PathIntegrate: Multivariate modelling approaches for pathway-based multi-omics data integration

PLoS Computational Biology, Journal Year: 2024, Volume and Issue: 20(3), P. e1011814 - e1011814

Published: March 25, 2024

As terabytes of multi-omics data are being generated, there is an ever-increasing need for methods facilitating the integration and interpretation such data. Current typically output lists, clusters, or subnetworks molecules related to outcome. Even with expert domain knowledge, discerning biological processes involved a time-consuming activity. Here we propose PathIntegrate, method integrating datasets based on pathways, designed exploit knowledge systems thus provide interpretable models studies. PathIntegrate employs single-sample pathway analysis transform from molecular pathway-level, applies predictive single-view multi-view model integrate Model outputs include pathways ranked by their contribution outcome prediction, each omics layer, importance molecule in pathway. Using semi-synthetic demonstrate benefit grouping into detect signals low signal-to-noise scenarios, as well ability precisely identify important at effect sizes. Finally, using COPD COVID-19 showcase how enables convenient complex high-dimensional datasets. available open-source Python package.

Language: Английский

---

Investigation of the oxidation rules and oxidative stability of seabuckthorn fruit oil during storage based on lipidomics and metabolomics

Food Chemistry, Journal Year: 2025, Volume and Issue: 476, P. 143238 - 143238

Published: Feb. 6, 2025

Language: Английский

---

On the influence of several factors on pathway enrichment analysis

Briefings in Bioinformatics, Journal Year: 2022, Volume and Issue: 23(3)

Published: April 14, 2022

Pathway enrichment analysis has become a widely used knowledge-based approach for the interpretation of biomedical data. Its popularity led to an explosion both methods and pathway databases. While elegance lies in its simplicity, multiple factors can impact results such analysis, which may not be accounted for. Researchers fail give influential aspects their due, resorting instead popular gene set collections, or default settings. Despite ongoing efforts establish guidelines, meaningful are still hampered by lack consensus gold standards around how should conducted. Nonetheless, concerns have prompted series benchmark studies specifically focused on evaluating influence various results. In this review, we organize summarize findings these benchmarks provide comprehensive overview factors. Our work covers broad spectrum factors, spanning from methodological assumptions those related prior biological knowledge, as definitions database choice. doing so, aim shed light lead insignificant, uninteresting even contradictory Finally, conclude review proposing future well solutions overcome some challenges, originate outlined

Language: Английский

---

RaMP-DB 2.0: a renovated knowledgebase for deriving biological and chemical insight from metabolites, proteins, and genes

Bioinformatics, Journal Year: 2022, Volume and Issue: 39(1)

Published: Nov. 12, 2022

Functional interpretation of high-throughput metabolomic and transcriptomic results is a crucial step in generating insight from experimental data. However, pathway functional information for genes metabolites are distributed among many siloed resources, limiting the scope analyses that rely on single knowledge source.RaMP-DB 2.0 web interface, relational database, API R package designed straightforward comprehensive multi-omic RaMP-DB has been upgraded with an expanded breadth depth chemical annotations (ClassyFire, LIPID MAPS, SMILES, InChIs, etc.), new data types related to lipids incorporated. To streamline entity resolution across multiple source databases, we have implemented semi-automated process, thereby lessening burden harmonization supporting more frequent updates. The associated now supports queries pathways, common reactions (e.g. metabolite-enzyme relationship), ontologies, classes structures, as well enrichment pathways (multi-omic) classes. Lastly, interface completely redesigned using Angular framework.The code used build all components freely available GitHub at https://github.com/ncats/ramp-db, https://github.com/ncats/RaMP-Client/ https://github.com/ncats/RaMP-Backend. application can be accessed https://rampdb.nih.gov/.Supplementary Bioinformatics online.

Language: Английский

---

PathBank 2.0—the pathway database for model organism metabolomics

Nucleic Acids Research, Journal Year: 2023, Volume and Issue: 52(D1), P. D654 - D662

Published: Nov. 14, 2023

Abstract PathBank (https://pathbank.org) and its predecessor database, the Small Molecule Pathway Database (SMPDB), have been providing comprehensive metabolite pathway information for metabolomics community since 2010. Over past 14 years, these databases grown evolved significantly to meet needs of respond continuing changes in computing technology. This year's update, 2.0, brings a number important improvements upgrades that should make database more useful appealing larger cross-section users. In particular, include: (i) significant increase primary or canonical pathways (from 1720 6951); (ii) massive total 110 234 605 359); (iii) quality diagrams descriptions; (iv) strong emphasis on drug metabolism mechanism pathways; (v) making most images slide-compatible manuscript-compatible; (vi) adding tools support better filtering selecting through complete taxonomy; (vii) analysis visualizing calculating enrichment. Many other minor updates content, interface general performance website also made. Overall, we believe greatly enhance PathBank's ease use potential applications interpreting data.

Language: Английский

---

Multi-omics approaches in psychoneuroimmunology and health research: Conceptual considerations and methodological recommendations

Brain Behavior and Immunity, Journal Year: 2023, Volume and Issue: 114, P. 475 - 487

Published: Aug. 4, 2023

Language: Английский

---

IDSL.GOA: gene ontology analysis for interpreting metabolomic datasets

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Jan. 14, 2024

Biological interpretation of metabolomic datasets often ends at a pathway analysis step to find the over-represented metabolic pathways in list statistically significant metabolites. However, definitions biochemical and metabolite coverage vary among different curated databases, leading missed interpretations. For lists genes, transcripts proteins, Gene Ontology (GO) terms over-presentation has become standardized approach for biological interpretation. But, GO not been achieved datasets. We present new knowledgebase (KB) online tool, Analysis by Integrated Data Science Laboratory Metabolomics Exposomics (IDSL.GOA) conduct over-representation list. The IDSL.GOA KB covers 2393 associated 3144 1,492 EC annotations, 2621 case study older versus young female brain cortex metabolome highlighted 82 being significantly overrepresented (FDR < 0.05). showed how identified key relevant processes that were yet covered other databases. Overall, we suggest should be limited only maps can also leverage as well. provides useful tool this purpose, allowing more comprehensive accurate data. accessed https://goa.idsl.me/ .

Language: Английский

---

Metabolomics of asthma, COPD, and asthma-COPD overlap: an overview

Critical Reviews in Clinical Laboratory Sciences, Journal Year: 2022, Volume and Issue: 60(2), P. 153 - 170

Published: Nov. 24, 2022

The two common progressive lung diseases, asthma and chronic obstructive pulmonary disease (COPD), are the leading causes of morbidity mortality worldwide. Asthma-COPD overlap, referred to as ACO, is another complex that manifests itself with features both COPD. has no clear diagnostic or therapeutic guidelines, thereby making diagnosis treatment challenging. Though a number studies on ACO have been documented, gaps in knowledge regarding pathophysiologic mechanism this disorder exist. Addressing issue an urgent need for improved management disease. Metabolomics, increasingly popular technique, reveals pathogenesis diseases holds promise biomarker discovery. This comprehensive narrative review, comprising 99 original research articles last five years (2017–2022), summarizes scientific advances terms metabolic alterations patients asthma, COPD, ACO. analytical tools, nuclear magnetic resonance (NMR), gas chromatography-mass spectrometry (GC-MS), liquid (LC-MS), commonly used study expression metabolome, discussed. Challenges frequently encountered during metabolite identification quality assessment highlighted. Bridging gap between phenotype metabotype envisioned future.

Language: Английский

---

Ten quick tips for computational analysis of medical images

PLoS Computational Biology, Journal Year: 2023, Volume and Issue: 19(1), P. e1010778 - e1010778

Published: Jan. 5, 2023

Medical imaging is a great asset for modern medicine, since it allows physicians to spatially interrogate disease site, resulting in precise intervention diagnosis and treatment, observe particular aspect of patients' conditions that otherwise would not be noticeable. Computational analysis medical images, moreover, can allow the discovery patterns correlations among cohorts patients with same disease, thus suggesting common causes or providing useful information better therapies cures. Machine learning deep applied particular, have produced new, unprecedented results pave way advanced frontiers discoveries. While computational images has become easier, however, possibility make mistakes generate inflated misleading too, hindering reproducibility deployment. In this article, we provide ten quick tips perform avoiding pitfalls noticed multiple studies past. We believe our guidelines, if taken into practice, help computational-medical community scientific research eventually positive impact on lives worldwide.

Language: Английский

---

Single sample pathway analysis in metabolomics: performance evaluation and application

BMC Bioinformatics, Journal Year: 2022, Volume and Issue: 23(1)

Published: Nov. 14, 2022

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

Language: Английский

---