decodes the sampled data into variants closely resembling the original sequence. CONCLUSION: Compared to other models, the person correlation coefficient between the predicted values of protein fitness obtained by T-VAE and the truth values was higher, and the mean absolute deviation was lower. In addition, the T-VAE model has a better representation learning ability for longer sequences when comparing the encoding of protein sequences of different lengths. These results show that our model has more advantages in representation learning for longer sequences. To verify the model's generative effect, we also calculate the sequence identity between the generated data and the input data. The sequence identity obtained by T-VAE improved by 12.9% compared to the baseline model.

To understand genome evolution in a group of microbes, we need to know the timing of events such as duplications, deletions and horizontal transfers. A common approach is to perform a gene-tree / species-tree reconciliation. While a number of software packages perform this type of analysis, none are geared toward a complete reconstruction for all families in an entire clade. Here we describe an update to the xenoGI software package which allows users to perform such an analysis using the newly developed DTLOR (duplication-transfer-loss-origin-rearrangement) reconciliation model starting from genome sequences as input.

BACKGROUND: Identifying variants associated with diseases is a challenging task in medical genetics research. Current studies that prioritize variants within individual genomes generally rely on known variants, evidence from literature and genomes, and patient symptoms and clinical signs. The functionalities of the existing tools, which rank variants based on given patient symptoms and clinical signs, are restricted to the coverage of ontologies such as the Human Phenotype Ontology (HPO). However, most clinicians do not limit themselves to HPO while describing patient symptoms/signs and their associated variants/genes. There is thus a need for an automated tool that can prioritize variants based on freely expressed patient symptoms and clinical signs. RESULTS: STARVar is a Symptom-based Tool for Automatic Ranking of Variants using evidence from literature and genomes. STARVar uses patient symptoms and clinical signs, either linked to HPO or expressed in free text format. It returns a ranked list of variants based on a combined score from two classifiers utilizing evidence from genomics and literature. STARVar improves over related tools on a set of synthetic patients. In addition, we demonstrated its distinct contribution to the domain on another synthetic dataset covering publicly available clinical genotype-phenotype associations by using symptoms and clinical signs expressed in free text format. CONCLUSIONS: STARVar stands as a unique and efficient tool that has the advantage of ranking variants with flexibly expressed patient symptoms in free-form text. Therefore, STARVar can be easily integrated into bioinformatics workflows designed to analyze disease-associated genomes. AVAILABILITY: STARVar is freely available from https://github.com/bio-ontology-research-group/STARVar .

BACKGROUND: The accelerating pace of biomedical publication has made it impractical to manually, systematically identify papers containing specific information and extract this information. This is especially challenging when the information itself resides beyond titles or abstracts. For emerging science, with a limited set of known papers of interest and an incomplete information model, this is of pressing concern. A timely example in retrospect is the identification of immune signatures (coherent sets of biomarkers) driving differential SARS-CoV-2 infection outcomes. IMPLEMENTATION: We built a classifier to identify papers containing domain-specific information from the document embeddings of the title and abstract. To train this classifier with limited data, we developed an iterative process leveraging pre-trained SPECTER document embeddings, SVM classifiers and web-enabled expert review to iteratively augment the training set. This training set was then used to create a classifier to identify papers containing domain-specific information. Finally, information was extracted from these papers through a semi-automated system that directly solicited the paper authors to respond via a web-based form. RESULTS: We demonstrate a classifier that retrieves papers with human COVID-19 immune signatures with a positive predictive value of 86%. The type of immune signature (e.g., gene expression vs. other types of profiling) was also identified with a positive predictive value of 74%. Semi-automated queries to the corresponding authors of these publications requesting signature information achieved a 31% response rate. CONCLUSIONS: Our results demonstrate the efficacy of using a SVM classifier with document embeddings of the title and abstract, to retrieve papers with domain-specific information, even when that information is rarely present in the abstract. Targeted author engagement based on classifier predictions offers a promising pathway to build a semi-structured representation of such information. Through this approach, partially automated literature mining can help rapidly create semi-structured knowledge repositories for automatic analysis of emerging health threats.

BACKGROUND: The rapid expansion of Whole-Genome Sequencing has revolutionized the fields of clinical and food microbiology. However, its implementation as a routine laboratory technique remains challenging due to the growth of data at a faster rate than can be effectively analyzed and critical gaps in bioinformatics knowledge. RESULTS: To address both issues, CamPype was developed as a new bioinformatics workflow for the genomics analysis of sequencing data of bacteria, especially Campylobacter, which is the main cause of gastroenteritis worldwide making a negative impact on the economy of the public health systems. CamPype allows fully customization of stages to run and tools to use, including read quality control filtering, read contamination, reads extension and assembly, bacterial typing, genome annotation, searching for antibiotic resistance genes, virulence genes and plasmids, pangenome construction and identification of nucleotide variants. All results are processed and resumed in an interactive HTML report for best data visualization and interpretation. CONCLUSIONS: The minimal user intervention of CamPype makes of this workflow an attractive resource for microbiology laboratories with no expertise in bioinformatics as a first line method for bacterial typing and epidemiological analyses, that would help to reduce the costs of disease outbreaks, or for comparative genomic analyses. CamPype is publicly available at https://github.com/JoseBarbero/CamPype .

BACKGROUND: The growing recognition of the microbiome's impact on human health and well-being has prompted extensive research into discovering the links between microbiome dysbiosis and disease (healthy) states. However, this valuable information is scattered in unstructured form within biomedical literature. The structured extraction and qualification of microbe-disease interactions are important. In parallel, recent advancements in deep-learning-based natural language processing algorithms have revolutionized language-related tasks such as ours. This study aims to leverage state-of-the-art deep-learning language models to extract microbe-disease relationships from biomedical literature. RESULTS: In this study, we first evaluate multiple pre-trained large language models within a zero-shot or few-shot learning context. In this setting, the models performed poorly out of the box, emphasizing the need for domain-specific fine-tuning of these language models. Subsequently, we fine-tune multiple language models (specifically, GPT-3, BioGPT, BioMedLM, BERT, BioMegatron, PubMedBERT, BioClinicalBERT, and BioLinkBERT) using labeled training data and evaluate their performance. Our experimental results demonstrate the state-of-the-art performance of these fine-tuned models ( specifically GPT-3, BioMedLM, and BioLinkBERT), achieving an average F1 score, precision, and recall of over [Formula: see text] compared to the previous best of  0.74. CONCLUSION: Overall, this study establishes that pre-trained language models excel as transfer learners when fine-tuned with domain and problem-specific data, enabling them to achieve state-of-the-art results even with limited training data for extracting microbiome-disease interactions from scientific publications.

BACKGROUND: Cancer subtype classification is helpful for personalized cancer treatment. Although, some approaches have been developed to classifying caner subtype based on high dimensional gene expression data, it is difficult to obtain satisfactory classification results. Meanwhile, some cancers have been well studied and classified to some subtypes, which are adopt by most researchers. Hence, this priori knowledge is significant for further identifying new meaningful subtypes. RESULTS: In this paper, we present a combined parallel random forest and autoencoder approach for cancer subtype identification based on high dimensional gene expression data, ForestSubtype. ForestSubtype first adopts the parallel RF and the priori knowledge of cancer subtype to train a module and extract significant candidate features. Second, ForestSubtype uses a random forest as the base module and ten parallel random forests to compute each feature weight and rank them separately. Then, the intersection of the features with the larger weights output by the ten parallel random forests is taken as our subsequent candidate features. Third, ForestSubtype uses an autoencoder to condenses the selected features into a two-dimensional data. Fourth, ForestSubtype utilizes k-means++ to obtain new cancer subtype identification results. In this paper, the breast cancer gene expression data obtained from The Cancer Genome Atlas are used for training and validation, and an independent breast cancer dataset from the Molecular Taxonomy of Breast Cancer International Consortium is used for testing. Additionally, we use two other cancer datasets for validating the generalizability of ForestSubtype. ForestSubtype outperforms the other two methods in terms of the distribution of clusters, internal and external metric results. The open-source code is available at https://github.com/lffyd/ForestSubtype . CONCLUSIONS: Our work shows that the combination of high-dimensional gene expression data and parallel random forests and autoencoder, guided by a priori knowledge, can identify new subtypes more effectively than existing methods of cancer subtype classification.

Deep learning-based medical image segmentation has made great progress over the past decades. Scholars have proposed many novel transformer-based segmentation networks to solve the problems of building long-range dependencies and global context connections in convolutional neural networks (CNNs). However, these methods usually replace the CNN-based blocks with improved transformer-based structures, which leads to the lack of local feature extraction ability, and these structures require a huge number of data for training. Moreover, those methods did not pay attention to edge information, which is essential in medical image segmentation. To address these problems, we proposed a new network structure, called P-TransUNet. This network structure combines the designed efficient P-Transformer and the fusion module, which extract distance-related long-range dependencies and local information respectively and produce the fused features. Besides, we introduced edge loss into training to focus the attention of the network on the edge of the lesion area to improve segmentation performance. Extensive experiments across four tasks of medical image segmentation demonstrated the effectiveness of P-TransUNet, and showed that our network outperforms other state-of-the-art methods.

BACKGROUND: Integration site (IS) analysis is a fundamental analytical platform for evaluating the safety and efficacy of viral vector based preclinical and clinical Gene Therapy (GT). A handful of groups have developed standardized bioinformatics pipelines to process IS sequencing data, to generate reports, and/or to perform comparative studies across different GT trials. Keeping up with the technological advances in the field of IS analysis, different computational pipelines have been published over the past decade. These pipelines focus on identifying IS from single-read sequencing or paired-end sequencing data either using read-based or using sonication fragment-based methods, but there is a lack of a bioinformatics tool that automatically includes unique molecular identifiers (UMI) for IS abundance estimations and allows comparing multiple quantification methods in one integrated pipeline. RESULTS: Here we present IS-Seq a bioinformatics pipeline that can process data from paired-end sequencing of both old restriction sites-based IS collection methods and new sonication-based IS retrieval systems while allowing the selection of different abundance estimation methods, including read-based, Fragment-based and UMI-based systems. CONCLUSIONS: We validated the performance of IS-Seq by testing it against the most popular  analytical workflow available in the literature (INSPIIRED) and using different scenarios. Lastly, by performing extensive simulation studies and a comprehensive wet-lab assessment of our IS-Seq pipeline we could show that in clinically relevant scenarios, UMI quantification provides better accuracy than the currently most widely used sonication fragment counts as a method for IS abundance estimation.

BACKGROUND: Next-generation sequencing technologies yield large numbers of genetic alterations, of which a subset are missense variants that alter an amino acid in the protein product. These variants can have a potentially destabilizing effect leading to an increased risk of misfolding and aggregation. Multiple software tools exist to predict the effect of single-nucleotide variants on proteins, however, a pipeline integrating these tools while starting from an NGS data output list of variants is lacking. RESULTS: The previous version SNPeffect 4.0 (De Baets in Nucleic Acids Res 40(D1):D935-D939, 2011) provided an online database containing pre-calculated variant effects and low-throughput custom variant analysis. Here, we built an automated and parallelized pipeline that analyzes the impact of missense variants on the aggregation propensity and structural stability of proteins starting from the Variant Call Format as input. The pipeline incorporates the AlphaFold Protein Structure Database to achieve high coverage for structural stability analyses using the FoldX force field. The effect on aggregation-propensity is analyzed using the established predictors TANGO and WALTZ. The pipeline focuses solely on the human proteome and can be used to analyze proteome stability/damage in a given sample based on sequencing results. CONCLUSION: We provide a bioinformatics pipeline that allows structural phenotyping from sequencing data using established stability and aggregation predictors including FoldX, TANGO, and WALTZ; and structural proteome coverage provided by the AlphaFold database. The pipeline and installation guide are freely available for academic users on https://github.com/vibbits/snpeffect and requires a computer cluster.

BACKGROUND: Local assembly with short and long reads has proven to be very useful in many applications: reconstruction of the sequence of a locus of interest, gap-filling in draft assemblies, as well as alternative allele reconstruction of large Structural Variants. Whereas linked-read technologies have a great potential to assemble specific loci as they provide long-range information while maintaining the power and accuracy of short-read sequencing, there is a lack of local assembly tools for linked-read data. RESULTS: We present MTG-Link, a novel local assembly tool dedicated to linked-reads. The originality of the method lies in its read subsampling step which takes advantage of the barcode information contained in linked-reads mapped in flanking regions. We validated our approach on several datasets from different linked-read technologies. We show that MTG-Link is able to assemble successfully large sequences, up to dozens of Kb. We also demonstrate that the read subsampling step of MTG-Link considerably improves the local assembly of specific loci compared to other existing short-read local assembly tools. Furthermore, MTG-Link was able to fully characterize large insertion variants and deletion breakpoints in a human genome and to reconstruct dark regions in clinically-relevant human genes. It also improved the contiguity of a 1.3 Mb locus of biological interest in several individual genomes of the mimetic butterfly Heliconius numata. CONCLUSIONS: MTG-Link is an efficient local assembly tool designed for different linked-read sequencing technologies. MTG-Link source code is available at https://github.com/anne-gcd/MTG-Link and as a Bioconda package.

BACKGROUND: Kernel methods have been proven to be a powerful tool for the integration and analysis of high-throughput technologies generated data. Kernels offer a nonlinear version of any linear algorithm solely based on dot products. The kernelized version of principal component analysis is a valid nonlinear alternative to tackle the nonlinearity of biological sample spaces. This paper proposes a novel methodology to obtain a data-driven feature importance based on the kernel PCA representation of the data. RESULTS: The proposed method, kernel PCA Interpretable Gradient (KPCA-IG), provides a data-driven feature importance that is computationally fast and based solely on linear algebra calculations. It has been compared with existing methods on three benchmark datasets. The accuracy obtained using KPCA-IG selected features is equal to or greater than the other methods' average. Also, the computational complexity required demonstrates the high efficiency of the method. An exhaustive literature search has been conducted on the selected genes from a publicly available Hepatocellular carcinoma dataset to validate the retained features from a biological point of view. The results once again remark on the appropriateness of the computed ranking. CONCLUSIONS: The black-box nature of kernel PCA needs new methods to interpret the original features. Our proposed methodology KPCA-IG proved to be a valid alternative to select influential variables in high-dimensional high-throughput datasets, potentially unravelling new biological and medical biomarkers.

MOTIVATION: Quantitative descriptions of multi-cellular structures from optical microscopy imaging are prime to understand the variety of three-dimensional (3D) shapes in living organisms. Experimental models of vertebrates, invertebrates and plants, such as zebrafish, killifish, Drosophila or Marchantia, mainly comprise multilayer tissues, and even if microscopes can reach the needed depth, their geometry hinders the selection and subsequent analysis of the optical volumes of interest. Computational tools to "peel" tissues by removing specific layers and reducing 3D volume into planar images, can critically improve visualization and analysis. RESULTS: We developed VolumePeeler, a versatile FIJI plugin for virtual 3D "peeling" of image stacks. The plugin implements spherical and spline surface projections. We applied VolumePeeler to perform peeling in 3D images of spherical embryos, as well as non-spherical tissue layers. The produced images improve the 3D volume visualization and enable analysis and quantification of geometrically challenging microscopy datasets. AVAILABILITY: ImageJ/FIJI software, source code, examples, and tutorials are openly available in https://cimt.uchile.cl/mcerda.

BACKGROUND: Network analysis is a powerful tool for studying gene regulation and identifying biological processes associated with gene function. However, constructing gene co-expression networks can be a challenging task, particularly when dealing with a large number of missing values. RESULTS: We introduce GeCoNet-Tool, an integrated gene co-expression network construction and analysis tool. The tool comprises two main parts: network construction and network analysis. In the network construction part, GeCoNet-Tool offers users various options for processing gene co-expression data derived from diverse technologies. The output of the tool is an edge list with the option of weights associated with each link. In network analysis part, the user can produce a table that includes several network properties such as communities, cores, and centrality measures. With GeCoNet-Tool, users can explore and gain insights into the complex interactions between genes.

Uveal melanoma arises from stromal melanocytes and is the most prevalent primary intraocular tumor in adults. It poses a significant diagnostic and therapeutic challenge due to its high malignancy and early onset of metastases. In recent years, there has been a growing interest in the role of diverse immune cells in tumor cell development and metastasis. Using The Cancer Genome Atlas and the gene expression omnibus databases, and the CIBERSORT method, we investigated the topography of intra-tumor immune infiltration in uveal melanoma in this research. We evaluated the prognosis of uveal melanoma patients using the M2 macrophage immune cell infiltration score in conjunction with clinical tumor patient data. We built a prognostic model based on the distinctive genes of M2 macrophages and combined it with patients' clinical data in the database; we ran a survival prognostic analysis to authenticate the model's accuracy. The functional study revealed the importance of macrophage-associated genes in the development of uveal melanoma. Moreover, the reliability of our prediction model was verified by combining tumor mutational load, immune checkpoint, and drug sensitivity, respectively. Our study provides a reference for the follow-up study of uveal melanoma.

BACKGROUND: Matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI MSI) displays significant potential for applications in cancer research, especially in tumor typing and subtyping. Lung cancer is the primary cause of tumor-related deaths, where the most lethal entities are adenocarcinoma (ADC) and squamous cell carcinoma (SqCC). Distinguishing between these two common subtypes is crucial for therapy decisions and successful patient management. RESULTS: We propose a new algebraic topological framework, which obtains intrinsic information from MALDI data and transforms it to reflect topological persistence. Our framework offers two main advantages. Firstly, topological persistence aids in distinguishing the signal from noise. Secondly, it compresses the MALDI data, saving storage space and optimizes computational time for subsequent classification tasks. We present an algorithm that efficiently implements our topological framework, relying on a single tuning parameter. Afterwards, logistic regression and random forest classifiers are employed on the extracted persistence features, thereby accomplishing an automated tumor (sub-)typing process. To demonstrate the competitiveness of our proposed framework, we conduct experiments on a real-world MALDI dataset using cross-validation. Furthermore, we showcase the effectiveness of the single denoising parameter by evaluating its performance on synthetic MALDI images with varying levels of noise. CONCLUSION: Our empirical experiments demonstrate that the proposed algebraic topological framework successfully captures and leverages the intrinsic spectral information from MALDI data, leading to competitive results in classifying lung cancer subtypes. Moreover, the framework's ability to be fine-tuned for denoising highlights its versatility and potential for enhancing data analysis in MALDI applications.

Pathogenic bacteria present a major threat to human health, causing various infections and illnesses, and in some cases, even death. The accurate identification of these bacteria is crucial, but it can be challenging due to the similarities between different species and genera. This is where automated classification using convolutional neural network (CNN) models can help, as it can provide more accurate, authentic, and standardized results.In this study, we aimed to create a larger and balanced dataset by image patching and applied different variations of CNN models, including training from scratch, fine-tuning, and weight adjustment, and data augmentation through random rotation, reflection, and translation. The results showed that the best results were achieved through augmentation and fine-tuning of deep models. We also modified existing architectures, such as InceptionV3 and MobileNetV2, to better capture complex features. The robustness of the proposed ensemble model was evaluated using two data splits (7:2:1 and 6:2:2) to see how performance changed as the training data was increased from 10 to 20%. In both cases, the model exhibited exceptional performance. For the 7:2:1 split, the model achieved an accuracy of 99.91%, F-Score of 98.95%, precision of 98.98%, recall of 98.96%, and MCC of 98.92%. For the 6:2:2 split, the model yielded an accuracy of 99.94%, F-Score of 99.28%, precision of 99.31%, recall of 98.96%, and MCC of 99.26%. This demonstrates that automatic classification using the ensemble model can be a valuable tool for diagnostic staff and microbiologists in accurately identifying pathogenic bacteria, which in turn can help control epidemics and minimize their social and economic impact.