Skip to main content
NCBI home page
BMC Bioinformatics logoLink to BMC Bioinformatics
. 2025 Oct 21;26:259. doi: 10.1186/s12859-025-06265-3

MultiModalGraphics: an R package for graphical integration of multi-omics datasets

Foziya Ahmed Mohammed 1,2,3,#, El Hadj Malick Fall 4,#, Kula Kekeba Tune 1,2, Rasha Hammamieh 5, Marti Jett 6, Seid Muhie 5,7,✉
PMCID: PMC12539187  PMID: 41120838

Abstract

Multimodal visualizations are essential for identifying and interpreting complex relationships in diverse, high-dimensional biological datasets. However, existing visualization tools often lack native capabilities for embedding explicit statistical and computational annotations, hindering effective quantitative interpretation. We introduce MultiModalGraphics, an R package designed specifically for creating annotated scatterplots and heatmaps of multi-omics and high-dimensional biological data. The package allows seamless embedding of statistical summaries such as fold-changes, p-values, q-values, and standard deviations, facilitating direct quantitative comparisons. MultiModalGraphics interoperates with Bioconductor packages including MultiAssayExperiment, limma, voom, and iClusterPlus, streamlining workflows from data preprocessing and differential expression analysis to visualization. Case studies on three distinct real-world multimodal datasets illustrate its practical utility. Source code, documentation, and example datasets are available via GitHub (https://github.com/famanalytics0/MultiModalGraphics) and under review for inclusion into Bioconductor.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-025-06265-3.

Keywords: Multimodal visualization, Annotated heatmap, Composite feature heatmap, Thresholded scatterplot

Background and motivation

Multimodal visualizations synthesize numerical, categorical, and temporal information into cohesive visual representation, revealing patterns and relationships across diverse biological data types otherwise obscured when analyzing data modalities independently. Visualizations enriched with quantitative annotations—such as statistical significances, fold-changes, or variance measures—facilitate consistent and reproducible interpretation of contributions and interactions across distinct modalities, thereby supporting data-driven insights and biological understanding.

Several specialized visualization packages are available within the R ecosystem. For instance, scatterplot3d [1] provides three-dimensional scatterplots suited to numerical datasets; MADE4 [2] integrates graphical approaches for multivariate analysis; and multipanelfigure [3] enables assembly of multiple graphical components into cohesive figures. Popular general-purpose packages such as ggplot2 [4], ComplexHeatmap [5], and their extensions like GGally [6] for pairwise visualization, EnrichedHeatmap [7], designed for comprehensive genomic data visualization; pheatmap [8], Heatmap3 [9] and heatmap.2 from gplots [10] package for heatmap visualization; iheatmapr [11] for interactive heatmap; and plot3D [12] for three-dimensional plots, have demonstrated wide applicability. Additionally, specialized tools tailored to biological data analysis include MOFA2 [13], which focuses on latent factor visualizations; Devis [14], targeted toward aggregated differential expression analysis; and Giotto [15] and STUtility [16], both designed specifically for spatial transcriptomics data integration and visualization. While highly effective within their specific scopes, these existing visualization packages typically lack native, out-of-the-box functionalities for directly embedding quantitative annotations—such as significance scores or fold-changes—within graphical outputs. Notable exceptions include VolcaNoseR [17] and EnhancedVolcano [18], which integrate enhanced labeling specifically for volcano plots. Yet these tools rarely offer inherent interoperability with upstream analytical packages (e.g., MultiAssayExperiment [19], limma [20] and voom [21]) commonly used within the Bioconductor ecosystem.

To bridge this gap, we developed MultiModalGraphics, an R package extending the powerful foundations of ggplot2 and ComplexHeatmap. MultiModalGraphics facilitates seamless embedding of statistical and numerical annotations directly within scatterplots and heatmaps, enhancing the interpretability and quantitative depth of multimodal visualizations. Key statistical annotations supported include p-values or multiple-testing adjusted q-values, fold-changes, standard deviations, error margins, and other user-defined statistical summaries. Additionally, the package permits user-driven customizations, enabling incorporation of text labels, equations, symbols and other graphical enhancements directly within the plots.

A notable strength of MultiModalGraphics is its inherent interoperability with widely used analytical frameworks in Bioconductor such as MultiAssayExperiment, limma, and voom. This interoperability enables streamlined workflows—starting from raw expression, RNA-seq, or other omics data matrices and associated metadata, proceeding through normalization and differential expression analysis, and leading directly to scatterplot and heatmap visualizations with natively embedded, user-defined statistical summaries.

Designed to be modality-agnostic, MultiModalGraphics accommodates potentially any feature-by-sample or condition matrix coupled with relevant categorical or phenotypic data (e.g., clinical groups, treatment conditions, timepoints). This design choice ensures broad applicability across high-dimensional biological datasets (multi-omics, genetics) and diverse clinical or phenotypic measurements.

MultiModalGraphics specifically addresses the following objectives:

  • Enhanced visualization Clearly communicating complex biological insights through direct statistical annotations.

  • Contextual integration Combining heterogeneous data types into cohesive visual representations.

  • Anomaly identification Facilitating rapid visual identification of biologically relevant patterns and outliers.

  • Analytical decision-making Supporting biological hypothesis generation and informed decision-making based on integrated visual evidence.

Key design features of MultiModalGraphics include:

  • Native statistical annotation Direct, seamless integration of statistical summaries (fold-changes, significance values) derived from established analytical pipelines (e.g., limma/voom).

  • Customized annotation capabilities Advanced customization via textual annotations, special symbols, and flexible positioning within scatterplots and heatmaps.

  • Graphical customization: User-adjustable visualization grids that facilitate exploration and interpretation of complex multivariate biological datasets.

  • Interactive compatibility Although inherently static, visualization objects are structured to support integration within interactive frameworks such as Shiny R package [22], allowing further extension toward interactive exploration and subsetting of data.

  • Modular design and extensibility Using carefully abstracted, modular S4 classes built upon ggplot2 and ComplexHeatmap, users maintain flexibility to iteratively refine or further annotate visualizations, even post-creation, enabling continuous, adaptive exploration of data insights.

Deployment status of MultiModalGraphics package

The MultiModalGraphics package is currently under review for inclusion in the Bioconductor ecosystem, which provides analytical and visualization tools for high-dimensional biological data. Integration into Bioconductor will ensure standardized documentation, testing, version control, long-term maintenance, and broad visibility within the bioinformatics research community.

Design and implementation

MultiModalGraphics is an R/Bioconductor package that leverages object-oriented S4 classes to visualize multimodal biological data with embedded statistical annotations. Originally developed for collaborative multi-omics studies [23–25], the source code has been refined and packaged into a publicly accessible tool that enables the generation of scatterplots and heatmaps enriched with quantitative measures. The package is fully interoperable with established Bioconductor infrastructure—including MultiAssayExperiment and limma/voom—providing a seamless pipeline from raw data processing to reproducible visualization.

The framework is organized around three core S4 classes—AnnotatedHeatmap, CompositeFeatureHeatmap, and ThresholdedScatterplot—each offering dedicated constructors and methods that streamline the creation of statistically annotated visualizations. To maintain clarity and readability, we provide a concise narrative overview of these classes in the main text, while detailed method listings are available in Supplementary Table S1.

Note Representative R code configurations for all classes and figures are provided in Supplementary Code Examples, ensuring full reproducibility of the visualizations shown in this section.

AnnotatedHeatmap (enhanced statistical heatmaps)

The AnnotatedHeatmap class extends ComplexHeatmap-based visualizations by overlaying statistical outputs—such as p-values, q-values, or variance measures—directly on the heatmap grid. This functionality enables researchers to distinguish statistically significant or trending features beyond color intensity alone. For example, significance levels can be displayed as stars or dots on individual heatmap tiles, allowing immediate recognition of biologically relevant patterns (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Example output from the AnnotatedHeatmap class, displaying statistical overlays (stars/dots) on heatmap tiles to indicate significant or trending molecular features. This enables immediate assessment of statistical relevance beyond color intensity alone. Data are drawn from integrated multimodal pan-cancer datasets (TCGA, ICGC, and cBioPortal), covering six cancer types: CESC (cervical squamous cell carcinoma and endocervical adenocarcinoma), OV (ovarian serous cystadenocarcinoma), PRAD (prostate adenocarcinoma), TGCT (testicular germ cell tumors), UCEC (uterine corpus endometrial carcinoma), and UCS (uterine carcinosarcoma). The corresponding R code configuration used to generate this plot is provided in Supplementary Code Examples

The class is compatible with multiple data sources, including expression matrices, precomputed statistical tables, and MultiAssayExperiment objects, ensuring broad applicability across multi-omics workflows. A representative configuration used to generate Fig. 1 is provided in Supplementary Code Examples, illustrating how statistical thresholds and visual markers are specified to overlay significance directly onto heatmap tiles.

CompositeFeatureHeatmap (pathway and group-level heatmaps)

The CompositeFeatureHeatmap class generates composite heatmaps that integrate multiple dimensions of information at the pathway or feature-group level. Tile colors typically encode pathway activity (e.g., z-scores), while overlaid dots convey enrichment significance (p-values) and dot size reflects feature counts. This multidimensional encoding provides an integrative view of biological activity across tissues, timepoints, or experimental conditions, enabling rapid identification of shared and distinct pathway-level signals (Fig. 2).

Fig. 2.

Fig. 2

Open in a new tab

Example output from the CompositeFeatureHeatmap class, summarizing pathway-level activity across multiple brain regions and timepoints. Tile colors encode z-scores, overlaid dots encode significance, and dot size reflects feature counts, allowing multidimensional comparisons across conditions. The visualization summarizes results from 20 gene expression datasets (each ~ 45 K features across 10 mice) collected from five mouse brain regions at four timepoints (GEO accession GSE45035). The regions include AY (amygdala), HC (hippocampus), MPFC (medial prefrontal cortex), SE (septal region), ST (corpus striatum), and VS (ventral striatum). The heatmap highlights inhibition of neurogenesis and synaptic plasticity pathways associated with differentially expressed genes across time. The corresponding R code configuration used to generate this plot is provided in Supplementary Code Examples

This approach is particularly valuable for summarizing large-scale datasets and supporting comparative analyses across biological contexts. A representative configuration used to generate Fig. 2 is provided in Supplementary Code Examples, demonstrating how z-scores, p-values, and feature counts are mapped to distinct graphical encodings to support pathway-level interpretation.

ThresholdedScatterplot (volcano and thresholded scatterplots)

The ThresholdedScatterplot class generalizes ggplot2-based scatterplots and volcano plots by embedding explicit statistical thresholds directly within the visualization. Fold-change and p-value cutoffs are automatically applied, and the number of upregulated, downregulated, and neutral features is annotated within each panel. This explicit integration of thresholds enhances transparency and allows rapid, quantitative interpretation of differential results across datasets (Fig. 3).

Fig. 3.

Fig. 3

Open in a new tab

Example output from the ThresholdedScatterplot class (volcano-style), illustrating log2 fold changes against − log10 p-values with thresholds for significance explicitly marked. Numbers of up- and down-regulated features are annotated in each panel, enabling transparent and quantitative interpretation of differential results. Data are from multi-tissue, multi-timepoint studies of differentially expressed genes across spleen, heart, blood, and seven brain regions at five timepoints (T5R1, T5R10, T10R1, T10R28, and T10R42; GEO accession GSE68077). Brain region abbreviations: AY (amygdala), HC (hippocampus), MPFC (medial prefrontal cortex), SE (septal region), ST (corpus striatum), and VS (ventral striatum). T indicates days of trauma exposure, and R indicates post-trauma tissue collection days. These datasets were downloaded from Gene Expression Omnibus GSE68077. The corresponding R code configuration used to generate this plot is provided in Supplementary Code

The method supports flexible inputs and produces publication-ready scatterplots that remain fully extensible for further customization. The configuration used to generate Fig. 3 is provided in Supplementary Code Examples, illustrating how thresholds and annotation options are applied to improve interpretability.

High-level wrapper: MultiModalPlot()

The package also provides a high-level wrapper function, MultiModalPlot(), designed to integrate multiple visualization types—such as heatmaps and scatterplots—into a single composite figure. The wrapper automatically detects input types, including raw matrices, precomputed statistical tables, and MultiAssayExperiment objects, and dispatches them to the appropriate visualization class. The configuration used to generate such composite multimodal figures is provided in Supplementary Code Examples.

Importantly, MultiModalPlot() is offered as an optional convenience interface to simplify exploratory workflows and reduce the coding burden for users who wish to generate multimodal panels quickly. It does not replace the core classes, which remain fully accessible for advanced customization. The wrapper returns the underlying S4 or ggplot objects, thereby ensuring that transparency and extensibility are preserved.

Practical guidance on when to use MultiModalPlot() versus the individual S4 classes is provided in Supplementary Table S2, which outlines their complementary roles, typical use cases, and transparency considerations.

Overall, MultiModalGraphics provides a flexible and extensible framework for integrating statistical annotations directly into visualizations. By combining dedicated S4 classes, an optional unifying wrapper, and interoperability with Bioconductor infrastructure, the package supports reproducible, interpretable, and customizable visualization of complex biological datasets.

Case studies and usage examples

The application of the MultiModalGraphics package is illustrated through the example plots (Figs. 1, 2 and 3), which showcase its versatility in visualizing complex and multimodal datasets:

Figure 1 presents visualization of multimodal datasets for six distinct cancer types, encompassing data from 312 Cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), 461 Ovarian serous cystadenocarcinoma (OV), 547 Prostate adenocarcinoma (PRAD), 157 Testicular Germ Cell Tumors (TGCT), 444 Uterine Corpus Endometrial Carcinoma (UCEC), 56 Uterine Carcinosarcoma (UCS) patients, whereby each cancer type in turn has seven different data modalities (types). Which means Fig. 1 visually summarizes 6 × 7 = 42 different datatypes. Messenger RNAs and proteins that showed more variations across the six cancer types and their upstream epigenetic marks (microRNAs and hypermethylated DNA probes) were found to be significantly associated with cellular proliferation and antiapoptotic pathways. The genetic data shows that each of these genes are mutated, though to a different extent of mutational frequencies, with both in terms of copy number alteration (CNA) and structural variation. This integrative approach reveals significant associations between epigenetic markers, and gene expression products (transcripts and proteins) involved in cellular proliferation and antiapoptotic pathways, highlighting the mutational landscape and the reported cancer relevance of these genes.

Figure 2 visualizes multimodal datasets that capture temporal and spatial variations in a mouse model simulating features of PTSD. It collates data from four distinct timepoints and five functionally unique brain regions, effectively summarizing findings from 20 datasets. The visualization underscores the suppression of pathways related to neurogenesis and synaptic plasticity, offering insights into their differential modulation across brain regions over time.

Figure 3 visualizes differentially expressed genes in spleen, heart, blood and seven distinct brain regions (down) at five different time points (across). By presenting both log2-fold changes and negative log10 p-values, and displaying the number of genes that meet specific expression thresholds, it provides a quantitative assessment of gene regulation patterns (number of up and down regulated genes), enhancing the interpretive value of the data. Showing the number of features that pass the user-specified cutoffs is an important feature of MultiModalGraphics. While it is technically possible to script similar annotations in general-purpose plotting libraries, our package offers seamless, out-of-the-box embedding of this quantitative annotation within the R environment, requiring no additional coding from the user. This convenience is not commonly available in existing scatterplot tools or volcano plot packages, where such annotations typically require manual scripting or custom code by the user.

These examples demonstrate how the MultiModalGraphics package can visually represent multimodal (multi-omics, genetic and other high-dimensional biological) datasets, integrating relational and quantitative information from multiple sources.

Functionality and performance

To illustrate how MultiModalGraphics extends existing visualization frameworks, Table 1 summarizes key features relative to ggplot2 [4] and ComplexHeatmap [5].

Table 1.

Key features relative to ggplot2 and ComplexHeatmap

Feature MultiModalGraphics ggplot2 ComplexHeatmap
Direct statistical annotation (p-values, q-values, fold changes) ✓ Manual layers Requires custom grobs
Composite feature grids (CompositeFeatureHeatmap) ✓ × ×
MultiAssayExperiment and Limma/voom pipeline integration ✓ × ×
Automatic count overlays in scatterplots ✓ Via stat_summary ×
Equation/special character/text embedding ✓ Limited ×
Modular S4 API ✓ No S4 API S4-only heatmap API
Open in a new tab

This structured comparison highlights how MultiModalGraphics provides out-of-the-box annotations for multimodal visualization—without low-level grob hacking—while leveraging the underlying power of ggplot2 and ComplexHeatmap

Performance and efficiency

Vectorization and efficient data handling: The MultiModalGraphics package is built upon vectorized statistical computations for filtering, differential expression analysis, and matrix operations, leveraging functions such as matrixStats::rowVars and vapply. Visualization components (volcano and heatmap plots) use precomputed, validated data frames and do not rely on explicit loops, resulting in faster plotting across thousands of features and multiple conditions.

Parallelization: Performance is enhanced through built-in parallelization using BiocParallel::bplapply. Differential expression and other per-cell/modality computations can be distributed across multiple cores or compute nodes. Users may control parallel execution using the BPPARAM argument and toggle between sequential, vectorized, and parallel workflows through the parallel and vectorized options.

Modular S4 object design

The package implements modular S4 classes (ThresholdedScatterplot, AnnotatedHeatmap, CompositeFeatureHeatmap) that separate data preparation, statistical computation, and visualization. Each method is atomic and reusable, enabling users to execute, debug, and benchmark individual steps independently. Empirical examples based on real-world datasets—including The Cancer Genome Atlas (curatedTCGAData, curatedPCaData, TCGAbiolinks), the scNMT multi-omics dataset, and the airway RNA-seq benchmark dataset—are provided in the updated README quick start guide to illustrate typical use cases.

For convenience, a high-level wrapper, MultiModalPlot(), is also available to generate composite multimodal visualizations with minimal code. This wrapper is strictly optional: it supports rapid exploratory analyses but returns the underlying S4 or ggplot objects, ensuring that advanced users maintain full control and can directly employ the individual classes for maximum customization.

The MultiModalGraphics source code has been optimized for efficiency through vectorization, with built-in options for parallel computing. Additional real-world examples and benchmarks are included in the package documentation to guide users in applying the framework to multi-omics and other high-dimensional biological datasets.

Maintenance and extensibility

MultiModalGraphics adheres to Bioconductor’s standards for active maintenance and ongoing development. The package will be regularly evaluated for updates, compatibility, and new feature integration, following Bioconductor’s release cycle. Community contributions are strongly encouraged via GitHub pull requests and issues, and all proposed changes undergo rigorous code review to ensure adherence to Bioconductor guidelines. Prioritization of new features and extensions is informed by user feedback, Bioconductor community discussions, and collaborative research needs, ensuring the package remains up to date, useful and responsive to evolving needs of users.

Conclusion

MultiModalGraphics extends multimodal visualization capabilities for biological data by providing scatterplots and heatmaps annotated with user-defined statistical summaries. The package is interoperable with established analytical pipelines, supporting streamlined analytical and visualization workflows. While MultiModalPlot() offers a convenient entry point for exploratory analysis and rapid multimodal figure generation, the core S4 classes remain the primary interface for advanced customization and method development. This dual design ensures accessibility for new users while preserving transparency, reproducibility, and flexibility for expert users.

Supplementary Information

Additional file1 (DOCX 21 kb) (21.1KB, docx)

Acknowledgements

F.A.M. was financially and logistically supported by Wolkite University, Wolkite, Ethiopia, and the Addis Ababa Science and Technology University, Addis Ababa, Ethiopia. E.M.F. was supported by FALLSYS, Lanham, MD, USA.

Author contributions

F.A.M., EH.M.F and S.M. developed the R package; S.M., F.A.M., K.K.T., M.J. and R.H. Conceptualization and supervision; F.A.M and S.M. wrote the first draft of the manuscript; F.A.M., EH.M.F., K.K.T., M.J., R.H. and S.M. revising the manuscript.

Data availability

All datasets analyzed in this study are publicly available from the repositories listed below. Permanent accession numbers and/or direct URLs are provided. A complete mapping of each dataset to its repository, accession, and direct link provided here and cited in the main text. Mouse PTSD datasets—Gene Expression Omnibus (GEO; NCBI/NIH)—Transcriptome characterization of social stressed C57B6 mice exhibiting PTSD-like features—GSE68077: [https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE68077] (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE68077)—Brain transcriptome profiles in mouse model simulating features of post-traumatic stress disorder—GSE45035: [https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE45035] (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE45035) Pan-cancer resources—The Cancer Genome Atlas (TCGA), via NCI Genomic Data Commons (GDC) Data Portal —TCGA-CESC (Cervical squamous cell carcinoma and endocervical adenocarcinoma): [https://portal.gdc.cancer.gov/projects/TCGA-CESC] (https://portal.gdc.cancer.gov/projects/TCGA-CESC) —TCGA-OV (Ovarian serous cystadenocarcinoma): [https://portal.gdc.cancer.gov/projects/TCGA-OV] (https://portal.gdc.cancer.gov/projects/TCGA-OV) —TCGA-PRAD (Prostate adenocarcinoma): [https://portal.gdc.cancer.gov/projects/TCGA-PRAD](https://portal.gdc.cancer.gov/projects/TCGA-PRAD)—TCGA-TGCT (Testicular germ cell tumors): [https://portal.gdc.cancer.gov/projects/TCGA-TGCT] (https://portal.gdc.cancer.gov/projects/TCGA-TGCT)—TCGA-UCEC (Uterine corpus endometrial carcinoma): [https://portal.gdc.cancer.gov/projects/TCGA-UCEC]) https://portal.gdc.cancer.gov/projects/TCGA-UCEC)—TCGA-UCS (Uterine carcinosarcoma): [https://portal.gdc.cancer.gov/projects/TCGA-UCShttps://portal.gdc.cancer.gov/projects/TCGA-UCS) (Program overview: [https://www.cancer.gov/ccg/research/genome-sequencing/tcga] (https://www.cancer.gov/ccg/research/genome-sequencing/tcga))—International Cancer Genome Consortium (ICGC)—ICGC ARGO (Accelerating Research in Genomic Oncology) Data Platform: [https://platform.icgc-argo.org/] (https://platform.icgc-argo.org) . Access to controlled data may require DAC approval)—cBioPortal for Cancer Genomics: [https://www.cbioportal.org/] (https://www.cbioportal.org) (Study-specific URLs corresponding to the TCGA projects above are provided in this section) In-text cross-references This section is cited in the main text (Methods/Implementation and Figure Legends). Source code, documentation and example datasets are open-source at GitHub: [https://github.com/famanalytics0/MultiModalGraphics] (https://github.com/famanalytics0/MultiModalGraphics) and Bioconductor.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Foziya Ahmed Mohammed and El Hadj Malick Fall have contributed equally to this work.

References

  • 1.Ligges U, Mächler M. Scatterplot3d-an r package for visualizing multivariate data. In. Technical report; 2002.
  • 2.Culhane AC, Thioulouse J, Perrière G, Higgins DG. MADE4: an R package for multivariate analysis of gene expression data. Bioinformatics. 2005;21(11):2789–90. [DOI] [PubMed] [Google Scholar]
  • 3.Graumann J, Cotton R. Multipanelfigure: simple assembly of multiple plots and images into a compound figure. J Stat Softw. 2018;84:1–10.30450020 [Google Scholar]
  • 4.Wickham H. ggplot2. Wiley Interdiscip Rev Comput Stat. 2011;3(2):180–5. [Google Scholar]
  • 5.Gu Z, Eils R, Schlesner M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics. 2016;32(18):2847–9. [DOI] [PubMed] [Google Scholar]
  • 6.Schloerke BCD, Larmarange J, Briatte F, Marbach M, Thoen E, Elberg A, Crowley J. GGally: Extension to 'ggplot2'. R package 2025, version 2.3.0.
  • 7.Gu Z, Eils R, Schlesner M, Ishaque N. Enrichedheatmap: an R/bioconductor package for comprehensive visualization of genomic signal associations. BMC Genom. 2018;19(1):234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.R K: pheatmap: Pretty Heatmaps. R package 2025, version 1.0.13.
  • 9.Zhao S YL, Guo Y, Sheng Q, Shyr Y heatmap3: an improved Heatmap package. R package 2021, version 1.1.9.
  • 10.Warnes GBB, Bonebakker L, Gentleman R, Huber W, Liaw A, Lumley T, Maechler M, Magnusson A, Moeller S, Schwartz M, Venables B, Galili T: gplots: various R Programming Tools for Plotting Data. R package 2024, version 3.2.0.
  • 11.Schep AN, Kummerfeld SK. Iheatmapr: interactive complex heatmaps in R. J Open Source Softw. 2017;2(16):359. [Google Scholar]
  • 12.Soetaert K: plot3D: Plotting multi-dimensional data. R package 2021, version 1.4.
  • 13.Argelaguet R, Velten B, Arnol D, Dietrich S, Zenz T, Marioni JC, et al. Multi-omics factor analysis-a framework for unsupervised integration of multi-omics data sets. Mol Syst Biol. 2018;14(6):e8124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Price A, Caciula A, Guo C, Lee B, Morrison J, Rasmussen A, et al. DEvis: an R package for aggregation and visualization of differential expression data. BMC Bioinf. 2019;20(1):110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Dries R, Zhu Q, Dong R, Eng CL, Li H, Liu K, et al. Giotto: a toolbox for integrative analysis and visualization of spatial expression data. Genome Biol. 2021;22(1):78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Bergenstrahle J, Larsson L, Lundeberg J. Seamless integration of image and molecular analysis for spatial transcriptomics workflows. BMC Genom. 2020;21(1):482. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Goedhart J, Luijsterburg MS. VolcaNoseR is a web app for creating, exploring, labeling and sharing volcano plots. Sci Rep. 2020;10(1):20560. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Blighe K, Rana S, Lewis M. EnhancedVolcano: publication-ready volcano plots with enhanced colouring and labeling. R Package Vers. 2019;1:10–18129. [Google Scholar]
  • 19.Ramos M, Schiffer L, Re A, Azhar R, Basunia A, Rodriguez C, et al. Software for the integration of multiomics experiments in bioconductor. Cancer Res. 2017;77(21):e39–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7):e47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Law CW, Chen Y, Shi W, Smyth GK. voom: precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol. 2014;15(2):R29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Chang W, Cheng J, Allaire J, Sievert C, Schloerke B, Xie Y, Allen J, McPherson J, Dipert A, Borges B. shiny: Web Application Framework for R. R package 2025, version 1.10.0.9001.
  • 23.Muhie S, Gautam A, Chakraborty N, Hoke A, Meyerhoff J, Hammamieh R, et al. Molecular indicators of stress-induced neuroinflammation in a mouse model simulating features of post-traumatic stress disorder. Transl Psychiatry. 2017;7(5):e1135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Muhie S, Gautam A, Meyerhoff J, Chakraborty N, Hammamieh R, Jett M. Brain transcriptome profiles in mouse model simulating features of post-traumatic stress disorder. Mol Brain. 2015;8:14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Muhie S, Gautam A, Yang R, Misganaw B, Daigle BJ Jr., Mellon SH, et al. Molecular signatures of post-traumatic stress disorder in war-zone-exposed veteran and active-duty soldiers. Cell Rep Med. 2023;4(5):101045. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Additional file1 (DOCX 21 kb) (21.1KB, docx)

Data Availability Statement

All datasets analyzed in this study are publicly available from the repositories listed below. Permanent accession numbers and/or direct URLs are provided. A complete mapping of each dataset to its repository, accession, and direct link provided here and cited in the main text. Mouse PTSD datasets—Gene Expression Omnibus (GEO; NCBI/NIH)—Transcriptome characterization of social stressed C57B6 mice exhibiting PTSD-like features—GSE68077: [https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE68077] (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE68077)—Brain transcriptome profiles in mouse model simulating features of post-traumatic stress disorder—GSE45035: [https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE45035] (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE45035) Pan-cancer resources—The Cancer Genome Atlas (TCGA), via NCI Genomic Data Commons (GDC) Data Portal —TCGA-CESC (Cervical squamous cell carcinoma and endocervical adenocarcinoma): [https://portal.gdc.cancer.gov/projects/TCGA-CESC] (https://portal.gdc.cancer.gov/projects/TCGA-CESC) —TCGA-OV (Ovarian serous cystadenocarcinoma): [https://portal.gdc.cancer.gov/projects/TCGA-OV] (https://portal.gdc.cancer.gov/projects/TCGA-OV) —TCGA-PRAD (Prostate adenocarcinoma): [https://portal.gdc.cancer.gov/projects/TCGA-PRAD](https://portal.gdc.cancer.gov/projects/TCGA-PRAD)—TCGA-TGCT (Testicular germ cell tumors): [https://portal.gdc.cancer.gov/projects/TCGA-TGCT] (https://portal.gdc.cancer.gov/projects/TCGA-TGCT)—TCGA-UCEC (Uterine corpus endometrial carcinoma): [https://portal.gdc.cancer.gov/projects/TCGA-UCEC]) https://portal.gdc.cancer.gov/projects/TCGA-UCEC)—TCGA-UCS (Uterine carcinosarcoma): [https://portal.gdc.cancer.gov/projects/TCGA-UCShttps://portal.gdc.cancer.gov/projects/TCGA-UCS) (Program overview: [https://www.cancer.gov/ccg/research/genome-sequencing/tcga] (https://www.cancer.gov/ccg/research/genome-sequencing/tcga))—International Cancer Genome Consortium (ICGC)—ICGC ARGO (Accelerating Research in Genomic Oncology) Data Platform: [https://platform.icgc-argo.org/] (https://platform.icgc-argo.org) . Access to controlled data may require DAC approval)—cBioPortal for Cancer Genomics: [https://www.cbioportal.org/] (https://www.cbioportal.org) (Study-specific URLs corresponding to the TCGA projects above are provided in this section) In-text cross-references This section is cited in the main text (Methods/Implementation and Figure Legends). Source code, documentation and example datasets are open-source at GitHub: [https://github.com/famanalytics0/MultiModalGraphics] (https://github.com/famanalytics0/MultiModalGraphics) and Bioconductor.

Articles from BMC Bioinformatics are provided here courtesy of BMC

RESOURCES