Bioinformatics advances最新文献

Imputation for Lipidomics and Metabolomics (ImpLiMet): a web-based application for optimization and method selection for missing data imputation.

IF 2.4

Bioinformatics advances Pub Date : 2025-01-21 eCollection Date: 2025-01-01 DOI: 10.1093/bioadv/vbae209

Huiting Ou, Anuradha Surendra, Graeme S V McDowell, Emily Hashimoto-Roth, Jianguo Xia, Steffany A L Bennett, Miroslava Čuperlović-Culf

{"title":"Imputation for Lipidomics and Metabolomics (ImpLiMet): a web-based application for optimization and method selection for missing data imputation.","authors":"Huiting Ou, Anuradha Surendra, Graeme S V McDowell, Emily Hashimoto-Roth, Jianguo Xia, Steffany A L Bennett, Miroslava Čuperlović-Culf","doi":"10.1093/bioadv/vbae209","DOIUrl":"10.1093/bioadv/vbae209","url":null,"abstract":"Motivation: Missing values are prevalent in high-throughput measurements due to various experimental or analytical reasons. Imputation, the process of replacing missing values in a dataset with estimated values, plays an important role in multivariate and machine learning analyses. The three missingness patterns, including missing completely at random, missing at random, and missing not at random, describe unique dependencies between the missing and observed data. The optimal imputation method for each dataset depends on the type of data, the cause of the missingness, and the nature of relationships between the missing and observed data. The challenge is to identify the optimal imputation solution for a given dataset.Results: ImpLiMet: is a user-friendly web-platform that enables users to impute missing data using eight different methods. For a given dataset, ImpLiMet suggests the optimal imputation solution through a grid search-based investigation of the error rate for imputation across three missingness data simulations. The effect of imputation can be visually assessed by histogram, kurtosis, and skewness, as well as principal component analysis comparing the impact of the chosen imputation method on the distribution and overall behavior of the data.Availability and implementation: ImpLiMet is freely available at https://complimet.ca/shiny/implimet/ and https://github.com/complimet/ImpLiMet.","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"5 1","pages":"vbae209"},"PeriodicalIF":2.4,"publicationDate":"2025-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11761345/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143048755","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

estiMAge: development of a DNA methylation clock to estimate the methylation age of single cells.

IF 2.4

Bioinformatics advances Pub Date : 2025-01-16 eCollection Date: 2025-01-01 DOI: 10.1093/bioadv/vbaf005

Zoe Saßmannshausen, Lisa Blank, Llorenç Solé-Boldo, Frank Lyko, Günter Raddatz

{"title":"estiMAge: development of a DNA methylation clock to estimate the methylation age of single cells.","authors":"Zoe Saßmannshausen, Lisa Blank, Llorenç Solé-Boldo, Frank Lyko, Günter Raddatz","doi":"10.1093/bioadv/vbaf005","DOIUrl":"10.1093/bioadv/vbaf005","url":null,"abstract":"Motivation: Since their introduction about 10 years ago, methylation clocks have provided broad insights into the biological age of different species, tissues, and in the context of several diseases or aging. However, their application to single-cell methylation data remains a major challenge, because of the inherent sparsity of such data, as many CpG sites are not covered. A methylation clock applicable on single-cell level could help to further disentangle the processes that drive the ticking of epigenetic clocks.Results: We have developed estiMAge (\"estimation of Methylation Age\"), a framework that exploits redundancy in methylation data to substitute missing CpGs of trained methylation clocks in single cells. Using Euclidean distance as a measure of similarity, we determine which CpGs covary with the required CpG sites of an epigenetic clock and can be used as surrogates for clock CpGs not covered in single-cell experiments. estiMAge is thus a tool that can be applied to standard epigenetic clocks built on elastic net regression, to achieve bulk and single-cell resolution. We show that estiMAge can accurately predict the ages of young and old hepatocytes and can be used to generate single-cell versions of publicly available epigenetic clocks.Availability and implementation: The source code and instructions for usage of estiMAge are available at https://github.com/DivEpigenetics/estiMAge.","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"5 1","pages":"vbaf005"},"PeriodicalIF":2.4,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11769677/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143048749","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Integrative co-registration of elemental imaging and histopathology for enhanced spatial multimodal analysis of tissue sections through TRACE. 结合元素成像和组织病理学，通过TRACE增强组织切片的空间多模态分析。

IF 2.4

Bioinformatics advances Pub Date : 2025-01-07 eCollection Date: 2025-01-01 DOI: 10.1093/bioadv/vbaf001

Yunrui Lu, Serin Han, Aruesha Srivastava, Neha Shaik, Matthew Chan, Alos Diallo, Naina Kumar, Nishita Paruchuri, Hrishikesh Deosthali, Vismay Ravikumar, Kevin Cornell, Elijah Stommel, Tracy Punshon, Brian Jackson, Fred Kolling, Linda Vahdat, Louis Vaickus, Jonathan Marotti, Sunita Ho, Joshua Levy

{"title":"Integrative co-registration of elemental imaging and histopathology for enhanced spatial multimodal analysis of tissue sections through TRACE.","authors":"Yunrui Lu, Serin Han, Aruesha Srivastava, Neha Shaik, Matthew Chan, Alos Diallo, Naina Kumar, Nishita Paruchuri, Hrishikesh Deosthali, Vismay Ravikumar, Kevin Cornell, Elijah Stommel, Tracy Punshon, Brian Jackson, Fred Kolling, Linda Vahdat, Louis Vaickus, Jonathan Marotti, Sunita Ho, Joshua Levy","doi":"10.1093/bioadv/vbaf001","DOIUrl":"10.1093/bioadv/vbaf001","url":null,"abstract":"Summary: Elemental imaging provides detailed profiling of metal bioaccumulation, offering more precision than bulk analysis by targeting specific tissue areas. However, accurately identifying comparable tissue regions from elemental maps is challenging, requiring the integration of hematoxylin and eosin (H&E) slides for effective comparison. Facilitating the streamlined co-registration of whole slide images (WSI) and elemental maps, TRACE enhances the analysis of tissue regions and elemental abundance in various pathological conditions. Through an interactive containerized web application, TRACE features real-time annotation editing, advanced statistical tools, and data export, supporting comprehensive spatial analysis. Notably, it allows for comparison of elemental abundances across annotated tissue structures and enables integration with other spatial data types through WSI co-registration.Availability and implementation: Available on the following platforms-GitHub: jlevy44/trace_app, PyPI: trace_app, Docker: joshualevy44/trace_app, Singularity: docker://joshualevy44/trace_app.","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"5 1","pages":"vbaf001"},"PeriodicalIF":2.4,"publicationDate":"2025-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11742137/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143017257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Predicting CRISPR-Cas9 off-target effects in human primary cells using bidirectional LSTM with BERT embedding. 利用BERT嵌入的双向LSTM预测CRISPR-Cas9在人原代细胞中的脱靶效应

IF 2.4

Bioinformatics advances Pub Date : 2024-12-30 eCollection Date: 2025-01-01 DOI: 10.1093/bioadv/vbae184

Orhan Sari, Ziying Liu, Youlian Pan, Xiaojian Shao

{"title":"Predicting CRISPR-Cas9 off-target effects in human primary cells using bidirectional LSTM with BERT embedding.","authors":"Orhan Sari, Ziying Liu, Youlian Pan, Xiaojian Shao","doi":"10.1093/bioadv/vbae184","DOIUrl":"https://doi.org/10.1093/bioadv/vbae184","url":null,"abstract":"Motivation: Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)-Cas9 system is a ground-breaking genome editing tool, which has revolutionized cell and gene therapies. One of the essential components involved in this system that ensures its success is the design of an optimal single-guide RNA (sgRNA) with high on-target cleavage efficiency and low off-target effects. This is challenging as many conditions need to be considered, and empirically testing every design is time-consuming and costly. In silico prediction using machine learning models provides high-performance alternatives.Results: We present CrisprBERT, a deep learning model incorporating a Bidirectional Encoder Representations from Transformers (BERT) architecture to provide a high-dimensional embedding for paired sgRNA and DNA sequences and Bidirectional Long Short-term Memory networks for learning, to predict the off-target effects of sgRNAs utilizing only the sgRNAs and their paired DNA sequences. We proposed doublet stack encoding to capture the local energy configuration of the Cas9 binding and applied the BERT model to learn the contextual embedding of the doublet pairs. Our results showed that the new model achieved better performance than state-of-the-art deep learning models regarding single split and leave-one-sgRNA-out cross-validations as well as independent testing.Availability and implementation: The CrisprBERT is available at GitHub: https://github.com/OSsari/CrisprBERT.","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"5 1","pages":"vbae184"},"PeriodicalIF":2.4,"publicationDate":"2024-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11696696/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142933934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Genal: a Python toolkit for genetic risk scoring and Mendelian randomization. 通用：用于遗传风险评分和孟德尔随机化的Python工具包。

IF 2.4

Bioinformatics advances Pub Date : 2024-12-24 eCollection Date: 2025-01-01 DOI: 10.1093/bioadv/vbae207

Cyprien A Rivier, Santiago Clocchiatti-Tuozzo, Shufan Huo, Victor Torres-Lopez, Daniela Renedo, Kevin N Sheth, Guido J Falcone, Julian N Acosta

{"title":"Genal: a Python toolkit for genetic risk scoring and Mendelian randomization.","authors":"Cyprien A Rivier, Santiago Clocchiatti-Tuozzo, Shufan Huo, Victor Torres-Lopez, Daniela Renedo, Kevin N Sheth, Guido J Falcone, Julian N Acosta","doi":"10.1093/bioadv/vbae207","DOIUrl":"https://doi.org/10.1093/bioadv/vbae207","url":null,"abstract":"Motivation: The expansion of genetic association data from genome-wide association studies has increased the importance of methodologies like Polygenic Risk Scores (PRS) and Mendelian Randomization (MR) in genetic epidemiology. However, their application is often impeded by complex, multi-step workflows requiring specialized expertise and the use of disparate tools with varying data formatting requirements. Existing solutions are frequently standalone packages or command-line based-largely due to dependencies on tools like PLINK-limiting accessibility for researchers without computational experience. Given Python's popularity and ease of use, there is a need for an integrated, user-friendly Python toolkit to streamline PRS and MR analyses.Results: We introduce Genal, a Python package that consolidates SNP-level data handling, cleaning, clumping, PRS computation, and MR analyses into a single, cohesive toolkit. By eliminating the need for multiple R packages and for command-line interaction by wrapping around PLINK, Genal lowers the barrier for medical scientists to perform complex genetic epidemiology studies. Genal draws on concepts from several well-established tools, ensuring that users have access to rigorous statistical techniques in the intuitive Python environment. Additionally, Genal leverages parallel processing for MR methods, including MR-PRESSO, significantly reducing the computational time required for these analyses.Availability and implementation: The package is available on Pypi (https://pypi.org/project/genal-python/), the code is openly available on Github with a tutorial: https://github.com/CypRiv/genal, and the documentation can be found on readthedocs: https://genal.rtfd.io.","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"5 1","pages":"vbae207"},"PeriodicalIF":2.4,"publicationDate":"2024-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11706532/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142959799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

QOMIC: quantum optimization for motif identification. QOMIC：用于图案识别的量子优化。

IF 2.4

Bioinformatics advances Pub Date : 2024-12-24 eCollection Date: 2025-01-01 DOI: 10.1093/bioadv/vbae208

Hoang M Ngo, Tamim Khatib, My T Thai, Tamer Kahveci

{"title":"QOMIC: quantum optimization for motif identification.","authors":"Hoang M Ngo, Tamim Khatib, My T Thai, Tamer Kahveci","doi":"10.1093/bioadv/vbae208","DOIUrl":"10.1093/bioadv/vbae208","url":null,"abstract":"Motivation: Network motif identification (MI) problem aims to find topological patterns in biological networks. Identifying disjoint motifs is a computationally challenging problem using classical computers. Quantum computers enable solving high complexity problems which do not scale using classical computers. In this article, we develop the first quantum solution, called QOMIC (Quantum Optimization for Motif IdentifiCation), to the MI problem. QOMIC transforms the MI problem using a integer model, which serves as the foundation to develop our quantum solution. We develop and implement the quantum circuit to find motif locations in the given network using this model.Results: Our experiments demonstrate that QOMIC outperforms the existing solutions developed for the classical computer, in term of motif counts. We also observe that QOMIC can efficiently find motifs in human regulatory networks associated with five neurodegenerative diseases: Alzheimer's, Parkinson's, Huntington's, Amyotrophic Lateral Sclerosis, and Motor Neurone Disease.Availability and implementation: Our implementation can be found in https://github.com/ngominhhoang/Quantum-Motif-Identification.git.","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"5 1","pages":"vbae208"},"PeriodicalIF":2.4,"publicationDate":"2024-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11725347/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142973619","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Identification of promising dipeptidyl peptidase-4 and protein tyrosine phosphatase 1B inhibitors from selected terpenoids through molecular modeling.

IF 2.4

Bioinformatics advances Pub Date : 2024-12-19 eCollection Date: 2025-01-01 DOI: 10.1093/bioadv/vbae205

Oludare M Ogunyemi, Gideon A Gyebi, Femi Olawale, Ibrahim M Ibrahim, Opeyemi Iwaloye, Modupe M Fabusiwa, Stephen Omowaye, Omotade I Oloyede, Charles O Olaiya

{"title":"Identification of promising dipeptidyl peptidase-4 and protein tyrosine phosphatase 1B inhibitors from selected terpenoids through molecular modeling.","authors":"Oludare M Ogunyemi, Gideon A Gyebi, Femi Olawale, Ibrahim M Ibrahim, Opeyemi Iwaloye, Modupe M Fabusiwa, Stephen Omowaye, Omotade I Oloyede, Charles O Olaiya","doi":"10.1093/bioadv/vbae205","DOIUrl":"10.1093/bioadv/vbae205","url":null,"abstract":"Motivation: Investigating novel drug-target interactions is crucial for expanding the chemical space of emerging therapeutic targets in human diseases. Herein, we explored the interactions of dipeptidyl peptidase-4 and protein tyrosine phosphatase 1B with selected terpenoids from African antidiabetic plants.Results: Using molecular docking, molecular dynamics simulations, molecular mechanics with generalized Born and surface area solvation-free energy, and density functional theory analyses, the study revealed dipeptidyl peptidase-4 as a promising target. Cucurbitacin B, 6-oxoisoiguesterin, and 20-epi-isoiguesterinol were identified as potential dipeptidyl peptidase-4 inhibitors with strong binding affinities. These triterpenoids interacted with key catalytic and hydrophobic pockets of dipeptidyl peptidase-4, demonstrating structural stability and flexibility under dynamic conditions, as indicated by dynamics simulation parameters. The free energy analysis further supported the binding affinities in dynamic environments. Quantum mechanical calculations revealed favorable highest occupied molecular orbital and lowest unoccupied molecular orbital energy profiles, indicating the suitability of the hits as proton donors and acceptors, which likely enhance their molecular interactions with the targets. Moreover, the terpenoids showed desirable drug-like properties, suggesting their potential as safe and effective dipeptidyl peptidase-4 inhibitors. These findings may pave the way for the development of novel antidiabetic agents and nutraceuticals based on these promising in silico hits.Availability and implementation: Not applicable.","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"5 1","pages":"vbae205"},"PeriodicalIF":2.4,"publicationDate":"2024-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11751579/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143026004","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploring the role of the Rab network in epithelial-to-mesenchymal transition. 探索 Rab 网络在上皮细胞向间质转化过程中的作用。

IF 2.4

Bioinformatics advances Pub Date : 2024-12-14 eCollection Date: 2025-01-01 DOI: 10.1093/bioadv/vbae200

Unmani Jaygude, Graham M Hughes, Jeremy C Simpson

{"title":"Exploring the role of the Rab network in epithelial-to-mesenchymal transition.","authors":"Unmani Jaygude, Graham M Hughes, Jeremy C Simpson","doi":"10.1093/bioadv/vbae200","DOIUrl":"10.1093/bioadv/vbae200","url":null,"abstract":"Motivation: Rab GTPases (Rabs) are crucial for membrane trafficking within mammalian cells, and their dysfunction is implicated in many diseases. This gene family plays a role in several crucial cellular processes. Network analyses can uncover the complete repertoire of interaction patterns across the Rab network, informing disease research, opening new opportunities for therapeutic interventions.Results: We examined Rabs and their interactors in the context of epithelial-to-mesenchymal transition (EMT), an indicator of cancer metastasizing to distant organs. A Rab network was first established from analysis of literature and was gradually expanded. Our Python module, resnet, assessed its network resilience and selected an optimally sized, resilient Rab network for further analyses. Pathway enrichment confirmed its role in EMT. We then identified 73 candidate genes showing a strong up-/down-regulation, across 10 cancer types, in patients with metastasized tumours compared to only primary-site tumours. We suggest that their encoded proteins might play a critical role in EMT, and further in vitro studies are needed to confirm their role as predictive markers of cancer metastasis. The use of resnet within the systematic analysis approach described here can be easily applied to assess other gene families and their role in biological events of interest.Availability and implementation: Source code for resnet is freely available at https://github.com/Unmani199/resnet.","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"5 1","pages":"vbae200"},"PeriodicalIF":2.4,"publicationDate":"2024-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11684074/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142907962","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SurfR: Riding the wave of RNA-seq data with a comprehensive bioconductor package to identify surface protein-coding genes. SurfR：利用综合生物导引软件包识别表面蛋白编码基因，引领 RNA-seq 数据浪潮。

IF 2.4

Bioinformatics advances Pub Date : 2024-12-14 eCollection Date: 2025-01-01 DOI: 10.1093/bioadv/vbae201

Aurora Maurizio, Anna Sofia Tascini, Marco J Morelli

{"title":"SurfR: Riding the wave of RNA-seq data with a comprehensive bioconductor package to identify surface protein-coding genes.","authors":"Aurora Maurizio, Anna Sofia Tascini, Marco J Morelli","doi":"10.1093/bioadv/vbae201","DOIUrl":"10.1093/bioadv/vbae201","url":null,"abstract":"Motivation: Proteins at the cell surface connect signaling networks and largely determine a cell's capacity to communicate and interact with its environment. In particular, variations in transcriptomic profiles are often observed between healthy and diseased cells, leading to distinct sets of cell-surface proteins. For these reasons, cell-surface proteins may act as biomarkers for the detection of cells of interest in tissues or body fluids, are often the target of pharmaceutical agents, and hold significant promise in the clinical practice for diagnosis, prognosis, treatment development, and evaluation of therapy response. Therefore, implementing robust methods to identify condition-specific cell-surface proteins is of pivotal importance to advance biomedical research.Results: We developed SurfR, an R/Bioconductor package providing a streamlined end-to-end workflow for computationally identifying surface protein-coding genes from expression data. Our user-friendly, comprehensive workflow performs systematic expression data retrieval from public databases, differential gene expression across conditions, integration of datasets, enrichment analysis, identification of targetable proteins on a condition of interest, and data visualization.Availability and implementation: SurfR is released under GNU-GPL-v3.0 License. Source code, documentation, examples, and tutorials are available through Bioconductor (http://www.bioconductor.org/packages/SurfR). RMD notebooks with the use cases code described in the manuscript can be found on GitHub (https://github.com/auroramaurizio/SurfR_UseCases).","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"5 1","pages":"vbae201"},"PeriodicalIF":2.4,"publicationDate":"2024-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11671034/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142904231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multitask knowledge-primed neural network for predicting missing metadata and host phenotype based on human microbiome. 基于人类微生物组预测缺失元数据和宿主表型的多任务知识启动神经网络。

IF 2.4

Bioinformatics advances Pub Date : 2024-12-13 eCollection Date: 2025-01-01 DOI: 10.1093/bioadv/vbae203

Mahsa Monshizadeh, Yuhui Hong, Yuzhen Ye

{"title":"Multitask knowledge-primed neural network for predicting missing metadata and host phenotype based on human microbiome.","authors":"Mahsa Monshizadeh, Yuhui Hong, Yuzhen Ye","doi":"10.1093/bioadv/vbae203","DOIUrl":"10.1093/bioadv/vbae203","url":null,"abstract":"Motivation: Microbial signatures in the human microbiome are closely associated with various human diseases, driving the development of machine learning models for microbiome-based disease prediction. Despite progress, challenges remain in enhancing prediction accuracy, generalizability, and interpretability. Confounding factors, such as host's gender, age, and body mass index, significantly influence the human microbiome, complicating microbiome-based predictions.Results: To address these challenges, we developed MicroKPNN-MT, a unified model for predicting human phenotype based on microbiome data, as well as additional metadata like age and gender. This model builds on our earlier MicroKPNN framework, which incorporates prior knowledge of microbial species into neural networks to enhance prediction accuracy and interpretability. In MicroKPNN-MT, metadata, when available, serves as additional input features for prediction. Otherwise, the model predicts metadata from microbiome data using additional decoders. We applied MicroKPNN-MT to microbiome data collected in mBodyMap, covering healthy individuals and 25 different diseases, and demonstrated its potential as a predictive tool for multiple diseases, which at the same time provided predictions for the missing metadata. Our results showed that incorporating real or predicted metadata helped improve the accuracy of disease predictions, and more importantly, helped improve the generalizability of the predictive models.Availability and implementation: https://github.com/mgtools/MicroKPNN-MT.","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"5 1","pages":"vbae203"},"PeriodicalIF":2.4,"publicationDate":"2024-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11676323/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142904220","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0