BACKGROUND: Independent component analysis is an unsupervised machine learning algorithm that separates a set of mixed signals into a set of statistically independent source signals. Applied to high-quality gene expression datasets, independent component analysis effectively reveals both the source signals of the transcriptome as co-regulated gene sets, and the activity levels of the underlying regulators across diverse experimental conditions. Two major variables that affect the final gene sets are the diversity of the expression profiles contained in the underlying data, and the user-defined number of independent components, or dimensionality, to compute. Availability of high-quality transcriptomic datasets has grown exponentially as high-throughput technologies have advanced; however, optimal dimensionality selection remains an open question. METHODS: We computed independent components across a range of dimensionalities for four gene expression datasets with varying dimensions (both in terms of number of genes and number of samples). We computed the correlation between independent components across different dimensionalities to understand how the overall structure evolves as the number of user-defined components increases. We then measured how well the resulting gene clusters reflected known regulatory mechanisms, and developed a set of metrics to assess the accuracy of the decomposition at a given dimension. RESULTS: We found that over-decomposition results in many independent components dominated by a single gene, whereas under-decomposition results in independent components that poorly capture the known regulatory structure. From these results, we developed a new method, called OptICA, for finding the optimal dimensionality that controls for both over- and under-decomposition. Specifically, OptICA selects the highest dimension that produces a low number of components that are dominated by a single gene. We show that OptICA outperforms two previously proposed methods for selecting the number of independent components across four transcriptomic databases of varying sizes. CONCLUSIONS: OptICA avoids both over-decomposition and under-decomposition of transcriptomic datasets resulting in the best representation of the organism's underlying transcriptional regulatory network.

BACKGROUND: Single-cell sequencing technology can address the amount of single-cell library data at the same time and display the heterogeneity of different cells. However, analyzing single-cell data is a computationally challenging problem. Because there are low counts in the gene expression region, it has a high chance of recognizing the non-zero entity as zero, which are called dropout events. At present, the mainstream dropout imputation methods cannot effectively recover the true expression of cells from dropout noise such as DCA, MAGIC, scVI, scImpute and SAVER. RESULTS: In this paper, we propose an autoencoder structure network, named GNNImpute. GNNImpute uses graph attention convolution to aggregate multi-level similar cell information and implements convolution operations on non-Euclidean space on scRNA-seq data. Distinct from current imputation tools, GNNImpute can accurately and effectively impute the dropout and reduce dropout noise. We use mean square error (MSE), mean absolute error (MAE), Pearson correlation coefficient (PCC) and Cosine similarity (CS) to measure the performance of different methods with GNNImpute. We analyze four real datasets, and our results show that the GNNImpute achieves 3.0130 MSE, 0.6781 MAE, 0.9073 PCC and 0.9134 CS. Furthermore, we use Adjusted rand index (ARI) and Normalized mutual information (NMI) to measure the clustering effect. The GNNImpute achieves 0.8199 (ARI) and 0.8368 (NMI), respectively. CONCLUSIONS: In this investigation, we propose a single-cell dropout imputation method (GNNImpute), which effectively utilizes shared information for imputing the dropout of scRNA-seq data. We test it with different real datasets and evaluate its effectiveness in MSE, MAE, PCC and CS. The results show that graph attention convolution and autoencoder structure have great potential in single-cell dropout imputation.

BACKGROUND: Software for nuclear magnetic resonance (NMR) spectrometers offer general functionality of instrument control and data processing; these applications are often developed with non-scripting languages. NMR users need to flexibly integrate rapidly developing NMR applications with emerging technologies. Scripting systems offer open environments for NMR users to write custom programs. However, existing scripting systems have limited capabilities for both extending the functionality of NMR software's non-script main program and using advanced native script libraries to support specialized application domains (e.g., biomacromolecules and metabolomics). Therefore, it is essential to design a novel scripting system to address both of these needs. RESULT: Here, a novel NMR scripting system named SpinSPJ is proposed. It works as a plug-in in the Java based NMR spectrometer software SpinStudioJ. In the scripting system, both Java based NMR methods and original CPython based libraries are supported. A module has been developed as a bridge to integrate the runtime environments of Java and CPython. The module works as an extension in the CPython environment and interacts with Java via the Java Native Interface. Leveraging this bridge, Java based instrument control and data processing methods of SpinStudioJ can be called with the CPython style. Compared with traditional scripting systems, SpinSPJ better supports both extending the non-script main program and implementing advanced NMR applications with a rich variety of script libraries. NMR researchers can easily call functions of instrument control and data processing as well as developing complex functionality (such as multivariate statistical analysis, deep learning, etc.) with CPython native libraries. CONCLUSION: SpinSPJ offers a user-friendly environment to implement custom functionality leveraging its powerful basic NMR and rich CPython libraries. NMR applications with emerging technologies can be easily integrated. The scripting system is free of charge and can be downloaded by visiting http://www.spinstudioj.net/spinspj .

BACKGROUND: Autophagy is closely related to the progression of breast cancer. The aim at this study is to establish a prognostic-related model comprised of hub autophagy genes (AGs) to assess patient prognosis. Simultaneously, the model can guide clinicians to make up individualized strategies and stratify patients aged 40-60 years based on risk level. METHODS: The hub AGs were identified with univariate COX regression and LASSO regression. The functions and alterations of these selected AGs were analyzed as well. Moreover, the multivariate COX regression and correlation analysis between hub AGs and clinicopathological parameters were done. RESULTS: Totally, 33 prognostic-related AGs were obtained from the univariate COX regression (P < 0.05). SERPINA1, HSPA8, HSPB8, MAP1LC3A, and DIRAS3 were identified to constitute the prognostic model by the LASSO regression. The survival curve of patients in the high-risk and low-risk groups was statistically significant (P < 0.05). The 3-year and 5-year ROC displayed that their AUC value reached 0.762 and 0.825, respectively. Stage and risk scores were independent risk factors relevant to prognosis. RB1CC1, RPS6KB1, and BIRC6 were identified as the most predominant mutant genes. It was found that AGs were mainly involved in regulating the endopeptidases synthesis and played important roles in the ErbB signal pathway. SERPIN1, risk score was closely related to the stage (P < 0.05); HSPA8, risk score were closely related to T stag (P < 0.05); HSPB8 was closely related to N stag (P < 0.05). CONCLUSIONS: Our prognostic model had the relatively robust predictive ability on prognosis for patients aged 40-60 years. If the stage was added into the prognostic model, the predictive ability would be more powerful.

BACKGROUND: The technical development of imaging techniques in life sciences has enabled the three-dimensional recording of living samples at increasing temporal resolutions. Dynamic 3D data sets of developing organisms allow for time-resolved quantitative analyses of morphogenetic changes in three dimensions, but require efficient and automatable analysis pipelines to tackle the resulting Terabytes of image data. Particle image velocimetry (PIV) is a robust and segmentation-free technique that is suitable for quantifying collective cellular migration on data sets with different labeling schemes. This paper presents the implementation of an efficient 3D PIV package using the Julia programming language-quickPIV. Our software is focused on optimizing CPU performance and ensuring the robustness of the PIV analyses on biological data. RESULTS: QuickPIV is three times faster than the Python implementation hosted in openPIV, both in 2D and 3D. Our software is also faster than the fastest 2D PIV package in openPIV, written in C++. The accuracy evaluation of our software on synthetic data agrees with the expected accuracies described in the literature. Additionally, by applying quickPIV to three data sets of the embryogenesis of Tribolium castaneum, we obtained vector fields that recapitulate the migration movements of gastrulation, both in nuclear and actin-labeled embryos. We show normalized squared error cross-correlation to be especially accurate in detecting translations in non-segmentable biological image data. CONCLUSIONS: The presented software addresses the need for a fast and open-source 3D PIV package in biological research. Currently, quickPIV offers efficient 2D and 3D PIV analyses featuring zero-normalized and normalized squared error cross-correlations, sub-pixel/voxel approximation, and multi-pass. Post-processing options include filtering and averaging of the resulting vector fields, extraction of velocity, divergence and collectiveness maps, simulation of pseudo-trajectories, and unit conversion. In addition, our software includes functions to visualize the 3D vector fields in Paraview.

BACKGROUND: Exogenous cDNA introduced into an experimental system, either intentionally or accidentally, can appear as added read coverage over that gene in next-generation sequencing libraries derived from this system. If not properly recognized and managed, this cross-contamination with exogenous signal can lead to incorrect interpretation of research results. Yet, this problem is not routinely addressed in current sequence processing pipelines. RESULTS: We present cDNA-detector, a computational tool to identify and remove exogenous cDNA contamination in DNA sequencing experiments. We demonstrate that cDNA-detector can identify cDNAs quickly and accurately from alignment files. A source inference step attempts to separate endogenous cDNAs (retrocopied genes) from potential cloned, exogenous cDNAs. cDNA-detector provides a mechanism to decontaminate the alignment from detected cDNAs. Simulation studies show that cDNA-detector is highly sensitive and specific, outperforming existing tools. We apply cDNA-detector to several highly-cited public databases (TCGA, ENCODE, NCBI SRA) and show that contaminant genes appear in sequencing experiments where they lead to incorrect coverage peak calls. CONCLUSIONS: cDNA-detector is a user-friendly and accurate tool to detect and remove cDNA detection in NGS libraries. This two-step design reduces the risk of true variant removal since it allows for manual review of candidates. We find that contamination with intentionally and accidentally introduced cDNAs is an underappreciated problem even in widely-used consortium datasets, where it can lead to spurious results. Our findings highlight the importance of sensitive detection and removal of contaminant cDNA from NGS libraries before downstream analysis.

BACKGROUND: The interpretation of results from transcriptome profiling experiments via RNA sequencing (RNA-seq) can be a complex task, where the essential information is distributed among different tabular and list formats-normalized expression values, results from differential expression analysis, and results from functional enrichment analyses. A number of tools and databases are widely used for the purpose of identification of relevant functional patterns, yet often their contextualization within the data and results at hand is not straightforward, especially if these analytic components are not combined together efficiently. RESULTS: We developed the GeneTonic software package, which serves as a comprehensive toolkit for streamlining the interpretation of functional enrichment analyses, by fully leveraging the information of expression values in a differential expression context. GeneTonic is implemented in R and Shiny, leveraging packages that enable HTML-based interactive visualizations for executing drilldown tasks seamlessly, viewing the data at a level of increased detail. GeneTonic is integrated with the core classes of existing Bioconductor workflows, and can accept the output of many widely used tools for pathway analysis, making this approach applicable to a wide range of use cases. Users can effectively navigate interlinked components (otherwise available as flat text or spreadsheet tables), bookmark features of interest during the exploration sessions, and obtain at the end a tailored HTML report, thus combining the benefits of both interactivity and reproducibility. CONCLUSION: GeneTonic is distributed as an R package in the Bioconductor project ( https://bioconductor.org/packages/GeneTonic/ ) under the MIT license. Offering both bird's-eye views of the components of transcriptome data analysis and the detailed inspection of single genes, individual signatures, and their relationships, GeneTonic aims at simplifying the process of interpretation of complex and compelling RNA-seq datasets for many researchers with different expertise profiles.

BACKGROUND: Biomolecular interactions that modulate biological processes occur mainly in cavities throughout the surface of biomolecular structures. In the data science era, structural biology has benefited from the increasing availability of biostructural data due to advances in structural determination and computational methods. In this scenario, data-intensive cavity analysis demands efficient scripting routines built on easily manipulated data structures. To fulfill this need, we developed pyKVFinder, a Python package to detect and characterize cavities in biomolecular structures for data science and automated pipelines. RESULTS: pyKVFinder efficiently detects cavities in biomolecular structures and computes their volume, area, depth and hydropathy, storing these cavity properties in NumPy arrays. Benefited from Python ecosystem interoperability and data structures, pyKVFinder can be integrated with third-party scientific packages and libraries for mathematical calculations, machine learning and 3D visualization in automated workflows. As proof of pyKVFinder's capabilities, we successfully identified and compared ADRP substrate-binding site of SARS-CoV-2 and a set of homologous proteins with pyKVFinder, showing its integrability with data science packages such as matplotlib, NGL Viewer, SciPy and Jupyter notebook. CONCLUSIONS: We introduce an efficient, highly versatile and easily integrable software for detecting and characterizing biomolecular cavities in data science applications and automated protocols. pyKVFinder facilitates biostructural data analysis with scripting routines in the Python ecosystem and can be building blocks for data science and drug design applications.

Genome-wide association studies have identified thousands of genetic susceptibility loci associated with cancer as well as other traits and diseases. Mapping germline variation in identified genetic susceptibility regions to alterations in nearby gene expression nominates candidate genes potentially related to disease risk for further functional investigation. We developed LDexpress as an online resource that integrates population-specific linkage disequilibrium data from the 1000 Genomes (1000G) project and tissue-specific expression data from the Genotype-Tissue Expression project to better study regional germline variation impacting gene expression. LDexpress is a publicly available web tool designed to be easy to use, flexible to conduct a wide range of variant queries, and quick to efficiently investigate dozens of query variants across multiple tissue types. We demonstrate the utility of LDexpress using example genomic queries and anticipate this tool will accelerate understanding of disease etiology by uncovering associations of regional germline variation to nearby gene expression.

BACKGROUND: Long non-coding RNAs (lncRNAs) play significant roles in varieties of physiological and pathological processes.The premise of the lncRNA functional study is that the lncRNAs are identified correctly. Recently, deep learning method like convolutional neural network (CNN) has been successfully applied to identify the lncRNAs. However, the traditional CNN considers little relationships among samples via an indirect way. RESULTS: Inspired by the Siamese Neural Network (SNN), here we propose a novel network named Class Similarity Network in coding RNA and lncRNA classification. Class Similarity Network considers more relationships among input samples in a direct way. It focuses on exploring the potential relationships between input samples and samples from both the same class and the different classes. To achieve this, Class Similarity Network trains the parameters specific to each class to obtain the high-level features and represents the general similarity to each class in a node. The comparison results on the validation dataset under the same conditions illustrate the superiority of our Class Similarity Network to the baseline CNN. Besides, our method performs effectively and achieves state-of-the-art performances on two test datasets. CONCLUSIONS: We construct Class Similarity Network in coding RNA and lncRNA classification, which is shown to work effectively on two different datasets by achieving accuracy, precision, and F1-score as 98.43%, 0.9247, 0.9374, and 97.54%, 0.9990, 0.9860, respectively.

BACKGROUND: Upon environmental stimuli, ribosomes are surmised to undergo compositional rearrangements due to abundance changes among proteins assembled into the complex, leading to modulated structural and functional characteristics. Here, we present the ComplexOme-Structural Network Interpreter ([Formula: see text]), a computational method to allow testing whether ribosomal proteins (rProteins) that exhibit abundance changes under specific conditions are spatially confined to particular regions within the large ribosomal complex. RESULTS: [Formula: see text] translates experimentally determined structures into graphs, with nodes representing proteins and edges the spatial proximity between them. In its first implementation, [Formula: see text] considers rProteins and ignores rRNA and other objects. Spatial regions are defined using a random walk with restart methodology, followed by a procedure to obtain a minimum set of regions that cover all proteins in the complex. Structural coherence is achieved by applying weights to the edges reflecting the physical proximity between purportedly contacting proteins. The weighting probabilistically guides the random-walk path trajectory. Parameter tuning during region selection provides the option to tailor the method to specific biological questions by yielding regions of different sizes with minimum overlaps. In addition, other graph community detection algorithms may be used for the [Formula: see text] workflow, considering that they yield different sized, non-overlapping regions. All tested algorithms result in the same node kernels under equivalent regions. Based on the defined regions, available abundance change information of proteins is mapped onto the graph and subsequently tested for enrichment in any of the defined spatial regions. We applied [Formula: see text] to the cytosolic ribosome structures of Saccharomyces cerevisiae, Oryctolagus cuniculus, and Triticum aestivum using datasets with available quantitative protein abundance change information. We found that in yeast, substoichiometric rProteins depleted from translating polysomes are significantly constrained to a ribosomal region close to the tRNA entry and exit sites. CONCLUSIONS: [Formula: see text] offers a computational method to partition multi-protein complexes into structural regions and a statistical approach to test for spatial enrichments of any given subsets of proteins. [Formula: see text] is applicable to any multi-protein complex given appropriate structural and abundance-change data. [Formula: see text] is publicly available as a GitHub repository https://github.com/MSeidelFed/COSNet_i and can be installed using the python installer pip.

BACKGROUND: Advances in sequencing technology have drastically reduced sequencing costs. As a result, the amount of sequencing data increases explosively. Since FASTQ files (standard sequencing data formats) are huge, there is a need for efficient compression of FASTQ files, especially quality scores. Several quality scores compression algorithms are recently proposed, mainly focused on lossy compression to boost the compression rate further. However, for clinical applications and archiving purposes, lossy compression cannot replace lossless compression. One of the main challenges for lossless compression is time complexity, where it takes thousands of seconds to compress a 1 GB file. Also, there are desired features for compression algorithms, such as random access. Therefore, there is a need for a fast lossless compressor with a reasonable compression rate and random access functionality. RESULTS: This paper proposes a Fast and Concurrent Lossless Quality scores Compressor (FCLQC) that supports random access and achieves a lower running time based on concurrent programming. Experimental results reveal that FCLQC is significantly faster than the baseline compressors on compression and decompression at the expense of compression ratio. Compared to LCQS (baseline quality score compression algorithm), FCLQC shows at least 31x compression speed improvement in all settings, where a performance degradation in compression ratio is up to 13.58% (8.26% on average). Compared to general-purpose compressors (such as 7-zip), FCLQC shows 3x faster compression speed while having better compression ratios, at least 2.08% (4.69% on average). Moreover, the speed of random access decompression also outperforms the others. The concurrency of FCLQC is implemented using Rust; the performance gain increases near-linearly with the number of threads. CONCLUSION: The superiority of compression and decompression speed makes FCLQC a practical lossless quality score compressor candidate for speed-sensitive applications of DNA sequencing data. FCLQC is available at https://github.com/Minhyeok01/FCLQC and is freely available for non-commercial usage.

BACKGROUND: Structural variations (SVs) occupy a prominent position in human genetic diversity, and deletions form an important type of SV that has been suggested to be associated with genetic diseases. Although various deletion calling methods based on long reads have been proposed, a new approach is still needed to mine features in long-read alignment information. Recently, deep learning has attracted much attention in genome analysis, and it is a promising technique for calling SVs. RESULTS: In this paper, we propose BreakNet, a deep learning method that detects deletions by using long reads. BreakNet first extracts feature matrices from long-read alignments. Second, it uses a time-distributed convolutional neural network (CNN) to integrate and map the feature matrices to feature vectors. Third, BreakNet employs a bidirectional long short-term memory (BLSTM) model to analyse the produced set of continuous feature vectors in both the forward and backward directions. Finally, a classification module determines whether a region refers to a deletion. On real long-read sequencing datasets, we demonstrate that BreakNet outperforms Sniffles, SVIM and cuteSV in terms of their F1 scores. The source code for the proposed method is available from GitHub at https://github.com/luojunwei/BreakNet . CONCLUSIONS: Our work shows that deep learning can be combined with long reads to call deletions more effectively than existing methods.

BACKGROUND: Existing computational methods for studying miRNA regulation are mostly based on bulk miRNA and mRNA expression data. However, bulk data only allows the analysis of miRNA regulation regarding a group of cells, rather than the miRNA regulation unique to individual cells. Recent advance in single-cell miRNA-mRNA co-sequencing technology has opened a way for investigating miRNA regulation at single-cell level. However, as currently single-cell miRNA-mRNA co-sequencing data is just emerging and only available at small-scale, there is a strong need of novel methods to exploit existing single-cell data for the study of cell-specific miRNA regulation. RESULTS: In this work, we propose a new method, CSmiR (Cell-Specific miRNA regulation) to combine single-cell miRNA-mRNA co-sequencing data and putative miRNA-mRNA binding information to identify miRNA regulatory networks at the resolution of individual cells. We apply CSmiR to the miRNA-mRNA co-sequencing data in 19 K562 single-cells to identify cell-specific miRNA-mRNA regulatory networks for understanding miRNA regulation in each K562 single-cell. By analyzing the obtained cell-specific miRNA-mRNA regulatory networks, we observe that the miRNA regulation in each K562 single-cell is unique. Moreover, we conduct detailed analysis on the cell-specific miRNA regulation associated with the miR-17/92 family as a case study. The comparison results indicate that CSmiR is effective in predicting cell-specific miRNA targets. Finally, through exploring cell-cell similarity matrix characterized by cell-specific miRNA regulation, CSmiR provides a novel strategy for clustering single-cells and helps to understand cell-cell crosstalk. CONCLUSIONS: To the best of our knowledge, CSmiR is the first method to explore miRNA regulation at a single-cell resolution level, and we believe that it can be a useful method to enhance the understanding of cell-specific miRNA regulation.

BACKGROUND: An increasing number of studies now produce multiple omics measurements that require using sophisticated computational methods for analysis. While each omics data can be examined separately, jointly integrating multiple omics data allows for deeper understanding and insights to be gained from the study. In particular, data integration can be performed horizontally, where biological entities from multiple omics measurements are mapped to common reactions and pathways. However, data integration remains a challenge due to the complexity of the data and the difficulty in interpreting analysis results. RESULTS: Here we present GraphOmics, a user-friendly platform to explore and integrate multiple omics datasets and support hypothesis generation. Users can upload transcriptomics, proteomics and metabolomics data to GraphOmics. Relevant entities are connected based on their biochemical relationships, and mapped to reactions and pathways from Reactome. From the Data Browser in GraphOmics, mapped entities and pathways can be ranked, sorted and filtered according to their statistical significance (p values) and fold changes. Context-sensitive panels provide information on the currently selected entities, while interactive heatmaps and clustering functionalities are also available. As a case study, we demonstrated how GraphOmics was used to interactively explore multi-omics data and support hypothesis generation using two complex datasets from existing Zebrafish regeneration and Covid-19 human studies. CONCLUSIONS: GraphOmics is fully open-sourced and freely accessible from https://graphomics.glasgowcompbio.org/ . It can be used to integrate multiple omics data horizontally by mapping entities across omics to reactions and pathways. Our demonstration showed that by using interactive explorations from GraphOmics, interesting insights and biological hypotheses could be rapidly revealed.

BACKGROUND: In population genomics, polymorphisms that are highly differentiated between geographically separated populations are often suggestive of Darwinian positive selection. Genomic scans have highlighted several such regions in African and non-African populations, but only a handful of these have functional data that clearly associates candidate variations driving the selection process. Fine-Mapping of Adaptive Variation (FineMAV) was developed to address this in a high-throughput manner using population based whole-genome sequences generated by the 1000 Genomes Project. It pinpoints positively selected genetic variants in sequencing data by prioritizing high frequency, population-specific and functional derived alleles. RESULTS: We developed a stand-alone software that implements the FineMAV statistic. To graphically visualise the FineMAV scores, it outputs the statistics as bigWig files, which is a common file format supported by many genome browsers. It is available as a command-line and graphical user interface. The software was tested by replicating the FineMAV scores obtained using 1000 Genomes Project African, European, East and South Asian populations and subsequently applied to whole-genome sequencing datasets from Singapore and China to highlight population specific variants that can be subsequently modelled. The software tool is publicly available at https://github.com/fadilla-wahyudi/finemav . CONCLUSIONS: The software tool described here determines genome-wide FineMAV scores, using low or high-coverage whole-genome sequencing datasets, that can be used to prioritize a list of population specific, highly differentiated candidate variants for in vitro or in vivo functional screens. The tool displays these scores on the human genome browsers for easy visualisation, annotation and comparison between different genomic regions in worldwide human populations.

BACKGROUND: At present, the bioinformatics research on the relationship between aging-related diseases and genes is mainly through the establishment of a machine learning multi-label model to classify each gene. Most of the existing methods for predicting pathogenic genes mainly rely on specific types of gene features, or directly encode multiple features with different dimensions, use the same encoder to concatenate and predict the final results, which will be subject to many limitations in the applicability of the algorithm. Possible shortcomings of the above include: incomplete coverage of gene features by a single type of biomics data, overfitting of small dimensional datasets by a single encoder, or underfitting of larger dimensional datasets. METHODS: We use the known gene disease association data and gene descriptors, such as gene ontology terms (GO), protein interaction data (PPI), PathDIP, Kyoto Encyclopedia of genes and genomes Genes (KEGG), etc, as input for deep learning to predict the association between genes and diseases. Our innovation is to use Mashup algorithm to reduce the dimensionality of PPI, GO and other large biological networks, and add new pathway data in KEGG database, and then combine a variety of biological information sources through modular Deep Neural Network (DNN) to predict the genes related to aging diseases. RESULT AND CONCLUSION: The results show that our algorithm is more effective than the standard neural network algorithm (the Area Under the ROC curve from 0.8795 to 0.9153), gradient enhanced tree classifier and logistic regression classifier. In this paper, we firstly use DNN to learn the similar genes associated with the known diseases from the complex multi-dimensional feature space, and then provide the evidence that the assumed genes are associated with a certain disease.

BACKGROUND: The volume of biomedical literature and clinical data is growing at an exponential rate. Therefore, efficient access to data described in unstructured biomedical texts is a crucial task for the biomedical industry and research. Named Entity Recognition (NER) is the first step for information and knowledge acquisition when we deal with unstructured texts. Recent NER approaches use contextualized word representations as input for a downstream classification task. However, distributed word vectors (embeddings) are very limited in Spanish and even more for the biomedical domain. METHODS: In this work, we develop several biomedical Spanish word representations, and we introduce two Deep Learning approaches for pharmaceutical, chemical, and other biomedical entities recognition in Spanish clinical case texts and biomedical texts, one based on a Bi-STM-CRF model and the other on a BERT-based architecture. RESULTS: Several Spanish biomedical embeddigns together with the two deep learning models were evaluated on the PharmaCoNER and CORD-19 datasets. The PharmaCoNER dataset is composed of a set of Spanish clinical cases annotated with drugs, chemical compounds and pharmacological substances; our extended Bi-LSTM-CRF model obtains an F-score of 85.24% on entity identification and classification and the BERT model obtains an F-score of 88.80% . For the entity normalization task, the extended Bi-LSTM-CRF model achieves an F-score of 72.85% and the BERT model achieves 79.97%. The CORD-19 dataset consists of scholarly articles written in English annotated with biomedical concepts such as disorder, species, chemical or drugs, gene and protein, enzyme and anatomy. Bi-LSTM-CRF model and BERT model obtain an F-measure of 78.23% and 78.86% on entity identification and classification, respectively on the CORD-19 dataset. CONCLUSION: These results prove that deep learning models with in-domain knowledge learned from large-scale datasets highly improve named entity recognition performance. Moreover, contextualized representations help to understand complexities and ambiguity inherent to biomedical texts. Embeddings based on word, concepts, senses, etc. other than those for English are required to improve NER tasks in other languages.

BACKGROUND: Biomedical named entity recognition (NER) is a fundamental task of biomedical text mining that finds the boundaries of entity mentions in biomedical text and determines their entity type. To accelerate the development of biomedical NER techniques in Spanish, the PharmaCoNER organizers launched a competition to recognize pharmacological substances, compounds, and proteins. Biomedical NER is usually recognized as a sequence labeling task, and almost all state-of-the-art sequence labeling methods ignore the meaning of different entity types. In this paper, we investigate some methods to introduce the meaning of entity types in deep learning methods for biomedical NER and apply them to the PharmaCoNER 2019 challenge. The meaning of each entity type is represented by its definition information. MATERIAL AND METHOD: We investigate how to use entity definition information in the following two methods: (1) SQuad-style machine reading comprehension (MRC) methods that treat entity definition information as query and biomedical text as context and predict answer spans as entities. (2) Span-level one-pass (SOne) methods that predict entity spans of one type by one type and introduce entity type meaning, which is represented by entity definition information. All models are trained and tested on the PharmaCoNER 2019 corpus, and their performance is evaluated by strict micro-average precision, recall, and F1-score. RESULTS: Entity definition information brings improvements to both SQuad-style MRC and SOne methods by about 0.003 in micro-averaged F1-score. The SQuad-style MRC model using entity definition information as query achieves the best performance with a micro-averaged precision of 0.9225, a recall of 0.9050, and an F1-score of 0.9137, respectively. It outperforms the best model of the PharmaCoNER 2019 challenge by 0.0032 in F1-score. Compared with the state-of-the-art model without using manually-crafted features, our model obtains a 1% improvement in F1-score, which is significant. These results indicate that entity definition information is useful for deep learning methods on biomedical NER. CONCLUSION: Our entity definition information enhanced models achieve the state-of-the-art micro-average F1 score of 0.9137, which implies that entity definition information has a positive impact on biomedical NER detection. In the future, we will explore more entity definition information from knowledge graph.

BACKGROUND: Natural language processing (NLP) and text mining technologies for the extraction and indexing of chemical and drug entities are key to improving the access and integration of information from unstructured data such as biomedical literature. METHODS: In this paper we evaluate two important tasks in NLP: the named entity recognition (NER) and Entity indexing using the SNOMED-CT terminology. For this purpose, we propose a combination of word embeddings in order to improve the results obtained in the PharmaCoNER challenge. RESULTS: For the NER task we present a neural network composed of BiLSTM with a CRF sequential layer where different word embeddings are combined as an input to the architecture. A hybrid method combining supervised and unsupervised models is used for the concept indexing task. In the supervised model, we use the training set to find previously trained concepts, and the unsupervised model is based on a 6-step architecture. This architecture uses a dictionary of synonyms and the Levenshtein distance to assign the correct SNOMED-CT code. CONCLUSION: On the one hand, the combination of word embeddings helps to improve the recognition of chemicals and drugs in the biomedical literature. We achieved results of 91.41% for precision, 90.14% for recall, and 90.77% for F1-score using micro-averaging. On the other hand, our indexing system achieves a 92.67% F1-score, 92.44% for recall, and 92.91% for precision. With these results in a final ranking, we would be in the first position.

BACKGROUND: The recognition of pharmacological substances, compounds and proteins is essential for biomedical relation extraction, knowledge graph construction, drug discovery, as well as medical question answering. Although considerable efforts have been made to recognize biomedical entities in English texts, to date, only few limited attempts were made to recognize them from biomedical texts in other languages. PharmaCoNER is a named entity recognition challenge to recognize pharmacological entities from Spanish texts. Because there are currently abundant resources in the field of natural language processing, how to leverage these resources to the PharmaCoNER challenge is a meaningful study. METHODS: Inspired by the success of deep learning with language models, we compare and explore various representative BERT models to promote the development of the PharmaCoNER task. RESULTS: The experimental results show that deep learning with language models can effectively improve model performance on the PharmaCoNER dataset. Our method achieves state-of-the-art performance on the PharmaCoNER dataset, with a max F1-score of 92.01%. CONCLUSION: For the BERT models on the PharmaCoNER dataset, biomedical domain knowledge has a greater impact on model performance than the native language (i.e., Spanish). The BERT models can obtain competitive performance by using WordPiece to alleviate the out of vocabulary limitation. The performance on the BERT model can be further improved by constructing a specific vocabulary based on domain knowledge. Moreover, the character case also has a certain impact on model performance.

BACKGROUND: Automated assignment of specific ontology concepts to mentions in text is a critical task in biomedical natural language processing, and the subject of many open shared tasks. Although the current state of the art involves the use of neural network language models as a post-processing step, the very large number of ontology classes to be recognized and the limited amount of gold-standard training data has impeded the creation of end-to-end systems based entirely on machine learning. Recently, Hailu et al. recast the concept recognition problem as a type of machine translation and demonstrated that sequence-to-sequence machine learning models have the potential to outperform multi-class classification approaches. METHODS: We systematically characterize the factors that contribute to the accuracy and efficiency of several approaches to sequence-to-sequence machine learning through extensive studies of alternative methods and hyperparameter selections. We not only identify the best-performing systems and parameters across a wide variety of ontologies but also provide insights into the widely varying resource requirements and hyperparameter robustness of alternative approaches. Analysis of the strengths and weaknesses of such systems suggest promising avenues for future improvements as well as design choices that can increase computational efficiency with small costs in performance. RESULTS: Bidirectional encoder representations from transformers for biomedical text mining (BioBERT) for span detection along with the open-source toolkit for neural machine translation (OpenNMT) for concept normalization achieve state-of-the-art performance for most ontologies annotated in the CRAFT Corpus. This approach uses substantially fewer computational resources, including hardware, memory, and time than several alternative approaches. CONCLUSIONS: Machine translation is a promising avenue for fully machine-learning-based concept recognition that achieves state-of-the-art results on the CRAFT Corpus, evaluated via a direct comparison to previous results from the 2019 CRAFT shared task. Experiments illuminating the reasons for the surprisingly good performance of sequence-to-sequence methods targeting ontology identifiers suggest that further progress may be possible by mapping to alternative target concept representations. All code and models can be found at: https://github.com/UCDenver-ccp/Concept-Recognition-as-Translation .

BACKGROUND: Bacterial genomes are being deposited into online databases at an increasing rate. Genome annotation represents one of the first efforts to understand organisms and their diseases. Some evolutionary relationships capable of being annotated only from genomes are conserved gene neighbourhoods (CNs), phylogenetic profiles (PPs), and gene fusions. At present, there is no standalone software that enables networks of interactions among proteins to be created using these three evolutionary characteristics with efficient and effective results. RESULTS: We developed GENPPI software for the ab initio prediction of interaction networks using predicted proteins from a genome. In our case study, we employed 50 genomes of the genus Corynebacterium. Based on the PP relationship, GENPPI differentiated genomes between the ovis and equi biovars of the species Corynebacterium pseudotuberculosis and created groups among the other species analysed. If we inspected only the CN relationship, we could not entirely separate biovars, only species. Our software GENPPI was determined to be efficient because, for example, it creates interaction networks from the central genomes of 50 species/lineages with an average size of 2200 genes in less than 40 min on a conventional computer. Moreover, the interaction networks that our software creates reflect correct evolutionary relationships between species, which we confirmed with average nucleotide identity analyses. Additionally, this software enables the user to define how he or she intends to explore the PP and CN characteristics through various parameters, enabling the creation of customized interaction networks. For instance, users can set parameters regarding the genus, metagenome, or pangenome. In addition to the parameterization of GENPPI, it is also the user's choice regarding which set of genomes they are going to study. CONCLUSIONS: GENPPI can help fill the gap concerning the considerable number of novel genomes assembled monthly and our ability to process interaction networks considering the noncore genes for all completed genome versions. With GENPPI, a user dictates how many and how evolutionarily correlated the genomes answer a scientific query.

BACKGROUND: Supervised classification methods have been used for many years for feature selection in metabolomics and other omics studies. We developed a novel primal-dual based classification method (PD-CR) that can perform classification with rejection and feature selection on high dimensional datasets. PD-CR projects data onto a low dimension space and performs classification by minimizing an appropriate quadratic cost. It simultaneously optimizes the selected features and the prediction accuracy with a new tailored, constrained primal-dual method. The primal-dual framework is general enough to encompass various robust losses and to allow for convergence analysis. Here, we compare PD-CR to three commonly used methods: partial least squares discriminant analysis (PLS-DA), random forests and support vector machines (SVM). We analyzed two metabolomics datasets: one urinary metabolomics dataset concerning lung cancer patients and healthy controls; and a metabolomics dataset obtained from frozen glial tumor samples with mutated isocitrate dehydrogenase (IDH) or wild-type IDH. RESULTS: PD-CR was more accurate than PLS-DA, Random Forests and SVM for classification using the 2 metabolomics datasets. It also selected biologically relevant metabolites. PD-CR has the advantage of providing a confidence score for each prediction, which can be used to perform classification with rejection. This substantially reduces the False Discovery Rate. CONCLUSION: PD-CR is an accurate method for classification of metabolomics datasets which can outperform PLS-DA, Random Forests and SVM while selecting biologically relevant features. Furthermore the confidence score provided with PD-CR can be used to perform classification with rejection and reduce the false discovery rate.

BACKGROUND: Affordable high-throughput DNA and RNA sequencing technologies are allowing genomic analysis of plant and animal populations and as a result empowering new systems genetics approaches to study complex traits. The availability of intuitive tools to browse and analyze the resulting large-scale genetic and genomic datasets remain a significant challenge. Furthermore, these integrative genomics approaches require innovative methods to dissect the flow and interconnectedness of biological information underlying complex trait variation. The Plant Genome Integrative Explorer (PlantGenIE.org) is a multi-species database and domain that houses online tools for model and woody plant species including Eucalyptus. Since the Eucalyptus Genome Integrative Explorer (EucGenIE) is integrated within PlantGenIE, it shares genome and expression analysis tools previously implemented within the various subdomains (ConGenIE, PopGenIE and AtGenIE). Despite the success in setting up integrative genomics databases, online tools for systems genetics modelling and high-resolution dissection of complex trait variation in plant populations have been lacking. RESULTS: We have developed qtlXplorer ( https://eucgenie.org/QTLXplorer ) for visualizing and exploring systems genetics data from genome-wide association studies including quantitative trait loci (QTLs) and expression-based QTL (eQTL) associations. This module allows users to, for example, find co-located QTLs and eQTLs using an interactive version of Circos, or explore underlying genes using JBrowse. It provides users with a means to build systems genetics models and generate hypotheses from large-scale population genomics data. We also substantially upgraded the EucGenIE resource and show how it enables users to combine genomics and systems genetics approaches to discover candidate genes involved in biotic stress responses and wood formation by focusing on two multigene families, laccases and peroxidases. CONCLUSIONS: qtlXplorer adds a new dimension, population genomics, to the EucGenIE and PlantGenIE environment. The resource will be of interest to researchers and molecular breeders working in Eucalyptus and other woody plant species. It provides an example of how systems genetics data can be integrated with functional genetics data to provide biological insight and formulate hypotheses. Importantly, integration within PlantGenIE enables novel comparative genomics analyses to be performed from population-scale data.

BACKGROUND: In bacteria, genes with related functions-such as those involved in the metabolism of the same compound or in infection processes-are often physically close on the genome and form groups called clusters. The enrichment of such clusters over various distantly related bacteria can be used to predict the roles of genes of unknown function that cluster with characterised genes. There is no obvious rule to define a cluster, given their variability in size and intergenic distances, and the definition of what comprises a "gene", since genes can gain and lose domains over time. Protein domains can cluster within a gene, or in adjacent genes of related function, and in both cases these are chromosomally clustered. Here, we model the distances between pairs of protein domain coding regions across a wide range of bacteria and archaea via a probabilistic two component mixture model, without imposing arbitrary thresholds in terms of gene numbers or distances. RESULTS: We trained our model using matched gene ontology terms to label functionally related pairs and assess the stability of the parameters of the model across 14,178 archaeal and bacterial strains. We found that the parameters of our mixture model are remarkably stable across bacteria and archaea, except for endosymbionts and obligate intracellular pathogens. Obligate pathogens have smaller genomes, and although they vary, on average do not show noticeably different clustering distances; the main difference in the parameter estimates is that a far greater proportion of the genes sharing ontology terms are clustered. This may reflect that these genomes are enriched for complexes encoded by clustered core housekeeping genes, as a proportion of the total genes. Given the overall stability of the parameter estimates, we then used the mean parameter estimates across the entire dataset to investigate which gene ontology terms are most frequently associated with clustered genes. CONCLUSIONS: Given the stability of the mixture model across species, it may be used to predict bacterial gene clusters that are shared across multiple species, in addition to giving insights into the evolutionary pressures on the chromosomal locations of genes in different species.

BACKGROUND: Next-generation sequencing platforms allow us to sequence millions of small fragments of DNA simultaneously, revolutionizing cancer research. Sequence analysis has revealed that cancer driver genes operate across multiple intricate pathways and networks with mutations often occurring in a mutually exclusive pattern. Currently, low-frequency mutations are understudied as cancer-relevant genes, especially in the context of networks. RESULTS: Here we describe a tool, gcMECM, that enables us to visualize the functionality of mutually exclusive genes in the subnetworks derived from mutation associations, gene-gene interactions, and graph clustering. These subnetworks have revealed crucial biological components in the canonical pathway, especially those mutated at low frequency. Examining the subnetwork, and not just the impact of a single gene, significantly increases the statistical power of clinical analysis and enables us to build models to better predict how and why cancer develops. CONCLUSIONS: gcMECM uses a computationally efficient and scalable algorithm to identify subnetworks in a canonical pathway with mutually exclusive mutation patterns and distinct biological functions.

BACKGROUND: Clinical notes are documents that contain detailed information about the health status of patients. Medical codes generally accompany them. However, the manual diagnosis is costly and error-prone. Moreover, large datasets in clinical diagnosis are susceptible to noise labels because of erroneous manual annotation. Therefore, machine learning has been utilized to perform automatic diagnoses. Previous state-of-the-art (SOTA) models used convolutional neural networks to build document representations for predicting medical codes. However, the clinical notes are usually long-tailed. Moreover, most models fail to deal with the noise during code allocation. Therefore, denoising mechanism and long-tailed classification are the keys to automated coding at scale. RESULTS: In this paper, a new joint learning model is proposed to extend our attention model for predicting medical codes from clinical notes. On the MIMIC-III-50 dataset, our model outperforms all the baselines and SOTA models in all quantitative metrics. On the MIMIC-III-full dataset, our model outperforms in the macro-F1, micro-F1, macro-AUC, and precision at eight compared to the most advanced models. In addition, after introducing the denoising mechanism, the convergence speed of the model becomes faster, and the loss of the model is reduced overall. CONCLUSIONS: The innovations of our model are threefold: firstly, the code-specific representation can be identified by adopted the self-attention mechanism and the label attention mechanism. Secondly, the performance of the long-tailed distributions can be boosted by introducing the joint learning mechanism. Thirdly, the denoising mechanism is suitable for reducing the noise effects in medical code prediction. Finally, we evaluate the effectiveness of our model on the widely-used MIMIC-III datasets and achieve new SOTA results.

BACKGROUND: More and more Cas9 variants with higher specificity are developed to avoid the off-target effect, which brings a significant volume of experimental data. Conventional machine learning performs poorly on these datasets, while the methods based on deep learning often lack interpretability, which makes researchers have to trade-off accuracy and interpretability. It is necessary to develop a method that can not only match deep learning-based methods in performance but also with good interpretability that can be comparable to conventional machine learning methods. RESULTS: To overcome these problems, we propose an intrinsically interpretable method called AttCRISPR based on deep learning to predict the on-target activity. The advantage of AttCRISPR lies in using the ensemble learning strategy to stack available encoding-based methods and embedding-based methods with strong interpretability. Comparison with the state-of-the-art methods using WT-SpCas9, eSpCas9(1.1), SpCas9-HF1 datasets, AttCRISPR can achieve an average Spearman value of 0.872, 0.867, 0.867, respectively on several public datasets, which is superior to these methods. Furthermore, benefits from two attention modules-one spatial and one temporal, AttCRISPR has good interpretability. Through these modules, we can understand the decisions made by AttCRISPR at both global and local levels without other post hoc explanations techniques. CONCLUSION: With the trained models, we reveal the preference for each position-dependent nucleotide on the sgRNA (short guide RNA) sequence in each dataset at a global level. And at a local level, we prove that the interpretability of AttCRISPR can be used to guide the researchers to design sgRNA with higher activity.

Erratum in BMC Bioinformatics. 2022 Jan 11;23(1):32.

BACKGROUND: Copy number variants (CNVs) are the gain or loss of DNA segments in the genome. Studies have shown that CNVs are linked to various disorders, including autism, intellectual disability, and schizophrenia. Consequently, the interest in studying a possible association of CNVs to specific disease traits is growing. However, due to the specific multi-dimensional characteristics of the CNVs, methods for testing the association between CNVs and the disease-related traits are still underdeveloped. We propose a novel multi-dimensional CNV kernel association test (MCKAT) in this paper. We aim to find significant associations between CNVs and disease-related traits using kernel-based methods. RESULTS: We address the multi-dimensionality in CNV characteristics. We first design a single pair CNV kernel, which contains three sub-kernels to summarize the similarity between two CNVs considering all CNV characteristics. Then, aggregate single pair CNV kernel to the whole chromosome CNV kernel, which summarizes the similarity between CNVs in two or more chromosomes. Finally, the association between the CNVs and disease-related traits is evaluated by comparing the similarity in the trait with kernel-based similarity using a score test in a random effect model. We apply MCKAT on genome-wide CNV datasets to examine the association between CNVs and disease-related traits, which demonstrates the potential usefulness the proposed method has for the CNV association tests. We compare the performance of MCKAT with CKAT, a uni-dimensional kernel method. Based on the results, MCKAT indicates stronger evidence, smaller p-value, in detecting significant associations between CNVs and disease-related traits in both rare and common CNV datasets. CONCLUSION: A multi-dimensional copy number variant kernel association test can detect statistically significant associated CNV regions with any disease-related trait. MCKAT can provide biologists with CNV hot spots at the cytogenetic band level that CNVs on them may have a significant association with disease-related traits. Using MCKAT, biologists can narrow their investigation from the whole genome, including many genes and CNVs, to more specific cytogenetic bands that MCKAT identifies. Furthermore, MCKAT can help biologists detect significantly associated CNVs with disease-related traits across a patient group instead of examining each subject's CNVs case by case.

The arterial aneurysm refers to localized dilation of blood vessel wall and is common in general population. The majority of aneurysm cases remains asymptomatic until a sudden rupture which is usually fatal and of extremely high mortality (~ 50-60%). Therefore, early diagnosis, prevention and management of aneurysm are in urgent need. Unfortunately, current understanding of disease driver genes of various aneurysm subtypes is still limited, and without appropriate biomarkers and drug targets no specialized drug has been developed for aneurysm treatment. In this research, aneurysm subtypes were analyzed based on protein-protein interaction network to better understand aneurysm pathogenesis. By measuring network-based proximity of aneurysm subtypes, we identified a relevant closest relationship between aortic aneurysm and aortic dissection. An improved random walk method was performed to prioritize candidate driver genes of each aneurysm subtype. Thereafter, transcriptomes of 6 human aneurysm subtypes were collected and differential expression genes were identified to further filter potential driver genes. Functional enrichment of above driver genes indicated a general role of ubiquitination and programmed cell death in aneurysm pathogenesis. Especially, we further observed participation of BCL-2-mediated apoptosis pathway and caspase-1 related pyroptosis in the development of cerebral aneurysm and aneurysmal subarachnoid hemorrhage in corresponding transcriptomes.