Skip to main content
v2026.11,986 entries · CC-BY 4.0

EnhancedVolcano: Publication-Ready Volcano Plots in R

EnhancedVolcano creates publication-ready volcano plots in R, featuring custom significance thresholds, custom color palettes, and ggrepel label connectors.

Written and maintained by CASRAI Editorial Board

Published · Last updated

On this page

EnhancedVolcano is an open-source Bioconductor package in R developed to construct highly customizable, publication-ready volcano plots from differential expression analysis datasets. By wrapping ggplot2 and integrating the ggrepel algorithm, EnhancedVolcano resolves the long-standing challenge of text collision in dense genomic scatter plots, allowing researchers to automate four-quadrant significance gating, custom color palettes, variable point sizing, and directional connector lines without requiring hundreds of lines of low-level graphic code.

What it is used for in research

In genomic and transcriptomic studies, high-throughput assays such as RNA sequencing (RNA-seq), single-cell transcriptomics, and mass spectrometry generate quantitative comparisons across thousands of molecular features. A central visualization in these studies is the volcano plot, a two-dimensional scatter plot that simultaneously illustrates both biological effect magnitude and statistical evidence. On the horizontal axis, researchers plot effect size—typically expressed as the log2 fold change (log2FC) between experimental cohorts. On the vertical axis, researchers plot statistical significance, expressed as the negative logarithm to base 10 of the p-value (-log10(p-value)). In this coordinate space, genes or proteins with large effect sizes and strong statistical significance populate the upper-left and upper-right quadrants, creating the characteristic eruption profile.

While base R graphics and bare ggplot2 pipelines can render scatter plots, generating clean, publication-grade figures from raw differential expression outputs presents substantial practical hurdles. Standard scatter plots frequently suffer from visual overcrowding in high-density genomic regions, where thousands of invariant genes cluster near the origin. More critically, labeling significant genes using standard graphic text geoms causes severe text collision, resulting in illegible overlapping labels. For deeper conceptual principles regarding threshold selection and quadrant definitions, researchers can consult CASRAI’s dedicated volcano plot RNA-seq interpretation guide.

EnhancedVolcano was engineered by Kevin Blighe, Sharmila Rana, and Myles Lewis to address these visualization bottlenecks within the Bioconductor software ecosystem. Built upon the statistical foundations of differential gene expression analysis pipelines, EnhancedVolcano provides a high-level interface that accepts differential expression tables from packages such as DESeq2, edgeR, and limma. It automatically segregates data points into four distinct significance tiers based on user-defined fold-change and p-value cut-offs, assigns distinct colors to each category, and interfaces directly with the ggrepel package for non-overlapping labels. This integration ensures that text labels are intelligently repelled away from data points and neighboring labels while maintaining clear visual linkages through customizable segment connectors.

Beyond bulk RNA-seq, EnhancedVolcano is widely utilized across biomarker discovery, oncology profiling, single-cell cluster characterization, and differential metabolomics. Its support for multi-attribute encoding—simultaneously modifying point color, size, shape, and transparency based on experimental metadata—allows researchers to communicate complex biological signals in a single composite graphic suitable for peer-reviewed journal submission.

How to install or access it

EnhancedVolcano is an official Bioconductor package and must be installed through the Bioconductor package management utility, BiocManager. Because Bioconductor packages undergo structured peer review, version synchronization, and continuous integration testing against specific R releases, EnhancedVolcano is not distributed through the Comprehensive R Archive Network (CRAN). Attempting to install the package using the base R command install.packages("EnhancedVolcano") will fail to locate the repository.

To install the stable release version of EnhancedVolcano, open an R session (version 4.0.0 or higher is required) and execute the following commands:

# Install BiocManager from CRAN if not already installed
if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

# Install EnhancedVolcano from Bioconductor
BiocManager::install("EnhancedVolcano")

This command automatically resolves and installs all required upstream dependencies, including ggplot2, ggrepel, and core Bioconductor infrastructure packages. In enterprise research environments, high-performance computing (HPC) clusters, or Docker containers where system-wide write permissions are restricted, ensure that your personal R library path (.libPaths()) is configured before running the installation.

For computational biologists who require developmental features or pre-release patches, the developmental branch can be installed directly from Bioconductor devel or the author’s GitHub repository:

# Option A: Install from Bioconductor devel
BiocManager::install(version = "devel")
BiocManager::install("EnhancedVolcano")

# Option B: Install directly from GitHub using remotes
if (!requireNamespace("remotes", quietly = TRUE))
    install.packages("remotes")
remotes::install_github("kevinblighe/EnhancedVolcano")

Once the installation completes successfully, load the package into your R environment and verify the installed release version using the following commands:

library(EnhancedVolcano)
packageVersion("EnhancedVolcano")

Step-by-step instructions

Creating a publication-ready volcano plot from differential expression data involves six structured steps, progressing from raw statistical output ingestion to high-resolution figure export:

  1. Format the input differential expression results matrix. EnhancedVolcano requires an input object (the toptable argument) that is a base R data.frame or matrix. Upstream pipelines such as DESeq2 return S4 DataFrame objects (e.g., DESeqResults), which must be coerced before plotting. Ensure your data frame includes row names matching gene identifiers, an effect-size column (e.g., log2FoldChange), and a significance column (e.g., pvalue or padj):
    # Convert DESeq2 results object to a standard data frame
    res_df <- as.data.frame(res)
    
    # Inspect data frame structure
    head(res_df[, c("log2FoldChange", "pvalue", "padj")])
  2. Generate a baseline volcano plot with automatic four-tier categorization. Run the primary EnhancedVolcano() function by specifying the data frame, label source, x-axis column, and y-axis column. By default, the function establishes a four-color categorization scheme: grey for non-significant genes, forest green for genes passing fold-change alone, royal blue for genes passing statistical significance alone, and red for genes passing both cut-offs:
    EnhancedVolcano(res_df,
        lab = rownames(res_df),
        x = 'log2FoldChange',
        y = 'padj',
        pCutoff = 10e-5,
        FCcutoff = 1.5,
        pointSize = 2.0,
        labSize = 4.0,
        title = 'Volcano Plot: Treatment vs. Control',
        subtitle = 'Differential expression analysis (DESeq2)',
        caption = 'Horizontal line: padj  1.5')
  3. Refine statistical thresholds and line styling. Adjust cut-off boundaries to match your study’s pre-registered statistical analysis plan. In genomics, researchers often prioritize adjusted p-values (q-values and false discovery rate control) over uncorrected nominal p-values to protect against false positives arising from thousands of simultaneous hypothesis tests. You can modify threshold values and line aesthetics using cutoffLineType and cutoffLineWidth:
    EnhancedVolcano(res_df,
        lab = rownames(res_df),
        x = 'log2FoldChange',
        y = 'padj',
        pCutoff = 0.01,
        FCcutoff = 2.0,
        cutoffLineType = 'longdash',
        cutoffLineWidth = 0.8,
        cutoffLineCol = 'black',
        legendPosition = 'right')
  4. Select specific candidate genes and activate connector lines. In whole-transcriptome studies, displaying labels for all statistically significant genes saturates the figure. EnhancedVolcano provides the selectLab parameter to restrict labeling to an explicitly declared vector of genes of interest (such as known oncogenes, transcription factors, or experimental targets). Combine this with drawConnectors = TRUE to draw precise lines from the text label to the data point:
    # Define candidate genes of biological interest
    candidates <- c("MYC", "TP53", "EGFR", "VEGFA", "IL6", "CDKN2A", "STAT3", "TNF")
    
    EnhancedVolcano(res_df,
        lab = rownames(res_df),
        x = 'log2FoldChange',
        y = 'padj',
        selectLab = candidates,
        pCutoff = 0.05,
        FCcutoff = 1.5,
        pointSize = 2.5,
        labSize = 4.5,
        labCol = 'black',
        labFace = 'bold',
        boxedLabels = FALSE,
        drawConnectors = TRUE,
        widthConnectors = 0.5,
        colConnectors = 'grey30',
        arrowheads = FALSE,
        maxoverlapsConnectors = Inf)
  5. Implement custom color palettes via conditional key-value mapping. When standard four-quadrant coloring does not reflect your experimental grouping—for instance, when highlighting specific biological pathways or distinguishing validated targets—use the colCustom argument. Supply a named vector of colors where names match legend categories and values define HEX codes or standard R color strings:
    # Construct custom color vector based on biological criteria
    custom_colors <- rep('grey75', nrow(res_df))
    names(custom_colors)  1.5 & res_df$padj < 0.01)]  1.5 & res_df$padj < 0.01)] <- 'Upregulated (FDR < 0.01)'
    
    # Highlight strongly downregulated genes
    custom_colors[which(res_df$log2FoldChange < -1.5 & res_df$padj < 0.01)] <- '#7570b3'
    names(custom_colors)[which(res_df$log2FoldChange < -1.5 & res_df$padj < 0.01)] <- 'Downregulated (FDR < 0.01)'
    
    # Highlight specific pathway targets
    pathway_genes <- rownames(res_df) %in% c("MYC", "TP53", "EGFR")
    custom_colors[which(pathway_genes)] <- '#1b9e77'
    names(custom_colors)[which(pathway_genes)] <- 'Key Pathway Regulators'
    
    # Render plot with custom palette
    p <- EnhancedVolcano(res_df,
        lab = rownames(res_df),
        x = 'log2FoldChange',
        y = 'padj',
        selectLab = c("MYC", "TP53", "EGFR"),
        colCustom = custom_colors,
        colAlpha = 0.85,
        pCutoff = 0.01,
        FCcutoff = 1.5,
        drawConnectors = TRUE,
        legendPosition = 'right')
  6. Export publication-quality vector and raster graphics. Because EnhancedVolcano() returns a standard ggplot object, you can seamlessly modify it with ggplot2 theme elements, combine it with sibling panels using the patchwork package in R, and export it using ggplot2::ggsave(). High-impact journals typically mandate vector PDF or EPS formats, or high-resolution TIFF images at a minimum of 300 to 600 DPI:
    # Assign the plot to an object
    p_final <- EnhancedVolcano(res_df,
        lab = rownames(res_df),
        x = 'log2FoldChange',
        y = 'padj',
        selectLab = candidates,
        pCutoff = 0.01,
        FCcutoff = 1.5,
        drawConnectors = TRUE)
    
    # Export as vector PDF for print submission
    ggplot2::ggsave("Figure_2_Volcano.pdf", plot = p_final, width = 8.5, height = 7.0, units = "in")
    
    # Export as 300 DPI TIFF with LZW compression
    ggplot2::ggsave("Figure_2_Volcano.tiff", plot = p_final, width = 8.5, height = 7.0, units = "in", dpi = 300, compression = "lzw")

Output and how to read it

The visual output generated by EnhancedVolcano is a Cartesian coordinate plot divided into analytical zones by user-configured threshold lines. Correctly interpreting this graphic requires understanding how geometric coordinates translate into biological and statistical significance:

  • Upper-Right Quadrant (Significant Upregulation): Points in this quadrant possess a positive log2 fold change exceeding +FCcutoff and a vertical position above the horizontal line at -log10(pCutoff). In transcriptomic experiments, these points represent transcripts that are significantly more abundant in the experimental condition relative to the control baseline. For example, a log2FC of +2.0 corresponds to a four-fold increase in expression.
  • Upper-Left Quadrant (Significant Downregulation): Points in this quadrant exhibit a negative log2 fold change dropping below -FCcutoff while remaining above the statistical significance boundary. These indicate genes or proteins that are significantly repressed or depleted in the treatment cohort. A log2FC of -2.0 represents a four-fold reduction (to 25% of baseline levels).
  • Central Top Region (Statistically Significant with Minimal Effect Size): Data points situated between -FCcutoff and +FCcutoff that sit above the horizontal significance line pass the false discovery threshold but exhibit small biological shifts. In large sample sizes or high-powered studies, minute transcriptional differences (e.g., a 10% change, log2FC ~ 0.14) frequently attain very low p-values due to statistical power; however, their biological impact may be minimal. EnhancedVolcano isolates these points in royal blue by default, preventing investigators from conflating statistical precision with biological effect magnitude.
  • Basal and Peripheral Bottom Zones (Non-Significant Transcripts): Points falling below the horizontal threshold fail to achieve statistical significance, regardless of their measured fold change. In RNA-seq datasets, genes with low read counts often display erratic, wide fold changes due to Poisson sampling noise; on the volcano plot, these appear in the far-left or far-right basal regions, correctly coded as non-significant (grey or green) because their variance precludes statistical confidence.

The directional connector lines produced when drawConnectors = TRUE provide unambiguous visual provenance. By utilizing the ggrepel force-directed simulation algorithm, EnhancedVolcano calculates repulsive forces between adjacent text boxes and attractive forces between each text box and its parent data point. When a connector is drawn, the line segment links the label directly to the centroid of the scatter coordinate, eliminating ambiguity regarding which point corresponds to a given gene identifier in dense clusters.

Troubleshooting

When implementing EnhancedVolcano across bioinformatics pipelines, researchers frequently encounter specific technical warnings and formatting errors. The following diagnostic solutions address the most common failure modes:

  • Error: toptable must be a data.frame or matrix:
    Root Cause: This error occurs when a user passes an S4 DataFrame object directly from DESeq2 (such as the output of DESeq2::results()) or an edgeR topTags object without converting it to base R format.
    Resolution: Coerce the results object using as.data.frame(res) prior to invoking EnhancedVolcano. For edgeR outputs, extract the underlying data frame using as.data.frame(topTags(lrt, n = Inf)).
  • Warning: ggrepel: X unlabelled data points (too many overlaps):
    Root Cause: By default, ggrepel halts label placement if a point is bounded by too many overlapping elements, preventing labels from obscuring data.
    Resolution: To force all requested labels to display, specify maxoverlapsConnectors = Inf within the function call. Alternatively, avoid visual crowding entirely by supplying a targeted list of candidate genes using selectLab rather than attempting to label every significant point across the entire genome.
  • Error or missing points due to non-finite values (NA, NaN, or Inf):
    Root Cause: Independent filtering in DESeq2 assigns NA values to adjusted p-values (padj) for genes with extreme outlier counts or near-zero base mean expression. Furthermore, software rounding extremely small p-values (e.g., p < 1e-300) to numeric 0 causes -log10(0) to evaluate to positive infinity (Inf), which disrupts axis scaling.
    Resolution: Clean the input table prior to plotting by removing NA entries and replacing exact zeros with the smallest double-precision floating-point number representable in R:
    # Filter out unadjusted NA rows
    res_clean <- res_df[!is.na(res_df$padj), ]
    
    # Clamp zero p-values to minimum machine precision
    res_clean$padj[res_clean$padj == 0] <- .Machine$double.xmin
  • Legend text truncation or viewport overlap:
    Root Cause: Extended custom legend titles or lengthy category descriptions can overflow the graphical device canvas when the legend is positioned on the right side.
    Resolution: Reposition the legend below the plot using legendPosition = 'bottom', reduce font and symbol dimensions using legendLabSize = 9 and legendIconSize = 3.5, or add custom margin expansion using standard ggplot2 syntax: + theme(legend.position = 'bottom', legend.box = 'horizontal').
  • Clipping of connector line labels near axis boundaries:
    Root Cause: When an extreme fold-change gene sits near the coordinate boundary, ggrepel may position the text box partially outside the plot panel, where default clipping masks it.
    Resolution: Disable coordinate clipping by chaining ggplot2’s coordinate function: + coord_cartesian(clip = 'off'), or manually expand the x-axis limits using xlim = c(-max_val, max_val) with sufficient headroom.

Free alternatives

While EnhancedVolcano is widely utilized in R-based genomics, several open-source and free alternative tools exist across different programming languages and graphical user interfaces. The data table below compares these alternatives across key technical attributes:

Tool Name Primary Language / Interface Collision-Free Text Repulsion Custom Attribute Encoding Bioconductor Pipeline Integration Best Suited Research Context
EnhancedVolcano R (Bioconductor package) Built-in (via ggrepel integration) Extensive (color, shape, size, alpha) Native (direct input from DESeq2 / edgeR) Automated, publication-grade scripting in R
Base ggplot2 + ggrepel R (CRAN packages) Manual (requires geom_text_repel) Unlimited (full manual ggplot2 syntax) Manual (user handles coercion and geoms) Custom figure styling with unique geometries
bioinfokit (visuz) Python (PyPI package) Supported (via adjustText wrapper) Moderate (pre-defined style themes) Indirect (requires pandas DataFrame export) Python-based bioinformatics and Jupyter workflows
VolcaNoseR Web Browser (R Shiny GUI) Built-in (automated repel in canvas) High (interactive menu controls) Manual file upload (CSV / TSV tables) Researchers seeking no-code, point-and-click plots
Galaxy Volcano Plot Web Browser (Galaxy Platform) Basic (R script wrapper on server) Standard (fixed color and cutoff forms) Native within Galaxy history workflows Cloud-based, non-programming genomics pipelines

Choosing between EnhancedVolcano and these alternatives depends largely on the technical infrastructure of your research laboratory. Investigators who maintain end-to-end R analysis scripts benefit most from EnhancedVolcano because it eliminates boilerplate code while retaining the flexibility to modify the returned ggplot object. Conversely, researchers operating in Python environments find bioinfokit or manual matplotlib with adjustText more suitable, while bench scientists without programming experience can leverage Shiny-based web tools like VolcaNoseR to explore differential expression data interactively.

For detailed technical specifications, arguments, and default parameters of EnhancedVolcano compared against other statistical plotting modules, the dataset table below compiles parameter-level configurations from the official Bioconductor documentation.

Frequently asked questions

How do I pass a DESeq2 results object into EnhancedVolcano?

A DESeq2 results object generated by results(dds) is an S4 DESeqResults structure, which EnhancedVolcano cannot read directly. To use it, convert the object to a standard base R data frame using res_df <- as.data.frame(res). Verify that gene symbols are assigned to rownames(res_df) or stored in a dedicated column, and pass x = 'log2FoldChange' and y = 'padj' into EnhancedVolcano().

Should I use nominal p-values or adjusted p-values for the y-axis?

In genomic and high-throughput research, researchers should generally use adjusted p-values (padj or false discovery rate q-values) rather than nominal raw p-values for the y-axis. Genome-wide testing evaluates tens of thousands of features simultaneously, inflating the risk of false positives. Setting y = 'padj' ensures that the vertical coordinate accurately reflects multiple-testing correction, although researchers should note that the axis label will read -log10(padj).

Why are some gene labels omitted or accompanied by a ggrepel warning?

Labels are omitted when the ggrepel algorithm detects that a label cannot be placed without overlapping adjacent labels or data points within its allowed iteration threshold. To resolve this, specify maxoverlapsConnectors = Inf to force placement, increase label repulsive forces using labSize, or restrict labeling to a focused subset of candidate genes using the selectLab argument.

How do I label only a specific subset of candidate genes?

Supply a character vector containing the exact gene symbols of interest to the selectLab argument, such as selectLab = c('BRCA1', 'TP53', 'PTEN'). EnhancedVolcano will suppress text labels for all other data points while continuing to display the full scatter background and statistical threshold lines.

How can I export EnhancedVolcano figures at 300 DPI for journal publication?

Because EnhancedVolcano returns a ggplot object, assign the output to an R variable (e.g., p <- EnhancedVolcano(...)) and export it using ggplot2::ggsave(). Specify filename = 'volcano.tiff', dpi = 300, compression = 'lzw', and explicit physical dimensions in inches, such as width = 8 and height = 7.

Can I customize the plot further using standard ggplot2 syntax?

Yes. The object returned by EnhancedVolcano() is a standard ggplot2 object. You can append additional ggplot2 layers using the + operator, including custom themes (+ theme_bw()), manual axis scales (+ scale_x_continuous()), coordinate adjustments (+ coord_flip()), or title modifications (+ labs()).

Sources

  • Blighe, K., Rana, S., and Lewis, M. (2018). EnhancedVolcano: Publication-ready volcano plots with enhance colouring and labeling. Bioconductor version 3.20. https://bioconductor.org/packages/release/bioc/html/EnhancedVolcano.html (Accessed 2026-11-15).
  • Blighe, K. (2024). EnhancedVolcano Vignette and User Guide. Bioconductor release documentation. https://bioconductor.org/packages/release/bioc/vignettes/EnhancedVolcano/inst/doc/EnhancedVolcano.html (Accessed 2026-11-15).
  • Blighe, K. (2024). EnhancedVolcano Source Code Repository and Issue Tracker. GitHub. https://github.com/kevinblighe/EnhancedVolcano (Accessed 2026-11-15).
  • Slowikowski, K. (2024). ggrepel: Automatically Position Non-Overlapping Text Labels with ggplot2. CRAN package repository. https://cran.r-project.org/package=ggrepel (Accessed 2026-11-15).
  • Love, M. I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology, 15(12), 550. https://genomebiology.biomedcentral.com/articles/10.1186/s13059-014-0550-8 (Accessed 2026-11-15).
  • Bioconductor Core Team (2024). Bioconductor Package Development and Contribution Guidelines. https://www.bioconductor.org/developers/package-guidelines/ (Accessed 2026-11-15).

Data: EnhancedVolcano Function Parameters, Defaults, and Aesthetic Mapping Specifications

Technical compilation of core function parameters in the EnhancedVolcano Bioconductor package, comparing data types, default settings, target aesthetic layers, and collision-handling behaviors as documented in official package vignettes and reference manuals.

EnhancedVolcano Function Parameters, Defaults, and Aesthetic Mapping Specifications
Parameter NameData TypeDefault SettingTarget Graphic LayerCollision Handling
toptabledata.frame or matrixNone (required)Input data coordinate sourceNot applicable
labcharacter vector or column nameNone (required)Point text label identificationInput to ggrepel label queue
xcharacter stringNone (required)Horizontal axis effect size (log2FC)Determines point X coordinate
ycharacter stringNone (required)Vertical axis significance (-log10 p)Determines point Y coordinate
pCutoffnumeric1e-05Horizontal significance threshold lineGates candidate label eligibility
FCcutoffnumeric1.0Vertical log2 fold change boundary linesGates candidate label eligibility
pointSizenumeric2.0Geometric point marker diameterDefines point exclusion radius
labSizenumeric5.0Text label typography point sizeDefines bounding box area in repel
selectLabcharacter vectorNULLRestricts text display to specified genesSuppresses non-matching label boxes
colCustomnamed character vectorNULLUser-defined point color palette overrideOverrides default four-tier coloring
drawConnectorslogicalFALSELine segment between point and text boxActivates ggrepel line segment geom
widthConnectorsnumeric0.5Thickness of connector line segmentControls connector stroke width
colConnectorscharacter stringblackColor palette for connecting line segmentsApplies stroke color to connectors
maxoverlapsConnectorsnumeric or Inf15Overlapping label tolerance thresholdInf forces placement of all labels
cutoffLineTypecharacter stringlongdashLinetype geometry for cutoff boundariesRenders dashed threshold indicators

Sources

Compiled by CASRAI from the sources above. Free to reuse under CC BY 4.0 with attribution.

Follow CASRAI

Research-administration guidance, standards updates and independent tool reviews.

Ask CASRAI · free to try

Ask about EnhancedVolcano: Publication-Ready Volcano Plots in R

2 free questions a day, no account.

An AI assistant for research: funder rules, compliance, methods and publishing. It links the sources behind its answers and labels answers that come from a web search.

Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.

Works on this site and inside Claude, Cursor and the AI tools you already use.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays open to read, with no account and no card.

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →