BACKGROUND: Precision medicine is a promising approach that has revolutionized disease prevention and individualized treatment. The DELFOS oracle is a model-driven genomics platform that aids clinicians in identifying relevant variations that are associated with diseases. In its previous version, the DELFOS oracle did not consider the high degree of variability of genomics data over time. However, changes in genomics data have had a profound impact on clinicians' work and pose the need for changing past, present, and future clinical actions. Therefore, our objective in this work is to consider changes in genomics data over time in the DELFOS oracle. METHODS: Our objective has been achieved through three steps. First, we studied the characteristics of each database from which the DELFOS oracle extracts data. Second, we characterized which genomics concepts of the conceptual schema that supports the DELFOS oracle change over time. Third, we updated the DELFOS Oracle so that it can manage the temporal dimension. To validate our approach, we carried out a use case to illustrate how the new version of the DELFOS oracle handles the temporal dimension. RESULTS: Three events can change genomics data, namely, the addition of a new variation, the addition of a new link between a variation and a phenotype, and the update of a link between a variation and a phenotype. These events have been linked to the entities of the conceptual model that are affected by them. Finally, a new version of the DELFOS oracle that can deal with the temporal dimension has been implemented. CONCLUSION: Huge amounts of genomics data that is associated with diseases change over time, impacting patients' diagnosis and treatment. Including this information in the DELFOS oracle added an extra layer of complexity, but using a model-driven based approach mitigated the cost of implementing the needed changes. The new version handles the temporal dimension appropriately and eases clinicians' work.

Pre-trained natural language processing models on a large natural language corpus can naturally transfer learned knowledge to protein domains by fine-tuning specific in-domain tasks. However, few studies focused on enriching such protein language models by jointly learning protein properties from strongly-correlated protein tasks. Here we elaborately designed a multi-task learning (MTL) architecture, aiming to decipher implicit structural and evolutionary information from three sequence-level classification tasks for protein family, superfamily and fold. Considering the co-existing contextual relevance between human words and protein language, we employed BERT, pre-trained on a large natural language corpus, as our backbone to handle protein sequences. More importantly, the encoded knowledge obtained in the MTL stage can be well transferred to more fine-grained downstream tasks of TAPE. Experiments on structure- or evolution-related applications demonstrate that our approach outperforms many state-of-the-art Transformer-based protein models, especially in remote homology detection.

BACKGROUND: Studying the co-occurrence network structure of microbial samples is one of the critical approaches to understanding the perplexing and delicate relationship between the microbe, host, and diseases. It is also critical to develop a tool for investigating co-occurrence networks and differential abundance analyses to reveal the disease-related taxa-taxa relationship. In addition, it is also necessary to tighten the co-occurrence network into smaller modules to increase the ability for functional annotation and interpretability of  these taxa-taxa relationships.  Also, it is critical to retain the phylogenetic relationship among the taxa to identify differential abundance patterns, which can be used to resolve contradicting functions reported by different studies. RESULTS: In this article, we present Correlation and Consensus-based Cross-taxonomy Network Analysis (C3NA), a user-friendly R package for investigating compositional microbial sequencing data to identify and compare co-occurrence patterns across different taxonomic levels. C3NA contains two interactive graphic user interfaces (Shiny applications), one of them dedicated to the comparison between two diagnoses, e.g., disease versus control. We used C3NA to analyze two well-studied diseases, colorectal cancer, and Crohn's disease. We discovered clusters of study and disease-dependent taxa that overlap with known functional taxa studied by other discovery studies and differential abundance analyses. CONCLUSION: C3NA offers a new microbial data analyses pipeline for refined and enriched taxa-taxa co-occurrence network analyses, and the usability was further expanded via the built-in Shiny applications for interactive investigation.

Disseminated intravascular coagulation (DIC) is a complex, life-threatening syndrome associated with the end-stage of different coagulation disorders. Early prediction of the risk of DIC development is an urgent clinical need to reduce adverse outcomes. However, effective approaches and models to identify early DIC are still lacking. In this study, a novel interpretable deep learning based time series is used to predict the risk of DIC. The study cohort included ICU patients from a 4300-bed academic hospital between January 1, 2019, and January 1, 2022. Experimental results show that our model achieves excellent performance (AUC: 0.986, Accuracy: 95.7%, and F1:0.935). Gradient-weighted Class Activation Mapping (Grad-CAM) was used to explain how predictive models identified patients with DIC. The decision basis of the model was displayed in the form of a heat map. The model can be used to identify high-risk patients with DIC early, which will help in the early intervention of DIC patients and improve the treatment effect.

Early detection of cancers has been much explored due to its paramount importance in biomedical fields. Among different types of data used to answer this biological question, studies based on T cell receptors (TCRs) are under recent spotlight due to the growing appreciation of the roles of the host immunity system in tumor biology. However, the one-to-many correspondence between a patient and multiple TCR sequences hinders researchers from simply adopting classical statistical/machine learning methods. There were recent attempts to model this type of data in the context of multiple instance learning (MIL). Despite the novel application of MIL to cancer detection using TCR sequences and the demonstrated adequate performance in several tumor types, there is still room for improvement, especially for certain cancer types. Furthermore, explainable neural network models are not fully investigated for this application. In this article, we propose multiple instance neural networks based on sparse attention (MINN-SA) to enhance the performance in cancer detection and explainability. The sparse attention structure drops out uninformative instances in each bag, achieving both interpretability and better predictive performance in combination with the skip connection. Our experiments show that MINN-SA yields the highest area under the ROC curve scores on average measured across 10 different types of cancers, compared to existing MIL approaches. Moreover, we observe from the estimated attentions that MINN-SA can identify the TCRs that are specific for tumor antigens in the same T cell repertoire.

BACKGROUND: The expression changes of some proteins are associated with cancer progression, and can be used as biomarkers in cancer diagnosis. Automated systems have been frequently applied in the large-scale detection of protein biomarkers and have provided a valuable complement for wet-laboratory experiments. For example, our previous work used an immunohistochemical image-based machine learning classifier of protein subcellular locations to screen biomarker proteins that change locations in colon cancer tissues. The tool could recognize the location of biomarkers but did not consider the effect of protein expression level changes on the screening process. RESULTS: In this study, we built an automated classification model that recognizes protein expression levels in immunohistochemical images, and used the protein expression levels in combination with subcellular locations to screen cancer biomarkers. To minimize the effect of non-informative sections on the immunohistochemical images, we employed the representative image patches as input and applied a Wasserstein distance method to determine the number of patches. For the patches and the whole images, we compared the ability of color features, characteristic curve features, and deep convolutional neural network features to distinguish different levels of protein expression and employed deep learning and conventional classification models. Experimental results showed that the best classifier can achieve an accuracy of 73.72% and an F1-score of 0.6343. In the screening of protein biomarkers, the detection accuracy improved from 63.64 to 95.45% upon the incorporation of the protein expression changes. CONCLUSIONS: Machine learning can distinguish different protein expression levels and speed up their annotation in the future. Combining information on the expression patterns and subcellular locations of protein can improve the accuracy of automatic cancer biomarker screening. This work could be useful in discovering new cancer biomarkers for clinical diagnosis and research.

BACKGROUND: In most parts of the world, especially in underdeveloped countries, acquired immunodeficiency syndrome (AIDS) still remains a major cause of death, disability, and unfavorable economic outcomes. This has necessitated intensive research to develop effective therapeutic agents for the treatment of human immunodeficiency virus (HIV) infection, which is responsible for AIDS. Peptide cleavage by HIV-1 protease is an essential step in the replication of HIV-1. Thus, correct and timely prediction of the cleavage site of HIV-1 protease can significantly speed up and optimize the drug discovery process of novel HIV-1 protease inhibitors. In this work, we built and compared the performance of selected machine learning models for the prediction of HIV-1 protease cleavage site utilizing a hybrid of octapeptide sequence information comprising bond composition, amino acid binary profile (AABP), and physicochemical properties as numerical descriptors serving as input variables for some selected machine learning algorithms. Our work differs from antecedent studies exploring the same subject in the combination of octapeptide descriptors and method used. Instead of using various subsets of the dataset for training and testing the models, we combined the dataset, applied a 3-way data split, and then used a "stratified" 10-fold cross-validation technique alongside the testing set to evaluate the models. RESULTS: Among the 8 models evaluated in the "stratified" 10-fold CV experiment, logistic regression, multi-layer perceptron classifier, linear discriminant analysis, gradient boosting classifier, Naive Bayes classifier, and decision tree classifier with AUC, F-score, and B. Acc. scores in the ranges of 0.91-0.96, 0.81-0.88, and 80.1-86.4%, respectively, have the closest predictive performance to the state-of-the-art model (AUC 0.96, F-score 0.80 and B. Acc. ~ 80.0%). Whereas, the perceptron classifier and the K-nearest neighbors had statistically lower performance (AUC 0.77-0.82, F-score 0.53-0.69, and B. Acc. 60.0-68.5%) at p < 0.05. On the other hand, logistic regression, and multi-layer perceptron classifier (AUC of 0.97, F-score > 0.89, and B. Acc. > 90.0%) had the best performance on further evaluation on the testing set, though linear discriminant analysis, gradient boosting classifier, and Naive Bayes classifier equally performed well (AUC > 0.94, F-score > 0.87, and B. Acc. > 86.0%). CONCLUSIONS: Logistic regression and multi-layer perceptron classifiers have comparable predictive performances to the state-of-the-art model when octapeptide sequence descriptors consisting of AABP, bond composition and standard physicochemical properties are used as input variables. In our future work, we hope to develop a standalone software for HIV-1 protease cleavage site prediction utilizing the linear regression algorithm and the aforementioned octapeptide sequence descriptors.

BACKGROUND: In recent years, huge improvements have been made in the context of sequencing genomic data under what is called Next Generation Sequencing (NGS). However, the DNA reads generated by current NGS platforms are not free of errors, which can affect the quality of downstream analysis. Although error correction can be performed as a preprocessing step to overcome this issue, it usually requires long computational times to analyze those large datasets generated nowadays through NGS. Therefore, new software capable of scaling out on a cluster of nodes with high performance is of great importance. RESULTS: In this paper, we present SparkEC, a parallel tool capable of fixing those errors produced during the sequencing process. For this purpose, the algorithms proposed by the CloudEC tool, which is already proved to perform accurate corrections, have been analyzed and optimized to improve their performance by relying on the Apache Spark framework together with the introduction of other enhancements such as the usage of memory-efficient data structures and the avoidance of any input preprocessing. The experimental results have shown significant improvements in the computational times of SparkEC when compared to CloudEC for all the representative datasets and scenarios under evaluation, providing an average and maximum speedups of 4.9[Formula: see text] and 11.9[Formula: see text], respectively, over its counterpart. CONCLUSION: As error correction can take excessive computational time, SparkEC provides a scalable solution for correcting large datasets. Due to its distributed implementation, SparkEC speed can increase with respect to the number of nodes in a cluster. Furthermore, the software is freely available under GPLv3 license and is compatible with different operating systems (Linux, Windows and macOS).

BACKGROUND: Whole genome sequencing using the long-read Oxford Nanopore Technologies (ONT) MinION sequencer provides a cost-effective option for structural variant (SV) detection in clinical applications. Despite the advantage of using long reads, however, accurate SV calling and phasing are still challenging. RESULTS: We introduce Duet, an SV detection tool optimized for SV calling and phasing using ONT data. The tool uses novel features integrated from both SV signatures and single-nucleotide polymorphism signatures, which can accurately distinguish SV haplotype from a false signal. Duet was benchmarked against state-of-the-art tools on multiple ONT sequencing datasets of sequencing coverage ranging from 8× to 40×. At low sequencing coverage of 8×, Duet performs better than all other tools in SV calling, SV genotyping and SV phasing. When the sequencing coverage is higher (20× to 40×), the F1-score for SV phasing is further improved in comparison to the performance of other tools, while its performance of SV genotyping and SV calling remains higher than other tools. CONCLUSION: Duet can perform accurate SV calling, SV genotyping and SV phasing using low-coverage ONT data, making it very useful for low-coverage genomes. It has great performance when scaled to high-coverage genomes, which is adaptable to various clinical applications. Duet is open source and is available at https://github.com/yekaizhou/duet .

Erratum in BMC Bioinformatics. 2023 Jan 11;24(1):15.

BACKGROUND: Chromosome conformation capture and its derivatives have provided substantial genetic data for understanding how chromatin self-organizes. These techniques have identified regions of high intrasequence interactions called topologically associated domains (TADs). TADs are structural and functional units that shape chromosomes and influence genomic expression. Many of these domains differ across cell development and can be impacted by diseases. Thus, analysis of the identified domains can provide insight into genome regulation. Hence, there are many approaches to identifying such domains across many cell lines. Despite the availability of multiple tools for TAD detection, TAD callers' speed, flexibility, result inconsistency, and reproducibility remain challenges in this research area. RESULTS: In this work, we developed a computational webserver called TADMaster that provides an analysis suite to directly evaluate the concordance level and robustness of two or more TAD data on any given genome region. The suite provides multiple visual and quantitative metrics to compare the identified domains' number, size, and various comparisons of shared domains, domain boundaries, and domain overlap. CONCLUSIONS: TADMaster is an efficient and easy-to-use web application that provides a set of consensus and unique TADs to inform the choice of TADs. It can be accessed at http://tadmaster.io and is also available as a containerized application that can be deployed and run locally on any platform or operating system.

BACKGROUND: Gene expression studies are an important tool in biological and biomedical research. The signal carried in expression profiles helps derive signatures for the prediction, diagnosis and prognosis of different diseases. Data science and specifically machine learning have many applications in gene expression analysis. However, as the dimensionality of genomics datasets grows, scalable solutions become necessary. METHODS: In this paper we review the main steps and bottlenecks in machine learning pipelines, as well as the main concepts behind scalable data science including those of concurrent and parallel programming. We discuss the benefits of the Dask framework and how it can be integrated with the Python scientific environment to perform data analysis in computational biology and bioinformatics. RESULTS: This review illustrates the role of Dask for boosting data science applications in different case studies. Detailed documentation and code on these procedures is made available at https://github.com/martaccmoreno/gexp-ml-dask . CONCLUSION: By showing when and how Dask can be used in transcriptomics analysis, this review will serve as an entry point to help genomic data scientists develop more scalable data analysis procedures.

BACKGROUND: The assembly of metagenomes decomposes members of complex microbe communities and allows the characterization of these genomes without laborious cultivation or single-cell metagenomics. Metagenome assembly is a process that is memory intensive and time consuming. Multi-terabyte sequences can become too large to be assembled on a single computer node, and there is no reliable method to predict the memory requirement due to data-specific memory consumption pattern. Currently, out-of-memory (OOM) is one of the most prevalent factors that causes metagenome assembly failures. RESULTS: In this study, we explored the possibility of using Persistent Memory (PMem) as a less expensive substitute for dynamic random access memory (DRAM) to reduce OOM and increase the scalability of metagenome assemblers. We evaluated the execution time and memory usage of three popular metagenome assemblers (MetaSPAdes, MEGAHIT, and MetaHipMer2) in datasets up to one terabase. We found that PMem can enable metagenome assemblers on terabyte-sized datasets by partially or fully substituting DRAM. Depending on the configured DRAM/PMEM ratio, running metagenome assemblies with PMem can achieve a similar speed as DRAM, while in the worst case it showed a roughly two-fold slowdown. In addition, different assemblers displayed distinct memory/speed trade-offs in the same hardware/software environment. CONCLUSIONS: We demonstrated that PMem is capable of expanding the capacity of DRAM to allow larger metagenome assembly with a potential tradeoff in speed. Because PMem can be used directly without any application-specific code modification, these findings are likely to be generalized to other memory-intensive bioinformatics applications.

BACKGROUND: Genome-scale metabolic reconstruction tools have been developed in the last decades. They have helped to reconstruct eukaryotic and prokaryotic metabolic models, which have contributed to fields, e.g., genetic engineering, drug discovery, prediction of phenotypes, and other model-driven discoveries. However, the use of these programs requires a high level of bioinformatic skills. Moreover, the functionalities required to build models are scattered throughout multiple tools, requiring knowledge and experience for utilizing several tools. RESULTS: Here we present ChiMera, which combines tools used for model reconstruction, prediction, and visualization. ChiMera uses CarveMe in the reconstruction module, generating a gap-filled draft reconstruction able to produce growth predictions using flux balance analysis for gram-positive and gram-negative bacteria. ChiMera also contains two modules for metabolic network visualization. The first module generates maps for the most important pathways, e.g., glycolysis, nucleotides and amino acids biosynthesis, fatty acid oxidation and biosynthesis and core-metabolism. The second module produces a genome-wide metabolic map, which can be used to retrieve KEGG pathway information for each compound in the model. A module to investigate gene essentiality and knockout is also present. CONCLUSIONS: Overall, ChiMera uses automation algorithms to combine a variety of tools to automatically perform model creation, gap-filling, flux balance analysis (FBA), and metabolic network visualization. ChiMera models readily provide metabolic insights that can aid genetic engineering projects, prediction of phenotypes, and model-driven discoveries.

BACKGROUND: In cell signaling pathways, proteins interact with each other to determine cell fate in response to either cell-extrinsic (micro-environmental) or intrinsic cues. One of the well-studied pathways, the mitogen-activated protein kinase (MAPK) signaling pathway, regulates cell processes such as differentiation, proliferation, apoptosis, and survival in response to various micro-environmental stimuli in eukaryotes. Upon micro-environmental stimulus, receptors on the cell membrane become activated. Activated receptors initiate a cascade of protein activation in the MAPK pathway. This activation involves protein binding, creating scaffold proteins, which are known to facilitate effective MAPK signaling transduction. RESULTS: This paper presents a novel mathematical model of a cell signaling pathway coordinated by protein scaffolding. The model is based on the extended Boolean network approach with stochastic processes. Protein production or decay in a cell was modeled considering the stochastic process, whereas the protein-protein interactions were modeled based on the extended Boolean network approach. Our model fills a gap in the binary set applied to previous models. The model simultaneously considers the stochastic process directly. Using the model, we simulated a simplified mitogen-activated protein kinase (MAPK) signaling pathway upon stimulation of both a single receptor at the initial time and multiple receptors at several time points. Our simulations showed that the signal is amplified as it travels down to the pathway from the receptor, generating substantially amplified downstream ERK activity. The noise generated by the stochastic process of protein self-activity in the model was also amplified as the signaling propagated through the pathway. CONCLUSIONS: The signaling transduction in a simplified MAPK signaling pathway could be explained by a mathematical model based on the extended Boolean network model with a stochastic process. The model simulations demonstrated signaling amplifications when it travels downstream, which was already observed in experimental settings. We also highlight the importance of stochastic activity in regulating protein inactivation.

BACKGROUND: Drug-target interactions (DTIs) prediction becomes more and more important for accelerating drug research and drug repositioning. Drug-target interaction network is a typical model for DTIs prediction. As many different types of relationships exist between drug and target, drug-target interaction network can be used for modeling drug-target interaction relationship. Recent works on drug-target interaction network are mostly concentrate on drug node or target node and neglecting the relationships between drug-target. RESULTS: We propose a novel prediction method for modeling the relationship between drug and target independently. Firstly, we use different level relationships of drugs and targets to construct feature of drug-target interaction. Then, we use line graph to model drug-target interaction. After that, we introduce graph transformer network to predict drug-target interaction. CONCLUSIONS: This method introduces a line graph to model the relationship between drug and target. After transforming drug-target interactions from links to nodes, a graph transformer network is used to accomplish the task of predicting drug-target interactions.

BACKGROUND: Biomedical named entity recognition (BioNER) is a basic and important task for biomedical text mining with the purpose of automatically recognizing and classifying biomedical entities. The performance of BioNER systems directly impacts downstream applications. Recently, deep neural networks, especially pre-trained language models, have made great progress for BioNER. However, because of the lack of high-quality and large-scale annotated data and relevant external knowledge, the capability of the BioNER system remains limited. RESULTS: In this paper, we propose a novel fully-shared multi-task learning model based on the pre-trained language model in biomedical domain, namely BioBERT, with a new attention module to integrate the auto-processed syntactic information for the BioNER task. We have conducted numerous experiments on seven benchmark BioNER datasets. The proposed best multi-task model obtains F1 score improvements of 1.03% on BC2GM, 0.91% on NCBI-disease, 0.81% on Linnaeus, 1.26% on JNLPBA, 0.82% on BC5CDR-Chemical, 0.87% on BC5CDR-Disease, and 1.10% on Species-800 compared to the single-task BioBERT model. CONCLUSION: The results demonstrate our model outperforms previous studies on all datasets. Further analysis and case studies are also provided to prove the importance of the proposed attention module and fully-shared multi-task learning method used in our model.

BACKGROUND: Single-cell RNA sequencing (scRNA-seq) technology has contributed significantly to diverse research areas in biology, from cancer to development. Since scRNA-seq data is high-dimensional, a common strategy is to learn low-dimensional latent representations better to understand overall structure in the data. In this work, we build upon scVI, a powerful deep generative model which can learn biologically meaningful latent representations, but which has limited explicit control of batch effects. Rather than prioritizing batch effect removal over conservation of biological variation, or vice versa, our goal is to provide a bird's eye view of the trade-offs between these two conflicting objectives. Specifically, using the well established concept of Pareto front from economics and engineering, we seek to learn the entire trade-off curve between conservation of biological variation and removal of batch effects. RESULTS: A multi-objective optimisation technique known as Pareto multi-task learning (Pareto MTL) is used to obtain the Pareto front between conservation of biological variation and batch effect removal. Our results indicate Pareto MTL can obtain a better Pareto front than the naive scalarization approach typically encountered in the literature. In addition, we propose to measure batch effect by applying a neural-network based estimator called Mutual Information Neural Estimation (MINE) and show benefits over the more standard maximum mean discrepancy measure. CONCLUSION: The Pareto front between conservation of biological variation and batch effect removal is a valuable tool for researchers in computational biology. Our results demonstrate the efficacy of applying Pareto MTL to estimate the Pareto front in conjunction with applying MINE to measure the batch effect.

BACKGROUND: For some understudied populations, genotype data is minimal for genotype-phenotype prediction. However, we can use the data of some other large populations to learn about the disease-causing SNPs and use that knowledge for the genotype-phenotype prediction of small populations. This manuscript illustrated that transfer learning is applicable for genotype data and genotype-phenotype prediction. RESULTS: Using HAPGEN2 and PhenotypeSimulator, we generated eight phenotypes for 500 cases/500 controls (CEU, large population) and 100 cases/100 controls (YRI, small populations). We considered 5 (4 phenotypes) and 10 (4 phenotypes) different risk SNPs for each phenotype to evaluate the proposed method. The improved accuracy with transfer learning for eight different phenotypes was between 2 and 14.2 percent. The two-tailed p-value between the classification accuracies for all phenotypes without transfer learning and with transfer learning was 0.0306 for five risk SNPs phenotypes and 0.0478 for ten risk SNPs phenotypes. CONCLUSION: The proposed pipeline is used to transfer knowledge for the case/control classification of the small population. In addition, we argue that this method can also be used in the realm of endangered species and personalized medicine. If the large population data is extensive compared to small population data, expect transfer learning results to improve significantly. We show that Transfer learning is capable to create powerful models for genotype-phenotype predictions in large, well-studied populations and fine-tune these models to populations were data is sparse.

BACKGROUND: Comprehensive metabolomic analyses have been conducted in various institutes and a large amount of metabolomic data are now publicly available. To help fully exploit such data and facilitate their interpretation, metabolomic data obtained from different facilities and different samples should be integrated and compared. However, large-scale integration of such data for biological discovery is challenging given that they are obtained from various types of sample at different facilities and by different measurement techniques, and the target metabolites and sensitivities to detect them also differ from study to study. RESULTS: We developed iDMET, a network-based approach to integrate metabolomic data from different studies based on the differential metabolomic profiles between two groups, instead of the metabolite profiles themselves. As an application, we collected cancer metabolomic data from 27 previously published studies and integrated them using iDMET. A pair of metabolomic changes observed in the same disease from two studies were successfully connected in the network, and a new association between two drugs that may have similar effects on the metabolic reactions was discovered. CONCLUSIONS: We believe that iDMET is an efficient tool for integrating heterogeneous metabolomic data and discovering novel relationships between biological phenomena.

BACKGROUND: Cancers are genetically heterogeneous, so anticancer drugs show varying degrees of effectiveness on patients due to their differing genetic profiles. Knowing patient's responses to numerous cancer drugs are needed for personalized treatment for cancer. By using molecular profiles of cancer cell lines available from Cancer Cell Line Encyclopedia (CCLE) and anticancer drug responses available in the Genomics of Drug Sensitivity in Cancer (GDSC), we will build computational models to predict anticancer drug responses from molecular features. RESULTS: We propose a novel deep neural network model that integrates multi-omics data available as gene expressions, copy number variations, gene mutations, reverse phase protein array expressions, and metabolomics expressions, in order to predict cellular responses to known anti-cancer drugs. We employ a novel graph embedding layer that incorporates interactome data as prior information for prediction. Moreover, we propose a novel attention layer that effectively combines different omics features, taking their interactions into account. The network outperformed feedforward neural networks and reported 0.90 for [Formula: see text] values for prediction of drug responses from cancer cell lines data available in CCLE and GDSC. CONCLUSION: The outstanding results of our experiments demonstrate that the proposed method is capable of capturing the interactions of genes and proteins, and integrating multi-omics features effectively. Furthermore, both the results of ablation studies and the investigations of the attention layer imply that gene mutation has a greater influence on the prediction of drug responses than other omics data types. Therefore, we conclude that our approach can not only predict the anti-cancer drug response precisely but also provides insights into reaction mechanisms of cancer cell lines and drugs as well.

BACKGROUND: Functional interaction networks, where edges connect genes likely to operate in the same biological process or pathway, can be inferred from CRISPR knockout screens in cancer cell lines. Genes with similar knockout fitness profiles across a sufficiently diverse set of cell line screens are likely to be co-functional, and these "coessentiality" networks are increasingly powerful predictors of gene function and biological modularity. While several such networks have been published, most use different algorithms for each step of the network construction process. RESULTS: In this study, we identify an optimal measure of functional interaction and test all combinations of options at each step-essentiality scoring, sample variance and covariance normalization, and similarity measurement-to identify best practices for generating a functional interaction network from CRISPR knockout data. We show that Bayes Factor and Ceres scores give the best results, that Ceres outperforms the newer Chronos scoring scheme, and that covariance normalization is a critical step in network construction. We further show that Pearson correlation, mathematically identical to ordinary least squares after covariance normalization, can be extended by using partial correlation to detect and amplify signals from "moonlighting" proteins which show context-dependent interaction with different partners. CONCLUSIONS: We describe a systematic survey of methods for generating coessentiality networks from the Cancer Dependency Map data and provide a partial correlation-based approach for exploring context-dependent interactions.

BACKGROUND: Research on gene duplication is abundant and comes from a wide range of approaches, from high-throughput analyses and experimental evolution to bioinformatics and theoretical models. Notwithstanding, a consensus is still lacking regarding evolutionary mechanisms involved in evolution through gene duplication as well as the conditions that affect them. We argue that a better understanding of evolution through gene duplication requires considering explicitly that genes do not act in isolation. It demands studying how the perturbation that gene duplication implies percolates through the web of gene interactions. Due to evolution's contingent nature, the paths that lead to the final fate of duplicates must depend strongly on the early stages of gene duplication, before gene copies have accumulated distinctive changes. METHODS: Here we use a widely-known model of gene regulatory networks to study how gene duplication affects network behavior in early stages. Such networks comprise sets of genes that cross-regulate. They organize gene activity creating the gene expression patterns that give cells their phenotypic properties. We focus on how duplication affects two evolutionarily relevant properties of gene regulatory networks: mitigation of the effect of new mutations and access to new phenotypic variants through mutation. RESULTS: Among other observations, we find that those networks that are better at maintaining the original phenotype after duplication are usually also better at buffering the effect of single interaction mutations and that duplication tends to enhance further this ability. Moreover, the effect of mutations after duplication depends on both the kind of mutation and genes involved in it. We also found that those phenotypes that had easier access through mutation before duplication had higher chances of remaining accessible through new mutations after duplication. CONCLUSION: Our results support that gene duplication often mitigates the impact of new mutations and that this effect is not merely due to changes in the number of genes. The work that we put forward helps to identify conditions under which gene duplication may enhance evolvability and robustness to mutations.

Bacteria can exceptionally evolve and develop pathogenic features making it crucial to determine novel pathogenic proteins for specific therapeutic interventions. Therefore, we have developed a machine-learning tool that predicts and functionally classifies pathogenic proteins into their respective pathogenic classes. Through construction of pathogenic proteins database and optimization of ML algorithms, Support Vector Machine was selected for the model construction. The developed SVM classifier yielded an accuracy of 81.72% on the blind-dataset and classified the proteins into three classes: Non-pathogenic proteins (Class-1), Antibiotic Resistance Proteins and Toxins (Class-2), and Secretory System Associated and capsular proteins (Class-3). The classifier provided an accuracy of 79% on real dataset-1, and 72% on real dataset-2. Based on the probability of prediction, users can estimate the pathogenicity and annotation of proteins under scrutiny. Tool will provide accurate prediction of pathogenic proteins in genomic and metagenomic datasets providing leads for experimental validations. Tool is available at: http://metagenomics.iiserb.ac.in/mp4 .

BACKGROUND: Histological feature representation is advantageous for computer aided diagnosis (CAD) and disease classification when using predictive techniques based on machine learning. Explicit feature representations in computer tissue models can assist explainability of machine learning predictions. Different approaches to feature representation within digital tissue images have been proposed. Cell-graphs have been demonstrated to provide precise and general constructs that can model both low- and high-level features. The basement membrane is high-level tissue architecture, and interactions across the basement membrane are involved in multiple disease processes. Thus, the basement membrane is an important histological feature to study from a cell-graph and machine learning perspective. RESULTS: We present a two stage machine learning pipeline for generating a cell-graph from a digital H &E stained tissue image. Using a combination of convolutional neural networks for visual analysis and graph neural networks exploiting node and edge labels for topological analysis, the pipeline is shown to predict both low- and high-level histological features in oral mucosal tissue with good accuracy. CONCLUSIONS: Convolutional and graph neural networks are complementary technologies for learning, representing and predicting local and global histological features employing node and edge labels. Their combination is potentially widely applicable in histopathology image analysis and can enhance explainability in CAD tools for disease prediction.

BACKGROUND: Building biological networks with a certain function is a challenge in systems biology. For the functionality of small (less than ten nodes) biological networks, most methods are implemented by exhausting all possible network topological spaces. This exhaustive approach is difficult to scale to large-scale biological networks. And regulatory relationships are complex and often nonlinear or non-monotonic, which makes inference using linear models challenging. RESULTS: In this paper, we propose a multi-layer perceptron-based differential equation method, which operates by training a fully connected neural network (NN) to simulate the transcription rate of genes in traditional differential equations. We verify whether the regulatory network constructed by the NN method can continue to achieve the expected biological function by verifying the degree of overlap between the regulatory network discovered by NN and the regulatory network constructed by the Hill function. And we validate our approach by adapting to noise signals, regulator knockout, and constructing large-scale gene regulatory networks using link-knockout techniques. We apply a real dataset (the mesoderm inducer Xenopus Brachyury expression) to construct the core topology of the gene regulatory network and find that Xbra is only strongly expressed at moderate levels of activin signaling. CONCLUSION: We have demonstrated from the results that this method has the ability to identify the underlying network topology and functional mechanisms, and can also be applied to larger and more complex gene network topologies.

BACKGROUND: Identifying polymorphism clades on phylogenetic trees could help detect punctual mutations that are associated with viral functions. With visualization tools coloring the tree, it is easy to visually find clades where most sequences have the same polymorphism state. However, with the fast accumulation of viral sequences, a computational tool to automate this process is urgently needed. RESULTS: Here, by implementing a branch-and-bound-like search method, we developed an R package named sitePath to identify polymorphism clades automatically. Based on the identified polymorphism clades, fixed and parallel mutations could be inferred. Furthermore, sitePath also integrated visualization tools to generate figures of the calculated results. In an example with the influenza A virus H3N2 dataset, the detected fixed mutations coincide with antigenic shift mutations. The highly specificity and sensitivity of sitePath in finding fixed mutations were achieved for a range of parameters and different phylogenetic tree inference software. CONCLUSIONS: The result suggests that sitePath can identify polymorphism clades per site. The clustering of sequences on a phylogenetic tree can be used to infer fixed and parallel mutations. High-quality figures of the calculated results could also be generated by sitePath.

BACKGROUND: Multiple processes impact the probability of retention of individual genes following whole genome duplication (WGD) events. In analyzing two consecutive whole genome duplication events that occurred in the lineage leading to Atlantic salmon, a new phylogenetic statistical analysis was developed to examine the contingency of retention in one event based upon retention in a previous event. This analysis is intended to evaluate mechanisms of duplicate gene retention and to provide software to generate the test statistic for any genome with pairs of WGDs in its history. RESULTS: Here a software package written in Python, 'WGDTree' for the analysis of duplicate gene retention following whole genome duplication events is presented. Using gene tree-species tree reconciliation to label gene duplicate nodes and differentiate between WGD and SSD duplicates, the tool calculates a statistic based upon the conditional probability of a gene duplicate being retained after a second whole genome duplication dependent upon the retention status after the first event. The package also contains methods for the simulation of gene trees with WGD events. After running simulations, the accuracy of the placement of events has been determined to be high. The conditional probability statistic has been calculated for Phalaenopsis equestris on a monocot species tree with a pair of consecutive WGD events on its lineage, showing the applicability of the method. CONCLUSIONS: A new software tool has been created for the analysis of duplicate genes in examination of retention mechanisms. The software tool has been made available on the Python package index and the source code can be found on GitHub here: https://github.com/cnickh/wgdtree .

As genotype databases increase in size, so too do the number of detectable segments of identity by descent (IBD): segments of the genome where two individuals share an identical copy of one of their two parental haplotypes, due to shared ancestry. We show that given a large enough genotype database, these segments of IBD collectively overlap entire chromosomes, including instances of IBD that span multiple chromosomes, and can be used to accurately separate the alleles inherited from each parent across the entire genome. The resulting phase is not an improvement over state-of-the-art local phasing methods, but provides accurate long-range phasing that indicates which of two haplotypes in different regions of the genome, including different chromosomes, was inherited from the same parent. We are able to separate the DNA inherited from each parent completely, across the entire genome, with 98% median accuracy in a test set of 30,000 individuals. We estimate the IBD data requirements for accurate genome-wide phasing, and we propose a method for estimating confidence in the resulting phase. We show that our methods do not require the genotypes of close family, and that they are robust to genotype errors and missing data. In fact, our method can impute missing data accurately and correct genotype errors.

BACKGROUND: Detection of newly transposed events by transposable elements (TEs) from next generation sequence (NGS) data is difficult, due to their multiple distribution sites over the genome containing older TEs. The previously reported Transposon Insertion Finder (TIF) detects TE transpositions on the reference genome from NGS short reads using end sequences of target TE. TIF requires the sequence of target TE and is not able to detect transpositions for TEs with an unknown sequence. RESULT: The new algorithm Transposable Element Finder (TEF) enables the detection of TE transpositions, even for TEs with an unknown sequence. TEF is a finding tool of transposed TEs, in contrast to TIF as a detection tool of transposed sites for TEs with a known sequence. The transposition event is often accompanied with a target site duplication (TSD). Focusing on TSD, two algorithms to detect both ends of TE, TSDs and target sites are reported here. One is based on the grouping with TSDs and direct comparison of k-mers from NGS without similarity search. The other is based on the junction mapping of TE end sequence candidates. Both methods succeed to detect both ends and TSDs of known active TEs in several tests with rice, Arabidopsis and Drosophila data and discover several new TEs in new locations. PCR confirmed the detected transpositions of TEs in several test cases in rice. CONCLUSIONS: TEF detects transposed TEs with TSDs as a result of TE transposition, sequences of both ends and their inserted positions of transposed TEs by direct comparison of NGS data between two samples. Genotypes of transpositions are verified by counting of junctions of head and tail, and non-insertion sequences in NGS reads. TEF is easy to run and independent of any TE library, which makes it useful to detect insertions from unknown TEs bypassed by common TE annotation pipelines.

BACKGROUND: Automatic and accurate recognition of various biomedical named entities from literature is an important task of biomedical text mining, which is the foundation of extracting biomedical knowledge from unstructured texts into structured formats. Using the sequence labeling framework and deep neural networks to implement biomedical named entity recognition (BioNER) is a common method at present. However, the above method often underutilizes syntactic features such as dependencies and topology of sentences. Therefore, it is an urgent problem to be solved to integrate semantic and syntactic features into the BioNER model. RESULTS: In this paper, we propose a novel biomedical named entity recognition model, named BioByGANS (BioBERT/SpaCy-Graph Attention Network-Softmax), which uses a graph to model the dependencies and topology of a sentence and formulate the BioNER task as a node classification problem. This formulation can introduce more topological features of language and no longer be only concerned about the distance between words in the sequence. First, we use periods to segment sentences and spaces and symbols to segment words. Second, contextual features are encoded by BioBERT, and syntactic features such as part of speeches, dependencies and topology are preprocessed by SpaCy respectively. A graph attention network is then used to generate a fusing representation considering both the contextual features and syntactic features. Last, a softmax function is used to calculate the probabilities and get the results. We conduct experiments on 8 benchmark datasets, and our proposed model outperforms existing BioNER state-of-the-art methods on the BC2GM, JNLPBA, BC4CHEMD, BC5CDR-chem, BC5CDR-disease, NCBI-disease, Species-800, and LINNAEUS datasets, and achieves F1-scores of 85.15%, 78.16%, 92.97%, 94.74%, 87.74%, 91.57%, 75.01%, 90.99%, respectively. CONCLUSION: The experimental results on 8 biomedical benchmark datasets demonstrate the effectiveness of our model, and indicate that formulating the BioNER task into a node classification problem and combining syntactic features into the graph attention networks can significantly improve model performance.

BACKGROUND: Cancer evolution consists of a stepwise acquisition of genetic and epigenetic changes, which alter the gene expression profiles of cells in a particular tissue and result in phenotypic alterations acted upon by natural selection. The recurrent appearance of specific genetic lesions across individual cancers and cancer types suggests the existence of certain "driver mutations," which likely make up the major contribution to tumors' selective advantages over surrounding normal tissue and as such are responsible for the most consequential aspects of the cancer cells' gene expression patterns and phenotypes. We hypothesize that such mutations are likely to cluster with specific dichotomous shifts in the expression of the genes they most closely control, and propose GMMchi, a Python package that leverages Gaussian Mixture Modeling to detect and characterize bimodal gene expression patterns across cancer samples, as a tool to analyze such correlations using 2 × 2 contingency table statistics. RESULTS: Using well-defined simulated data, we were able to confirm the robust performance of GMMchi, reaching 85% accuracy with a sample size of n = 90. We were also able to demonstrate a few examples of the application of GMMchi with respect to its capacity to characterize background florescent signals in microarray data, filter out uninformative background probe sets, as well as uncover novel genetic interrelationships and tumor characteristics. Our approach to analysing gene expression analysis in cancers provides an additional lens to supplement traditional continuous-valued statistical analysis by maximizing the information that can be gathered from bulk gene expression data. CONCLUSIONS: We confirm that GMMchi robustly and reliably extracts bimodal patterns from both colorectal cancer (CRC) cell line-derived microarray and tumor-derived RNA-Seq data and verify previously reported gene expression correlates of some well-characterized CRC phenotypes. AVAILABILITY: The Python package GMMchi and our cell line microarray data used in this paper is available for downloading on GitHub at https://github.com/jeffliu6068/GMMchi .

BACKGROUND: Ligand-protein interactions play a key role in defining protein function, and detecting natural ligands for a given protein is thus a very important bioengineering task. In particular, with the rapid development of AI-based structure prediction algorithms, batch structural models with high reliability and accuracy can be obtained at low cost, giving rise to the urgent requirement for the prediction of natural ligands based on protein structures. In recent years, although several structure-based methods have been developed to predict ligand-binding pockets and ligand-binding sites, accurate and rapid methods are still lacking, especially for the prediction of ligand-binding regions and the spatial extension of ligands in the pockets. RESULTS: In this paper, we proposed a multilayer dynamics perturbation analysis (MDPA) method for predicting ligand-binding regions based solely on protein structure, which is an extended version of our previously developed fast dynamic perturbation analysis (FDPA) method. In MDPA/FDPA, ligand binding tends to occur in regions that cause large changes in protein conformational dynamics. MDPA, examined using a standard validation dataset of ligand-protein complexes, yielded an averaged ligand-binding site prediction Matthews coefficient of 0.40, with a prediction precision of at least 50% for 71% of the cases. In particular, for 80% of the cases, the predicted ligand-binding region overlaps the natural ligand by at least 50%. The method was also compared with other state-of-the-art structure-based methods. CONCLUSIONS: MDPA is a structure-based method to detect ligand-binding regions on protein surface. Our calculations suggested that a range of spaces inside the protein pockets has subtle interactions with the protein, which can significantly impact on the overall dynamics of the protein. This work provides a valuable tool as a starting point upon which further docking and analysis methods can be used for natural ligand detection in protein functional annotation. The source code of MDPA method is freely available at: https://github.com/mingdengming/mdpa .

BACKGROUND: Genome-wide association studies (GWAS) are a powerful method to detect associations between variants and phenotypes. A GWAS requires several complex computations with large data sets, and many steps may need to be repeated with varying parameters. Manual running of these analyses can be tedious, error-prone and hard to reproduce. RESULTS: The H3AGWAS workflow from the Pan-African Bioinformatics Network for H3Africa is a powerful, scalable and portable workflow implementing pre-association analysis, implementation of various association testing methods and post-association analysis of results. CONCLUSIONS: The workflow is scalable-laptop to cluster to cloud (e.g., SLURM, AWS Batch, Azure). All required software is containerised and can run under Docker or Singularity.

BACKGROUND: Genotyping and sequencing technologies produce increasingly large numbers of genetic markers with potentially high rates of missing or erroneous data. Therefore, the construction of linkage maps is more and more complex. Moreover, the size of segregating populations remains constrained by cost issues and is less and less commensurate with the numbers of SNPs available. Thus, guaranteeing a statistically robust marker order requires that maps include only a carefully selected subset of SNPs. RESULTS: In this context, the SeSAM software allows automatic genetic map construction using seriation and placement approaches, to produce (1) a high-robustness framework map which includes as many markers as possible while keeping the order robustness beyond a given statistical threshold, and (2) a high-density total map including the framework plus almost all polymorphic markers. During this process, care is taken to limit the impact of genotyping errors and of missing data on mapping quality. SeSAM can be used with a wide range of biparental populations including from outcrossing species for which phases are inferred on-the-fly by maximum-likelihood during map elongation. The package also includes functions to simulate data sets, convert data formats, detect putative genotyping errors, visualize data and map quality (including graphical genotypes), and merge several maps into a consensus. SeSAM is also suitable for interactive map construction, by providing lower-level functions for 2-point and multipoint EM analyses. The software is implemented in a R package including functions in C++. CONCLUSIONS: SeSAM is a fully automatic linkage mapping software designed to (1) produce a framework map as robust as desired by optimizing the selection of a subset of markers, and (2) produce a high-density map including almost all polymorphic markers. The software can be used with a wide range of biparental mapping populations including cases from outcrossing. SeSAM is freely available under a GNU GPL v3 license and works on Linux, Windows, and macOS platforms. It can be downloaded together with its user-manual and quick-start tutorial from ForgeMIA (SeSAM project) at https://forgemia.inra.fr/gqe-acep/sesam/-/releases.

BACKGROUND: Tandem repeats (TR), highly variable genomic variants, are widely used in individual identification, disease diagnostics, and evolutionary studies. The recent advances in sequencing technologies and bioinformatic tools facilitate calling TR haplotypes genome widely. Both length-based and sequence-based TR alleles are used in different applications. However, sequence-based TR alleles could provide the highest precision in characterizing TR haplotypes. The need to identify the differences at the single nucleotide level between or among TR haplotypes with an easy-use bioinformatic tool is essential. RESULTS: In this study, we developed a Universal STR Allele Toolkit (USAT) for TR haplotype analysis, which takes TR haplotype output from existing tools to perform allele size conversion, sequence comparison of haplotypes, figure plotting, comparison for allele distribution, and interactive visualization. An exemplary application of USAT for analysis of the CODIS core STR loci for DNA forensics with benchmarking human individuals demonstrated the capabilities of USAT. USAT has user-friendly graphic interfaces and runs fast in major computing operating systems with parallel computing enabled. CONCLUSION: USAT is a user-friendly bioinformatics software for interpretation, visualization, and comparisons of TRs.

BACKGROUND: Sequencing is the key method to study the impact of short RNAs, which include micro RNAs, tRNA-derived RNAs, and piwi-interacting RNA, among others. The first step to make use of these reads is to map them to a genome. Existing mapping tools have been developed for long RNAs in mind, and, so far, no tool has been conceived for short RNAs. However, short RNAs have several distinctive features which make them different from messenger RNAs: they are shorter, they are often redundant, they can be produced by duplicated loci, and they may be edited at their ends. RESULTS: In this work, we present a new tool, srnaMapper, that exhaustively maps these reads with all these features in mind, and is most efficient when applied to reads no longer than 50 base pairs. We show, on several datasets, that srnaMapper is very efficient considering computation time and edition error handling: it retrieves all the hits, with arbitrary number of errors, in time comparable with non-exhaustive tools.

BACKGROUND: Accurate annotation of protein function is the key to understanding life at the molecular level and has great implications for biomedicine and pharmaceuticals. The rapid developments of high-throughput technologies have generated huge amounts of protein-protein interaction (PPI) data, which prompts the emergence of computational methods to determine protein function. Plagued by errors and noises hidden in PPI data, these computational methods have undertaken to focus on the prediction of functions by integrating the topology of protein interaction networks and multi-source biological data. Despite effective improvement of these computational methods, it is still challenging to build a suitable network model for integrating multiplex biological data. RESULTS: In this paper, we constructed a heterogeneous biological network by initially integrating original protein interaction networks, protein-domain association data and protein complexes. To prove the effectiveness of the heterogeneous biological network, we applied the propagation algorithm on this network, and proposed a novel iterative model, named Propagate on Heterogeneous Biological Networks (PHN) to score and rank functions in descending order from all functional partners, Finally, we picked out top L of these predicted functions as candidates to annotate the target protein. Our comprehensive experimental results demonstrated that PHN outperformed seven other competing approaches using cross-validation. Experimental results indicated that PHN performs significantly better than competing methods and improves the Area Under the Receiver-Operating Curve (AUROC) in Biological Process (BP), Molecular Function (MF) and Cellular Components (CC) by no less than 33%, 15% and 28%, respectively. CONCLUSIONS: We demonstrated that integrating multi-source data into a heterogeneous biological network can preserve the complex relationship among multiplex biological data and improve the prediction accuracy of protein function by getting rid of the constraints of errors in PPI networks effectively. PHN, our proposed method, is effective for protein function prediction.

Classification of different cancer types is an essential step in designing a decision support model for early cancer predictions. Using various machine learning (ML) techniques with ensemble learning is one such method used for classifications. In the present study, various ML algorithms were explored on twenty exome datasets, belonging to 5 cancer types. Initially, a data clean-up was carried out on 4181 variants of cancer with 88 features, and a derivative dataset was obtained using natural language processing and probabilistic distribution. An exploratory dataset analysis using principal component analysis was then performed in 1 and 2D axes to reduce the high-dimensionality of the data. To significantly reduce the imbalance in the derivative dataset, oversampling was carried out using SMOTE. Further, classification algorithms such as K-nearest neighbour and support vector machine were used initially on the oversampled dataset. A 4-layer artificial neural network model with 1D batch normalization was also designed to improve the model accuracy. Ensemble ML techniques such as bagging along with using KNN, SVM and MLPs as base classifiers to improve the weighted average performance metrics of the model. However, due to small sample size, model improvement was challenging. Therefore, a novel method to augment the sample size using generative adversarial network (GAN) and triplet based variational auto encoder (TVAE) was employed that reconstructed the features and labels generating the data. The results showed that from initial scrutiny, KNN showed a weighted average of 0.74 and SVM 0.76. Oversampling ensured that the accuracy of the derivative dataset improved significantly and the ensemble classifier augmented the accuracy to 82.91%, when the data was divided into 70:15:15 ratio (training, test and holdout datasets). The overall evaluation metric value when GAN and TVAE increased the sample size was found to be 0.92 with an overall comparison model of 0.66. Therefore, the present study designed an effective model for classifying cancers which when implemented to real world samples, will play a major role in early cancer diagnosis.

BACKGROUND: Clinical studies show that microorganisms are closely related to human health, and the discovery of potential associations between microbes and drugs will facilitate drug research and development. However, at present, few computational methods for predicting microbe-drug associations have been proposed. RESULTS: In this work, we proposed a novel computational model named GSAMDA based on the graph attention network and sparse autoencoder to infer latent microbe-drug associations. In GSAMDA, we first built a heterogeneous network through integrating known microbe-drug associations, microbe similarities and drug similarities. And then, we adopted a GAT-based autoencoder and a sparse autoencoder module respectively to learn topological representations and attribute representations for nodes in the newly constructed heterogeneous network. Finally, based on these two kinds of node representations, we constructed two kinds of feature matrices for microbes and drugs separately, and then, utilized them to calculate possible association scores for microbe-drug pairs. CONCLUSION: A novel computational model is proposed for predicting potential microbe-drug associations based on graph attention network and sparse autoencoder. Compared with other five state-of-the-art competitive methods, the experimental results illustrated that our model can achieve better performance. Moreover, case studies on two categories of representative drugs and microbes further demonstrated the effectiveness of our model as well.

BACKGROUND: Genomics and virology are unquestionably important, but complex, domains being investigated by a large number of scientists. The need to facilitate and support work within these domains requires sharing of databases, although it is often difficult to do so because of the different ways in which data is represented across the databases. To foster semantic interoperability, models are needed that provide a deep understanding and interpretation of the concepts in a domain, so that the data can be consistently interpreted among researchers. RESULTS: In this research, we propose the use of conceptual models to support semantic interoperability among databases and assess their ontological clarity to support their effective use. This modeling effort is illustrated by its application to the Viral Conceptual Model (VCM) that captures and represents the sequencing of viruses, inspired by the need to understand the genomic aspects of the virus responsible for COVID-19. For achieving semantic clarity on the VCM, we leverage the "ontological unpacking" method, a process of ontological analysis that reveals the ontological foundation of the information that is represented in a conceptual model. This is accomplished by applying the stereotypes of the OntoUML ontology-driven conceptual modeling language.As a result, we propose a new OntoVCM, an ontologically grounded model, based on the initial VCM, but with guaranteed interoperability among the data sources that employ it. CONCLUSIONS: We propose and illustrate how the unpacking of the Viral Conceptual Model resolves several issues related to semantic interoperability, the importance of which is recognized by the "I" in FAIR principles. The research addresses conceptual uncertainty within the domain of SARS-CoV-2 data and knowledge.The method employed provides the basis for further analyses of complex models currently used in life science applications, but lacking ontological grounding, subsequently hindering the interoperability needed for scientists to progress their research.

BACKGROUND: Studies that utilize RNA Sequencing (RNA-Seq) in conjunction with designs that introduce dependence between observations (e.g. longitudinal sampling) require specialized analysis tools to accommodate this additional complexity. This R package contains a set of utilities to fit linear mixed effects models to transformed RNA-Seq counts that properly account for this dependence when performing statistical analyses. RESULTS: In a simulation study comparing lmerSeq and two existing methodologies that also work with transformed RNA-Seq counts, we found that lmerSeq was comprehensively better in terms of nominal error rate control and statistical power. CONCLUSIONS: Existing R packages for analyzing transformed RNA-Seq data with linear mixed models are limited in the variance structures they allow and/or the transformation methods they support. The lmerSeq package offers more flexibility in both of these areas and gave substantially better results in our simulations.

BACKGROUND: Current methods of high-dimensional unsupervised clustering of mass cytometry data lack means to monitor and evaluate clustering results. Whether unsupervised clustering is correct is typically evaluated by agreement with dimensionality reduction techniques or based on benchmarking with manually classified cells. The ambiguity and lack of reproducibility of sequential gating has been replaced with ambiguity in interpretation of clustering results. On the other hand, spurious overclustering of data leads to loss of statistical power. We have developed INFLECT, an R-package designed to give insight in clustering results and provide an optimal number of clusters. In our approach, a mass cytometry dataset is overclustered intentionally to ensure the smallest phenotypically different subsets are captured using FlowSOM. A range of metacluster number endpoints are generated and evaluated using marker interquartile range and distribution unimodality checks. The fraction of marker distributions that pass these checks is taken as a measure of clustering success. The fraction of unimodal distributions within metaclusters is plotted against the number of generated metaclusters and reaches a plateau of diminishing returns. The inflection point at which this occurs gives an optimal point of capturing cellular heterogeneity versus statistical power. RESULTS: We applied INFLECT to four publically available mass cytometry datasets of different size and number of markers. The unimodality score consistently reached a plateau, with an inflection point dependent on dataset size and number of dimensions. We tested both ConsenusClusterPlus metaclustering and hierarchical clustering. While hierarchical clustering is less computationally expensive and thus faster, it achieved similar results to ConsensusClusterPlus. The four datasets consisted of labeled data and we compared INFLECT metaclustering to published results. INFLECT identified a higher optimal number of metaclusters for all datasets. We illustrated the underlying heterogeneity within labels, showing that these labels encompass distinct types of cells. CONCLUSION: INFLECT addresses a knowledge gap in high-dimensional cytometry analysis, namely assessing clustering results. This is done through monitoring marker distributions for interquartile range and unimodality across a range of metacluster numbers. The inflection point is the optimal trade-off between cellular heterogeneity and statistical power, applied in this work for FlowSOM clustering on mass cytometry datasets.

BACKGROUND: Decisions in healthcare usually rely on the goodness and completeness of data that could be coupled with heuristics to improve the decision process itself. However, this is often an incomplete process. Structured interviews denominated Delphi surveys investigate experts' opinions and solve by consensus complex matters like those underlying surgical decision-making. Natural Language Processing (NLP) is a field of study that combines computer science, artificial intelligence, and linguistics. NLP can then be used as a valuable help in building a correct context in surgical data, contributing to the amelioration of surgical decision-making. RESULTS: We applied NLP coupled with machine learning approaches to predict the context (words) owning high accuracy from the words nearest to Delphi surveys, used as input. CONCLUSIONS: The proposed methodology has increased the usefulness of Delphi surveys favoring the extraction of keywords that can represent a specific clinical context. It permits the characterization of the clinical context suggesting words for the evaluation process of the data.

BACKGROUND: RNA-seq has become a standard technology to quantify mRNA. The measured values usually vary by several orders of magnitude, and while the detection of differences at high values is statistically well grounded, the significance of the differences for rare mRNAs can be weakened by the presence of biological and technical noise. RESULTS: We have developed a method for cleaning RNA-seq data, which improves the detection of differentially expressed genes and specifically genes with low to moderate transcription. Using a data modeling approach, parameters of randomly distributed mRNA counts are identified and reads, most probably originating from technical noise, are removed. We demonstrate that the removal of this random component leads to the significant increase in the number of detected differentially expressed genes, more significant pvalues and no bias towards low-count genes. CONCLUSION: Application of RNAdeNoise to our RNA-seq data on polysome profiling and several published RNA-seq datasets reveals its suitability for different organisms and sequencing technologies such as Illumina and BGI, shows improved detection of differentially expressed genes, and excludes the subjective setting of thresholds for minimal RNA counts. The program, RNA-seq data, resulted gene lists and examples of use are in the supplementary data and at https://github.com/Deyneko/RNAdeNoise .

BACKGROUND: Unstructured text in medical records, such as Electronic Health Records, contain an enormous amount of valuable information for research; however, it is difficult to extract and structure important information because of frequent typographical errors. Therefore, improving the quality of data with errors for text analysis is an essential task. To date, few prior studies have been conducted addressing this. Here, we propose a new methodology for extracting important information from unstructured medical texts by overcoming the typographical problem in surgical pathology records related to lung cancer. METHODS: We propose a typo correction model that considers context, based on the Masked Language Model, to solve the problem of typographical errors in real-world medical data. In addition, a word dictionary was used for the typo correction model based on PubMed abstracts. After refining the data through typo correction, fine tuning was performed on pre-trained BERT model. Next, deep learning-based Named Entity Recognition (NER) was performed. By solving the quality problem of medical data, we sought to improve the accuracy of information extraction in unstructured text data. RESULTS: We compared the performance of the proposed typo correction model based on contextual information with an existing SymSpell model. We confirmed that our proposed model outperformed the existing model in a typographical correction task. The F1-score of the model improved by approximately 5% and 9% when compared with the model without contextual information in the NCBI-disease and surgical pathology record datasets, respectively. In addition, the F1-score of NER after typo correction increased by 2% in the NCBI-disease dataset. There was a significant performance difference of approximately 25% between the before and after typo correction in the Surgical pathology record dataset. This confirmed that typos influenced the information extraction of the unstructured text. CONCLUSION: We verified that typographical errors in unstructured text negatively affect the performance of natural language processing tasks. The proposed method of a typo correction model outperformed the existing SymSpell model. This study shows that the proposed model is robust and can be applied in real-world environments by focusing on the typos that cause difficulties in analyzing unstructured medical text.

BACKGROUND: Identification of deleterious genetic variants using DNA sequencing data relies on increasingly detailed filtering strategies to isolate the small subset of variants that are more likely to underlie a disease phenotype. Datasets reflecting population allele frequencies of different types of variants serve as powerful filtering tools, especially in the context of rare disease analysis. While such population-scale allele frequency datasets now exist for structural variants (SVs), it remains a challenge to match SV calls between multiple datasets, thereby complicating estimates of a putative SV's population allele frequency. RESULTS: We introduce SVAFotate, a software tool that enables the annotation of SVs with variant allele frequency and related information from existing SV datasets. As a result, VCF files annotated by SVAFotate offer a variety of metrics to aid in the stratification of SVs as common or rare in the broader human population. CONCLUSIONS: Here we demonstrate the use of SVAFotate in the classification of SVs with regards to their population frequency and illustrate how SVAFotate's annotations can be used to filter and prioritize SVs. Lastly, we detail how best to utilize these SV annotations in the analysis of genetic variation in studies of rare disease.

BACKGROUND: Cuproptosis, a newly discovered mode of cell death, has been less studied in hepatocellular carcinoma (HCC). Exploring the molecular characteristics of different subtypes of HCC based on cuproptosis-related genes (CRGs) is meaningful to HCC. In addition, immunotherapy plays a pivotal role in treating HCC. Exploring the sensitivity of immunotherapy and building predictive models are critical for HCC. METHODS: The 357 HCC samples from the TCGA database were classified into three subtypes, Cluster 1, Cluster 2, and Cluster 3, based on the expression levels of ten CRGs genes using consensus clustering. Six machine learning algorithms were used to build models that identified the three subtypes. The molecular features of the three subtypes were analyzed and compared from some perspectives. Moreover, based on the differentially expressed genes (DEGs) between Cluster 1 and Cluster 3, a prognostic scoring model was constructed using LASSO regression and Cox regression, and the scoring model was used to predict the efficacy of immunotherapy in the IMvigor210 cohort. RESULTS: Cluster 3 had the worst overall survival compared to Cluster 1 and Cluster 2 (P = 0.0048). The AUC of the Catboost model used to identify Cluster 3 was 0.959. Cluster 3 was significantly different from the other two subtypes in gene mutation, tumor mutation burden, tumor microenvironment, the expression of immune checkpoint inhibitor genes and N6-methyladenosine regulatory genes, and the sensitivity to sorafenib. We believe Cluster 3 is more sensitive to immunotherapy from the above analysis results. Therefore, based on the DEGs between Cluster 1 and Cluster 3, we obtained a 7-gene scoring prognostic model, which achieved meaningful results in predicting immunotherapy efficacy in the IMvigor210 cohort (P = 0.013). CONCLUSIONS: Our study provides new ideas for molecular characterization and immunotherapy of HCC from machine learning and bioinformatics. Moreover, we successfully constructed a prognostic model of immunotherapy.

BACKGROUND: Mass screening programs for cervical cancer prevention in the Nordic countries have strongly reduced cancer incidence and mortality at the population level. An alternative to the current mass screening is a more personalised screening strategy adapting the recommendations to each individual. However, this necessitates reliable risk prediction models accounting for disease dynamics and individual data. Herein we propose a novel matrix factorisation framework to classify females by the time-varying risk of being diagnosed with cervical cancer. We cast the problem as a time-series prediction model where the data from females in the Norwegian screening population are represented as sparse vectors in time and then combined into a single matrix. Using novel temporal regularisation and discrepancy terms for the cervical cancer screening context, we reconstruct complete screening profiles from this scarce matrix and use these to predict the next exam results indicating the risk of cervical cancer. The algorithm is validated on both synthetic and registry screening data by measuring the probability of agreement (PoA) between Kaplan-Meier estimates. RESULTS: In numerical experiments on synthetic data, we demonstrate that the novel regularisation and discrepancy term can improve the data reconstruction ability as well as prediction performance over varying data scarcity. Using a hold-out set of screening data, we compare several numerical models and find that the proposed framework attains the strongest PoA. We observe strong correlations between the empirical survival curves from our method and the hold-out data, and evaluate the ability of our framework to predict the females' next results for up to five years ahead in time using only their current screening histories as input. CONCLUSIONS: We have proposed a matrix factorization model for predicting future screening results and evaluated its performance in a female cohort to demonstrate the potential for developing prediction models for more personalized cervical cancer screening.

BACKGROUND: The t-distributed Stochastic Neighbor Embedding (t-SNE) algorithm has emerged as one of the leading methods for visualising high-dimensional (HD) data in a wide variety of fields, especially for revealing cluster structure in HD single-cell transcriptomics data. However, t-SNE often fails to correctly represent hierarchical relationships between clusters and creates spurious patterns in the embedding. In this work we generalised t-SNE using shape-aware graph distances to mitigate some of the limitations of the t-SNE. Although many methods have been recently proposed to circumvent the shortcomings of t-SNE, notably Uniform manifold approximation (UMAP) and Potential of heat diffusion for affinity-based transition embedding (PHATE), we see a clear advantage of the proposed graph-based method. RESULTS: The superior performance of the proposed method is first demonstrated on simulated data, where a significant improvement compared to t-SNE, UMAP and PHATE, based on quantitative validation indices, is observed when visualising imbalanced, nonlinear, continuous and hierarchically structured data. Thereafter the ability of the proposed method compared to the competing methods to create faithfully low-dimensional embeddings is shown on two real-world data sets, the single-cell transcriptomics data and the MNIST image data. In addition, the only hyper-parameter of the method can be automatically chosen in a data-driven way, which is consistently optimal across all test cases in this study. CONCLUSIONS: In this work we show that the proposed shape-aware stochastic neighbor embedding method creates low-dimensional visualisations that robustly and accurately reveal key structures of high-dimensional data.

BACKGROUND: Single sample pathway analysis (ssPA) transforms molecular level omics data to the pathway level, enabling the discovery of patient-specific pathway signatures. Compared to conventional pathway analysis, ssPA overcomes the limitations by enabling multi-group comparisons, alongside facilitating numerous downstream analyses such as pathway-based machine learning. While in transcriptomics ssPA is a widely used technique, there is little literature evaluating its suitability for metabolomics. Here we provide a benchmark of established ssPA methods (ssGSEA, GSVA, SVD (PLAGE), and z-score) alongside the evaluation of two novel methods we propose: ssClustPA and kPCA, using semi-synthetic metabolomics data. We then demonstrate how ssPA can facilitate pathway-based interpretation of metabolomics data by performing a case-study on inflammatory bowel disease mass spectrometry data, using clustering to determine subtype-specific pathway signatures. RESULTS: While GSEA-based and z-score methods outperformed the others in terms of recall, clustering/dimensionality reduction-based methods provided higher precision at moderate-to-high effect sizes. A case study applying ssPA to inflammatory bowel disease data demonstrates how these methods yield a much richer depth of interpretation than conventional approaches, for example by clustering pathway scores to visualise a pathway-based patient subtype-specific correlation network. We also developed the sspa python package (freely available at https://pypi.org/project/sspa/ ), providing implementations of all the methods benchmarked in this study. CONCLUSION: This work underscores the value ssPA methods can add to metabolomic studies and provides a useful reference for those wishing to apply ssPA methods to metabolomics data.

BACKGROUND: A high recurrence rate has always been a serious problem for treatment of hepatocellular carcinoma (HCC). Exploring predictors of postoperative and posttransplantation recurrence in patients with HCC can guide treatment strategies for clinicians. RESULTS: In this study, logistic regression and multivariate Cox regression models were constructed with microRNA expression profile data from The Cancer Genome Atlas (TCGA) and gene expression omnibus (GEO). The accuracy of predictions was assessed using receiver operating characteristic curve (ROC) and Kaplan‒Meier survival curve analyses. The results showed that the combination of 10 miRNAs (including hsa-miR-509-3p, hsa-miR-769-3p, hsa-miR-671-3p, hsa-miR-296-5p, hsa-miR-767-5p, hsa-miR-421, hsa-miR-193a-3p, hsa-miR-139-3p, hsa-miR-342-3p, and hsa-miR-193a-5p) accurately predicted postoperative and posttransplantation malignancy recurrence in HCC patients and was also valuable for prognostic evaluation of HCC patients. The 10-miRNA prediction model might assist doctors in making prognoses for HCC patients who have a high probability of relapse following surgery and in offering additional, individualized treatment to lessen that risk.

MOTIVATION: Variable selection is a common statistical approach to identifying genes associated with clinical outcomes of scientific interest. There are thousands of genes in genomic studies, while only a limited number of individual samples are available. Therefore, it is important to develop a method to identify genes associated with outcomes of interest that can control finite-sample false discovery rate (FDR) in high-dimensional data settings. RESULTS: This article proposes a novel method named Grace-AKO for graph-constrained estimation (Grace), which incorporates aggregation of multiple knockoffs (AKO) with the network-constrained penalty. Grace-AKO can control FDR in finite-sample settings and improve model stability simultaneously. Simulation studies show that Grace-AKO has better performance in finite-sample FDR control than the original Grace model. We apply Grace-AKO to the prostate cancer data in The Cancer Genome Atlas program by incorporating prostate-specific antigen (PSA) pathways in the Kyoto Encyclopedia of Genes and Genomes as the prior information. Grace-AKO finally identifies 47 candidate genes associated with PSA level, and more than 75% of the detected genes can be validated.

Enhancers are small regions of DNA that bind to proteins, which enhance the transcription of genes. The enhancer may be located upstream or downstream of the gene. It is not necessarily close to the gene to be acted on, because the entanglement structure of chromatin allows the positions far apart in the sequence to have the opportunity to contact each other. Therefore, identifying enhancers and their strength is a complex and challenging task. In this article, a new prediction method based on deep learning is proposed to identify enhancers and enhancer strength, called iEnhancer-DCLA. Firstly, we use word2vec to convert k-mers into number vectors to construct an input matrix. Secondly, we use convolutional neural network and bidirectional long short-term memory network to extract sequence features, and finally use the attention mechanism to extract relatively important features. In the task of predicting enhancers and their strengths, this method has improved to a certain extent in most evaluation indexes. In summary, we believe that this method provides new ideas in the analysis of enhancers.

Microorganisms in the human body have a great impact on human health. Therefore, mastering the potential relationship between microorganisms and diseases is helpful to understand the pathogenesis of diseases and is of great significance to the prevention, diagnosis, and treatment of diseases. In order to predict the potential microbial disease relationship, we propose a new computational model. Firstly, a bi-directional heterogeneous microbial disease network is constructed by integrating multiple similarities, including Gaussian kernel similarity, microbial function similarity, disease semantic similarity, and disease symptom similarity. Secondly, the neighbor information of the network is learned by random walk; Finally, the selection model is used for information aggregation, and the microbial disease node pair is analyzed. Our method is superior to the existing methods in leave-one-out cross-validation and five-fold cross-validation. Moreover, in case studies of different diseases, our method was proven to be effective.

BACKGROUND: Despite numerous molecular and computational advances, roughly half of patients with a rare disease remain undiagnosed after exome or genome sequencing. A particularly challenging barrier to diagnosis is identifying variants that cause deleterious alternative splicing at intronic or exonic loci outside of canonical donor or acceptor splice sites. RESULTS: Several existing tools predict the likelihood that a genetic variant causes alternative splicing. We sought to extend such methods by developing a new metric that aids in discerning whether a genetic variant leads to deleterious alternative splicing. Our metric combines genetic variation in the Genome Aggregate Database with alternative splicing predictions from SpliceAI to compare observed and expected levels of splice-altering genetic variation. We infer genic regions with significantly less splice-altering variation than expected to be constrained. The resulting model of regional splicing constraint captures differential splicing constraint across gene and exon categories, and the most constrained genic regions are enriched for pathogenic splice-altering variants. Building from this model, we developed ConSpliceML. This ensemble machine learning approach combines regional splicing constraint with multiple per-nucleotide alternative splicing scores to guide the prediction of deleterious splicing variants in protein-coding genes. ConSpliceML more accurately distinguishes deleterious and benign splicing variants than state-of-the-art splicing prediction methods, especially in "cryptic" splicing regions beyond canonical donor or acceptor splice sites. CONCLUSION: Integrating a model of genetic constraint with annotations from existing alternative splicing tools allows ConSpliceML to prioritize potentially deleterious splice-altering variants in studies of rare human diseases.

BACKGROUND: Single marker analysis (SMA) with linear mixed models for genome wide association studies has uncovered the contribution of genetic variants to many observed phenotypes. However, SMA has weak false discovery control. In addition, when a few variants have large effect sizes, SMA has low statistical power to detect small and medium effect sizes, leading to low recall of true causal single nucleotide polymorphisms (SNPs). RESULTS: We present the Bayesian Iterative Conditional Stochastic Search (BICOSS) method that controls false discovery rate and increases recall of variants with small and medium effect sizes. BICOSS iterates between a screening step and a Bayesian model selection step. A simulation study shows that, when compared to SMA, BICOSS dramatically reduces false discovery rate and allows for smaller effect sizes to be discovered. Finally, two real world applications show the utility and flexibility of BICOSS. CONCLUSIONS: When compared to widely used SMA, BICOSS provides higher recall of true SNPs while dramatically reducing false discovery rate.

The rapid evolution of image processing equipment and techniques ensures the development of novel picture analysis methodologies. One of the most powerful yet computationally possible algebraic techniques for measuring the topological characteristics of functions is persistent homology. It's an algebraic invariant that can capture topological details at different spatial resolutions. Persistent homology investigates the topological features of a space using a set of sampled points, such as pixels. It can track the appearance and disappearance of topological features caused by changes in the nested space created by an operation known as filtration, in which a parameter scale, in our case the intensity of pixels, is increased to detect changes in the studied space over a range of varying scales. In addition, at the level of machine learning there were many studies and articles witnessing recently the combination between homological persistence and machine learning algorithms. On another level, prostate cancer is diagnosed referring to a scoring criterion describing the severity of the cancer called Gleason score. The classical Gleason system defines five histological growth patterns (grades). In our study we propose to study the Gleason score on some glands issued from a new optical microscopy technique called SLIM. This new optical microscopy technique that combines two classic ideas in light imaging: Zernike's phase contrast microscopy and Gabor's holography. Persistent homology features are computed on these images. We suggested machine learning methods to classify these images into the corresponding Gleason score. Machine learning techniques applied on homological persistence features was very effective in the detection of the right Gleason score of the prostate cancer in these kinds of images and showed an accuracy of above 95%.

BACKGROUND: Personalized therapy has been at the forefront of cancer care, making cancer treatment more effective. Since cancer patients respond individually to drug therapy, predicting the sensitivity of each patient to specific drugs is very helpful to apply therapeutic agents. Traditional methods focus on node (molecular) information but ignore relevant interactions among different nodes, which has very limited application in complex situations, such as cancer drug responses in real clinical practice. RESULTS: Treatment evaluation with Quantified Network (TreeQNet) is a webserver which could predict sensitivity to drugs for patients through the innovative use of proteomic and phosphoproteomic network from tumor tissues. CONCLUSION: TreeQNet service: http://bioinfo.ustc.edu.cn/ . TreeQNet source code: https://github.com/Really00/treeqnet-web-front/ .

BACKGROUND: Huge amounts of molecular interaction data are continuously produced and stored in public databases. Although many bioinformatics tools have been proposed in the literature for their analysis, based on their modeling through different types of biological networks, several problems still remain unsolved when the problem turns on a large scale. RESULTS: We propose DIAMIN, that is, a high-level software library to facilitate the development of applications for the efficient analysis of large-scale molecular interaction networks. DIAMIN relies on distributed computing, and it is implemented in Java upon the framework Apache Spark. It delivers a set of functionalities implementing different tasks on an abstract representation of very large graphs, providing a built-in support for methods and algorithms commonly used to analyze these networks. DIAMIN has been tested on data retrieved from two of the most used molecular interactions databases, resulting to be highly efficient and scalable. As shown by different provided examples, DIAMIN can be exploited by users without any distributed programming experience, in order to perform various types of data analysis, and to implement new algorithms based on its primitives. CONCLUSIONS: The proposed DIAMIN has been proved to be successful in allowing users to solve specific biological problems that can be modeled relying on biological networks, by using its functionalities. The software is freely available and this will hopefully allow its rapid diffusion through the scientific community, to solve both specific data analysis and more complex tasks.

BACKGROUND: False discovery rate (FDR) estimation is very important in proteomics. The target-decoy strategy (TDS), which is often used for FDR estimation, estimates the FDR under the assumption that when spectra are identified incorrectly, the probabilities of the spectra matching the target or decoy peptides are identical. However, no spectra matching target or decoy peptide probabilities are identical. We propose cTDS (target-decoy strategy with candidate peptides) for accurate estimation of the FDR using the probability that the spectrum is identified incorrectly as a target or decoy peptide. RESULTS: Most spectrum cases result in a probability of having the spectrum identified incorrectly as a target or decoy peptide of close to 0.5, but only about 1.14-4.85% of the total spectra have an exact probability of 0.5. We used an entrapment sequence method to demonstrate the accuracy of cTDS. For fixed FDR thresholds (1-10%), the false match rate (FMR) in cTDS is closer than the FMR in TDS. We compared the number of peptide-spectrum matches (PSMs) obtained with TDS and cTDS at a 1% FDR threshold with the HEK293 dataset. In the first and third replications, the number of PSMs obtained with cTDS for the reverse, pseudo-reverse, shuffle, and de Bruijn databases exceeded those obtained with TDS (about 0.001-0.132%), with the pseudo-shuffle database containing less compared to TDS (about 0.05-0.126%). In the second replication, the number of PSMs obtained with cTDS for all databases exceeds that obtained with TDS (about 0.013-0.274%). CONCLUSIONS: When spectra are actually identified incorrectly, most probabilities of the spectra matching a target or decoy peptide are not identical. Therefore, we propose cTDS, which estimates the FDR more accurately using the probability of the spectrum being identified incorrectly as a target or decoy peptide.

BACKGROUND: Kinetic modeling is a powerful tool for understanding the dynamic behavior of biochemical systems. For kinetic modeling, determination of a number of kinetic parameters, such as the Michaelis constant (Km), is necessary, and global optimization algorithms have long been used for parameter estimation. However, the conventional global optimization approach has three problems: (i) It is computationally demanding. (ii) It often yields unrealistic parameter values because it simply seeks a better model fitting to experimentally observed behaviors. (iii) It has difficulty in identifying a unique solution because multiple parameter sets can allow a kinetic model to fit experimental data equally well (the non-identifiability problem). RESULTS: To solve these problems, we propose the Machine Learning-Aided Global Optimization (MLAGO) method for Km estimation of kinetic modeling. First, we use a machine learning-based Km predictor based only on three factors: EC number, KEGG Compound ID, and Organism ID, then conduct a constrained global optimization-based parameter estimation by using the machine learning-predicted Km values as the reference values. The machine learning model achieved relatively good prediction scores: RMSE = 0.795 and R2 = 0.536, making the subsequent global optimization easy and practical. The MLAGO approach reduced the error between simulation and experimental data while keeping Km values close to the machine learning-predicted values. As a result, the MLAGO approach successfully estimated Km values with less computational cost than the conventional method. Moreover, the MLAGO approach uniquely estimated Km values, which were close to the measured values. CONCLUSIONS: MLAGO overcomes the major problems in parameter estimation, accelerates kinetic modeling, and thus ultimately leads to better understanding of complex cellular systems. The web application for our machine learning-based Km predictor is accessible at https://sites.google.com/view/kazuhiro-maeda/software-tools-web-apps , which helps modelers perform MLAGO on their own parameter estimation tasks.