AMEND 2.0: module identification and multi-omic data integration with multiplex-heterogeneous graphs

Background Multi-omic studies provide comprehensive insight into biological systems by evaluating cellular changes between normal and pathological conditions at multiple levels of measurement. Biological networks, which represent interactions or associations between biomolecules, have been highly ef...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:BMC bioinformatics Jg. 26; H. 1; S. 39 - 29
Hauptverfasser: Boyd, Samuel S., Slawson, Chad, Thompson, Jeffrey A.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: London BioMed Central 05.02.2025
BioMed Central Ltd
Springer Nature B.V
BMC
Schlagworte:
Active module identification
Algorithms
Applications software
Bias
Bioinformatics
Biological effects
Biological networks
Biological research
Biological systems
Biology, Experimental
Biomedical and Life Sciences
Biomolecules
Cancer
Carcinoma, Renal Cell - genetics
Carcinoma, Renal Cell - metabolism
Computational Biology - methods
Computational Biology/Bioinformatics
Computer Appl. in Life Sciences
Data analysis
Data integration
Datasets
Electronic data processing
Genomic analysis
Graphs
Humans
Identification methods
Kidney cancer
Kidney Neoplasms - genetics
Kidney Neoplasms - metabolism
Life Sciences
Methods
Microarrays
Modules
Multi-omic data integration
Multiomics
Multiplexing
Network analysis
Probability
Programming languages
Proteins
Public software
Random walk
Renal cell carcinoma
Software
Task complexity
United States
Multi-omic data integration
Active module identification
Biological networks
ISSN:1471-2105, 1471-2105
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Background Multi-omic studies provide comprehensive insight into biological systems by evaluating cellular changes between normal and pathological conditions at multiple levels of measurement. Biological networks, which represent interactions or associations between biomolecules, have been highly effective in facilitating omic analysis. However, current network-based methods lack generalizability to accommodate multiple data types across a range of diverse experiments. Results We present AMEND 2.0, an updated active module identification method which can analyze multiplex and/or heterogeneous networks integrated with multi-omic data in a highly generalizable framework, in contrast to existing methods, which are mostly appropriate for at most two specific omic types. It is powered by Random Walk with Restart for multiplex-heterogeneous networks, with additional capabilities including degree bias adjustment and biased random walk for multi-objective module identification. AMEND was applied to two real-world multi-omic datasets: renal cell carcinoma data from The cancer genome atlas and an O-GlcNAc Transferase knockout study. Additional analyses investigate the performance of various subroutines of AMEND on tasks of node ranking and degree bias adjustment. Conclusions While the analysis of multi-omic datasets in a network context is poised to provide deeper understanding of health and disease, new methods are required to fully take advantage of this increasingly complex data. The current study combines several network analysis techniques into a single versatile method for analyzing biological networks with multi-omic data that can be applied in many diverse scenarios. Software is freely available in the R programming language at https://github.com/samboyd0/AMEND .
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1471-2105
1471-2105
DOI:10.1186/s12859-025-06063-x