@bgicli/bgicli 2.1.0 → 2.2.0

package/data/skills/bio-data-visualization-specialized-omics-plots/SKILL.md

@@ -0,0 +1,251 @@
+<!--
+# COPYRIGHT NOTICE
+# This file is part of the "Universal Biomedical Skills" project.
+# Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>
+# All Rights Reserved.
+#
+# This code is proprietary and confidential.
+# Unauthorized copying of this file, via any medium is strictly prohibited.
+#
+# Provenance: Authenticated by MD BABU MIA
+
+-->
+
+---
+name: bio-data-visualization-specialized-omics-plots
+description: Reusable plotting functions for common omics visualizations. Custom ggplot2/matplotlib implementations of volcano, MA, PCA, enrichment dotplots, boxplots, and survival curves. Use when creating volcano, MA, or enrichment plots.
+tool_type: mixed
+primary_tool: ggplot2
+measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+# Specialized Omics Plots
+
+## Scope
+
+This skill provides **reusable plotting functions** for common omics visualizations that can be applied across different analysis types:
+- Volcano plots (any DE result)
+- MA plots (any log-fold-change data)
+- PCA plots (any high-dimensional data)
+- Enrichment dotplots (manual, not enrichplot)
+- Expression boxplots with statistics
+- Survival curves
+
+**For DESeq2/edgeR built-in functions** (plotMA, plotPCA, plotDispEsts), see `differential-expression/de-visualization`.
+**For enrichplot-specific functions** (dotplot, cnetplot, emapplot, gseaplot2), see `pathway-analysis/enrichment-visualization`.
+
+## Volcano Plot (R)
+
+```r
+library(ggplot2)
+library(ggrepel)
+
+volcano_plot <- function(res, fdr = 0.05, lfc = 1, top_n = 10) {
+    res <- res %>%
+        mutate(
+            significance = case_when(
+                padj < fdr & log2FoldChange > lfc ~ 'Up',
+                padj < fdr & log2FoldChange < -lfc ~ 'Down',
+                TRUE ~ 'NS'
+            ),
+            label = ifelse(rank(padj) <= top_n & significance != 'NS', gene, '')
+        )
+
+    ggplot(res, aes(log2FoldChange, -log10(pvalue), color = significance)) +
+        geom_point(alpha = 0.6, size = 1.5) +
+        geom_text_repel(aes(label = label), color = 'black', size = 3, max.overlaps = 20) +
+        scale_color_manual(values = c('Up' = '#E64B35', 'Down' = '#4DBBD5', 'NS' = 'grey60')) +
+        geom_vline(xintercept = c(-lfc, lfc), linetype = 'dashed', color = 'grey40') +
+        geom_hline(yintercept = -log10(fdr), linetype = 'dashed', color = 'grey40') +
+        labs(x = expression(Log[2]~Fold~Change), y = expression(-Log[10]~P-value)) +
+        theme_bw() + theme(panel.grid = element_blank())
+}
+```
+
+## Volcano Plot (Python)
+
+```python
+import matplotlib.pyplot as plt
+import numpy as np
+
+def volcano_plot(df, fdr=0.05, lfc=1, ax=None):
+    if ax is None:
+        fig, ax = plt.subplots(figsize=(8, 6))
+
+    sig_up = (df['padj'] < fdr) & (df['log2FoldChange'] > lfc)
+    sig_down = (df['padj'] < fdr) & (df['log2FoldChange'] < -lfc)
+    ns = ~(sig_up | sig_down)
+
+    ax.scatter(df.loc[ns, 'log2FoldChange'], -np.log10(df.loc[ns, 'pvalue']),
+               c='grey', alpha=0.5, s=10, label='NS')
+    ax.scatter(df.loc[sig_up, 'log2FoldChange'], -np.log10(df.loc[sig_up, 'pvalue']),
+               c='#E64B35', alpha=0.7, s=15, label='Up')
+    ax.scatter(df.loc[sig_down, 'log2FoldChange'], -np.log10(df.loc[sig_down, 'pvalue']),
+               c='#4DBBD5', alpha=0.7, s=15, label='Down')
+
+    ax.axhline(-np.log10(fdr), ls='--', c='grey', lw=0.8)
+    ax.axvline(-lfc, ls='--', c='grey', lw=0.8)
+    ax.axvline(lfc, ls='--', c='grey', lw=0.8)
+
+    ax.set_xlabel('Log2 Fold Change')
+    ax.set_ylabel('-Log10 P-value')
+    ax.legend()
+    return ax
+```
+
+## MA Plot (R)
+
+```r
+ma_plot <- function(res, fdr = 0.05) {
+    res <- res %>%
+        mutate(significant = padj < fdr & !is.na(padj))
+
+    ggplot(res, aes(log10(baseMean), log2FoldChange, color = significant)) +
+        geom_point(alpha = 0.5, size = 1) +
+        scale_color_manual(values = c('FALSE' = 'grey60', 'TRUE' = '#E64B35')) +
+        geom_hline(yintercept = 0, color = 'black', linewidth = 0.5) +
+        labs(x = expression(Log[10]~Mean~Expression), y = expression(Log[2]~Fold~Change)) +
+        theme_bw() + theme(panel.grid = element_blank(), legend.position = 'none')
+}
+```
+
+## PCA Plot (R)
+
+```r
+pca_plot <- function(vsd, intgroup = 'condition', ntop = 500) {
+    rv <- rowVars(assay(vsd))
+    select <- order(rv, decreasing = TRUE)[seq_len(min(ntop, length(rv)))]
+    pca <- prcomp(t(assay(vsd)[select, ]))
+    percentVar <- round(100 * pca$sdev^2 / sum(pca$sdev^2), 1)
+
+    pca_df <- data.frame(PC1 = pca$x[, 1], PC2 = pca$x[, 2], colData(vsd))
+
+    ggplot(pca_df, aes(PC1, PC2, color = .data[[intgroup]])) +
+        geom_point(size = 3) +
+        stat_ellipse(level = 0.95, linetype = 'dashed') +
+        labs(x = paste0('PC1 (', percentVar[1], '%)'),
+             y = paste0('PC2 (', percentVar[2], '%)')) +
+        theme_bw() + theme(panel.grid = element_blank())
+}
+```
+
+## PCA Plot (Python)
+
+```python
+from sklearn.decomposition import PCA
+import matplotlib.pyplot as plt
+
+def pca_plot(df, metadata, color_by, ax=None):
+    if ax is None:
+        fig, ax = plt.subplots(figsize=(8, 6))
+
+    pca = PCA(n_components=2)
+    pcs = pca.fit_transform(df.T)
+
+    for group in metadata[color_by].unique():
+        mask = metadata[color_by] == group
+        ax.scatter(pcs[mask, 0], pcs[mask, 1], label=group, alpha=0.8, s=50)
+
+    ax.set_xlabel(f'PC1 ({pca.explained_variance_ratio_[0]*100:.1f}%)')
+    ax.set_ylabel(f'PC2 ({pca.explained_variance_ratio_[1]*100:.1f}%)')
+    ax.legend()
+    return ax
+```
+
+## Dotplot for Enrichment (R)
+
+```r
+library(ggplot2)
+
+enrichment_dotplot <- function(enrich_result, top_n = 20) {
+    df <- enrich_result %>%
+        arrange(p.adjust) %>%
+        head(top_n) %>%
+        mutate(Description = factor(Description, levels = rev(Description)),
+               GeneRatio_numeric = sapply(strsplit(GeneRatio, '/'), function(x) as.numeric(x[1])/as.numeric(x[2])))
+
+    ggplot(df, aes(GeneRatio_numeric, Description, size = Count, color = p.adjust)) +
+        geom_point() +
+        scale_color_gradient(low = '#E64B35', high = '#4DBBD5', trans = 'log10') +
+        scale_size_continuous(range = c(3, 10)) +
+        labs(x = 'Gene Ratio', y = NULL, color = 'Adj. P-value', size = 'Count') +
+        theme_bw() + theme(panel.grid.major.y = element_blank())
+}
+```
+
+## Boxplot with Statistics (R)
+
+```r
+library(ggpubr)
+
+expression_boxplot <- function(df, gene, group_var) {
+    ggboxplot(df, x = group_var, y = gene, color = group_var,
+              add = 'jitter', palette = 'npg') +
+        stat_compare_means(method = 't.test', label = 'p.signif') +
+        labs(y = paste0(gene, ' Expression')) +
+        theme(legend.position = 'none')
+}
+```
+
+## UMAP/tSNE Plot (Python)
+
+```python
+import scanpy as sc
+import matplotlib.pyplot as plt
+
+def umap_plot(adata, color, ax=None, **kwargs):
+    if ax is None:
+        fig, ax = plt.subplots(figsize=(8, 6))
+
+    sc.pl.umap(adata, color=color, ax=ax, show=False, **kwargs)
+    return ax
+
+# With custom styling
+sc.pl.umap(adata, color='leiden', palette='tab20', frameon=False,
+           title='', legend_loc='on data', legend_fontsize=8)
+```
+
+## Correlation Plot (R)
+
+```r
+library(corrplot)
+
+cor_mat <- cor(t(top_genes_mat), method = 'pearson')
+corrplot(cor_mat, method = 'color', type = 'lower', order = 'hclust',
+         tl.col = 'black', tl.cex = 0.7, col = colorRampPalette(c('#4DBBD5', 'white', '#E64B35'))(100))
+```
+
+## Violin Plot with Split (R)
+
+```r
+ggplot(df, aes(cluster, expression, fill = condition)) +
+    geom_split_violin(alpha = 0.7) +
+    geom_boxplot(width = 0.2, position = position_dodge(0.5), outlier.shape = NA) +
+    scale_fill_manual(values = c('#4DBBD5', '#E64B35')) +
+    theme_bw()
+```
+
+## Survival Curves (R)
+
+```r
+library(survival)
+library(survminer)
+
+fit <- survfit(Surv(time, status) ~ group, data = df)
+ggsurvplot(fit, data = df, risk.table = TRUE, pval = TRUE,
+           palette = c('#4DBBD5', '#E64B35'),
+           legend.labs = c('Low', 'High'))
+```
+
+## Related Skills
+
+- data-visualization/ggplot2-fundamentals - Base plotting
+- data-visualization/color-palettes - Color selection
+- differential-expression/de-visualization - DE-specific plots
+- pathway-analysis/enrichment-visualization - Enrichment plots
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

package/data/skills/bio-data-visualization-upset-plots/SKILL.md

@@ -0,0 +1,228 @@
+<!--
+# COPYRIGHT NOTICE
+# This file is part of the "Universal Biomedical Skills" project.
+# Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>
+# All Rights Reserved.
+#
+# This code is proprietary and confidential.
+# Unauthorized copying of this file, via any medium is strictly prohibited.
+#
+# Provenance: Authenticated by MD BABU MIA
+
+-->
+
+---
+name: bio-data-visualization-upset-plots
+description: Create UpSet plots to visualize set intersections as an alternative to Venn diagrams using UpSetR or upsetplot. Use when comparing overlapping gene sets, peak sets, or sample groups with more than 3 sets.
+tool_type: mixed
+primary_tool: UpSetR
+measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+# UpSet Plots
+
+## UpSetR (R) - Basic Plot
+
+```r
+library(UpSetR)
+
+# From binary matrix (rows = elements, columns = sets)
+upset(fromExpression(data), order.by = 'freq', nsets = 6)
+
+# From list of sets
+gene_sets <- list(
+    SetA = c('Gene1', 'Gene2', 'Gene3', 'Gene4'),
+    SetB = c('Gene2', 'Gene3', 'Gene5', 'Gene6'),
+    SetC = c('Gene1', 'Gene3', 'Gene6', 'Gene7'),
+    SetD = c('Gene3', 'Gene4', 'Gene7', 'Gene8')
+)
+upset(fromList(gene_sets), order.by = 'freq', nsets = 4)
+```
+
+## UpSetR Customization
+
+```r
+# Customized appearance
+upset(fromList(gene_sets),
+      nsets = 6,
+      nintersects = 40,
+      order.by = 'freq',
+      decreasing = TRUE,
+      mb.ratio = c(0.6, 0.4),  # Matrix to bar ratio
+      point.size = 3,
+      line.size = 1.5,
+      mainbar.y.label = 'Intersection Size',
+      sets.x.label = 'Set Size',
+      text.scale = c(1.5, 1.3, 1.3, 1, 1.5, 1.3),
+      set_size.show = TRUE,
+      set_size.scale_max = 500)
+
+# Custom set colors
+upset(fromList(gene_sets),
+      sets.bar.color = c('#E64B35', '#4DBBD5', '#00A087', '#3C5488'),
+      main.bar.color = '#7E6148',
+      matrix.color = '#7E6148')
+```
+
+## UpSetR with Queries
+
+```r
+# Highlight specific intersections
+upset(fromList(gene_sets),
+      order.by = 'freq',
+      queries = list(
+          list(query = intersects,
+               params = list('SetA', 'SetB'),
+               color = '#E64B35',
+               active = TRUE),
+          list(query = intersects,
+               params = list('SetA', 'SetC', 'SetD'),
+               color = '#4DBBD5',
+               active = TRUE)
+      ))
+
+# Highlight elements matching criteria
+# Requires attribute data frame with element names as row names
+upset(fromList(gene_sets),
+      queries = list(
+          list(query = elements,
+               params = list('logFC', 1, 2),  # column, min, max
+               color = 'red',
+               active = TRUE)
+      ))
+```
+
+## UpSetR with Metadata Boxplots
+
+```r
+# Add attribute plots below intersection matrix
+# Requires data frame with set membership columns + attribute columns
+upset(data,
+      order.by = 'freq',
+      boxplot.summary = c('logFC', 'pvalue'))
+
+# Custom attribute plots
+upset(data,
+      order.by = 'freq',
+      attribute.plots = list(
+          gridrows = 50,
+          plots = list(
+              list(plot = histogram, x = 'logFC', queries = FALSE),
+              list(plot = scatter_plot, x = 'logFC', y = 'pvalue', queries = TRUE)
+          ),
+          ncols = 2
+      ))
+```
+
+## upsetplot (Python) - Basic
+
+```python
+from upsetplot import from_memberships, plot, UpSet
+import matplotlib.pyplot as plt
+
+# From membership lists
+memberships = [
+    ['SetA', 'SetB'],
+    ['SetA'],
+    ['SetB', 'SetC'],
+    ['SetA', 'SetB', 'SetC'],
+    ['SetC'],
+    ['SetA', 'SetC']
+]
+data = from_memberships(memberships)
+
+# Basic plot
+plot(data, show_counts=True)
+plt.savefig('upset.png', dpi=150, bbox_inches='tight')
+```
+
+## upsetplot from DataFrame
+
+```python
+import pandas as pd
+from upsetplot import from_contents, UpSet
+
+# From dict of sets
+gene_sets = {
+    'SetA': ['Gene1', 'Gene2', 'Gene3', 'Gene4'],
+    'SetB': ['Gene2', 'Gene3', 'Gene5', 'Gene6'],
+    'SetC': ['Gene1', 'Gene3', 'Gene6', 'Gene7']
+}
+data = from_contents(gene_sets)
+
+upset = UpSet(data, subset_size='count', show_counts=True, sort_by='cardinality')
+upset.plot()
+plt.savefig('upset.png', dpi=150, bbox_inches='tight')
+```
+
+## upsetplot Customization
+
+```python
+from upsetplot import UpSet
+
+upset = UpSet(data,
+              subset_size='count',
+              show_counts=True,
+              show_percentages=True,
+              sort_by='cardinality',  # or 'degree'
+              sort_categories_by='cardinality',
+              facecolor='#4DBBD5',
+              element_size=40,
+              intersection_plot_elements=10)
+
+fig = plt.figure(figsize=(12, 8))
+upset.plot(fig=fig)
+```
+
+## upsetplot with Metadata
+
+```python
+# Add data attributes for additional plots
+df = pd.DataFrame({
+    'SetA': [True, True, False, True, False],
+    'SetB': [True, False, True, True, False],
+    'SetC': [False, True, True, False, True],
+    'logFC': [1.2, -0.8, 2.1, 0.5, -1.5],
+    'pvalue': [0.01, 0.05, 0.001, 0.2, 0.03]
+})
+df = df.set_index(['SetA', 'SetB', 'SetC'])
+
+upset = UpSet(df, subset_size='count')
+upset.add_stacked_bars(by='significant', colors=['gray', 'red'])
+# Or: upset.add_catplot(value='logFC', kind='box')
+upset.plot()
+```
+
+## Save UpSet Plots
+
+```r
+# R - to PDF
+pdf('upset_plot.pdf', width = 10, height = 6)
+upset(fromList(gene_sets), order.by = 'freq')
+dev.off()
+
+# R - to PNG
+png('upset_plot.png', width = 10, height = 6, units = 'in', res = 300)
+upset(fromList(gene_sets), order.by = 'freq')
+dev.off()
+```
+
+```python
+# Python
+fig = plt.figure(figsize=(10, 6))
+upset.plot(fig=fig)
+plt.savefig('upset.pdf', bbox_inches='tight')
+plt.savefig('upset.png', dpi=300, bbox_inches='tight')
+```
+
+## Related Skills
+
+- data-visualization/heatmaps-clustering - Alternative for smaller sets
+- pathway-analysis/enrichment-visualization - Gene set overlaps
+- differential-expression/de-results - DE gene set comparisons
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

package/data/skills/bio-data-visualization-volcano-customization/SKILL.md

@@ -0,0 +1,233 @@
+<!--
+# COPYRIGHT NOTICE
+# This file is part of the "Universal Biomedical Skills" project.
+# Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>
+# All Rights Reserved.
+#
+# This code is proprietary and confidential.
+# Unauthorized copying of this file, via any medium is strictly prohibited.
+#
+# Provenance: Authenticated by MD BABU MIA
+
+-->
+
+---
+name: bio-data-visualization-volcano-customization
+description: Create publication-ready volcano plots with custom thresholds, gene labels, and highlighting using ggplot2, EnhancedVolcano, or matplotlib. Use when visualizing differential expression or association results with gene annotations.
+tool_type: mixed
+primary_tool: ggplot2
+measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+# Volcano Plot Customization
+
+## ggplot2 Basic Volcano
+
+```r
+library(ggplot2)
+library(ggrepel)
+
+# Add significance category column
+df$significance <- case_when(
+    df$padj < 0.05 & df$log2FoldChange > 1 ~ 'Up',
+    df$padj < 0.05 & df$log2FoldChange < -1 ~ 'Down',
+    TRUE ~ 'NS'
+)
+
+ggplot(df, aes(x = log2FoldChange, y = -log10(pvalue))) +
+    geom_point(aes(color = significance), alpha = 0.6, size = 1.5) +
+    scale_color_manual(values = c(Up = '#E64B35', Down = '#4DBBD5', NS = 'gray70')) +
+    geom_hline(yintercept = -log10(0.05), linetype = 'dashed', color = 'gray40') +
+    geom_vline(xintercept = c(-1, 1), linetype = 'dashed', color = 'gray40') +
+    theme_classic() +
+    labs(x = 'log2 Fold Change', y = '-log10(p-value)', color = 'Regulation')
+```
+
+## ggplot2 with Gene Labels
+
+```r
+# Label top significant genes
+top_genes <- df %>%
+    filter(padj < 0.05, abs(log2FoldChange) > 1) %>%
+    arrange(pvalue) %>%
+    head(20)
+
+ggplot(df, aes(x = log2FoldChange, y = -log10(pvalue))) +
+    geom_point(aes(color = significance), alpha = 0.6, size = 1.5) +
+    scale_color_manual(values = c(Up = '#E64B35', Down = '#4DBBD5', NS = 'gray70')) +
+    geom_text_repel(
+        data = top_genes,
+        aes(label = gene),
+        size = 3,
+        max.overlaps = 20,
+        box.padding = 0.5,
+        segment.color = 'gray50'
+    ) +
+    theme_classic()
+
+# Label specific genes of interest
+genes_of_interest <- c('TP53', 'BRCA1', 'MYC', 'EGFR')
+highlight_df <- df %>% filter(gene %in% genes_of_interest)
+
+ggplot(df, aes(x = log2FoldChange, y = -log10(pvalue))) +
+    geom_point(aes(color = significance), alpha = 0.4, size = 1.5) +
+    geom_point(data = highlight_df, color = 'black', size = 3) +
+    geom_text_repel(data = highlight_df, aes(label = gene), fontface = 'bold') +
+    theme_classic()
+```
+
+## EnhancedVolcano (R)
+
+```r
+library(EnhancedVolcano)
+
+# Basic EnhancedVolcano
+EnhancedVolcano(df,
+    lab = df$gene,
+    x = 'log2FoldChange',
+    y = 'pvalue',
+    pCutoff = 0.05,
+    FCcutoff = 1,
+    title = 'Treatment vs Control',
+    subtitle = 'DE genes highlighted')
+
+# Customized EnhancedVolcano
+EnhancedVolcano(df,
+    lab = df$gene,
+    x = 'log2FoldChange',
+    y = 'pvalue',
+    pCutoff = 0.05,
+    FCcutoff = 1,
+    xlim = c(-5, 5),
+    ylim = c(0, 50),
+    pointSize = 2,
+    labSize = 3,
+    colAlpha = 0.6,
+    col = c('gray70', '#4DBBD5', '#00A087', '#E64B35'),
+    legendLabels = c('NS', 'Log2FC', 'p-value', 'p-value and Log2FC'),
+    legendPosition = 'right',
+    drawConnectors = TRUE,
+    widthConnectors = 0.5,
+    maxoverlapsConnectors = 20,
+    selectLab = genes_of_interest,  # Only label specific genes
+    boxedLabels = TRUE)
+```
+
+## EnhancedVolcano with Custom Keyvals
+
+```r
+# Custom point colors by category
+keyvals <- ifelse(df$log2FoldChange > 2 & df$padj < 0.01, '#E64B35',
+           ifelse(df$log2FoldChange < -2 & df$padj < 0.01, '#4DBBD5',
+           ifelse(df$padj < 0.05, '#00A087', 'gray70')))
+names(keyvals)[keyvals == '#E64B35'] <- 'Highly Up'
+names(keyvals)[keyvals == '#4DBBD5'] <- 'Highly Down'
+names(keyvals)[keyvals == '#00A087'] <- 'Moderate'
+names(keyvals)[keyvals == 'gray70'] <- 'NS'
+
+EnhancedVolcano(df,
+    lab = df$gene,
+    x = 'log2FoldChange',
+    y = 'pvalue',
+    colCustom = keyvals,
+    legendPosition = 'right')
+```
+
+## matplotlib Volcano (Python)
+
+```python
+import matplotlib.pyplot as plt
+import numpy as np
+
+fig, ax = plt.subplots(figsize=(8, 6))
+
+# Color by significance
+colors = np.where((df['padj'] < 0.05) & (df['log2FoldChange'] > 1), '#E64B35',
+         np.where((df['padj'] < 0.05) & (df['log2FoldChange'] < -1), '#4DBBD5', 'gray'))
+
+ax.scatter(df['log2FoldChange'], -np.log10(df['pvalue']),
+           c=colors, alpha=0.6, s=20, edgecolors='none')
+
+# Threshold lines
+ax.axhline(-np.log10(0.05), color='gray', linestyle='--', linewidth=1)
+ax.axvline(-1, color='gray', linestyle='--', linewidth=1)
+ax.axvline(1, color='gray', linestyle='--', linewidth=1)
+
+ax.set_xlabel('log2 Fold Change')
+ax.set_ylabel('-log10(p-value)')
+plt.tight_layout()
+```
+
+## matplotlib with Labels
+
+```python
+from adjustText import adjust_text
+
+# Get top genes to label
+top_idx = df.nsmallest(15, 'pvalue').index
+
+fig, ax = plt.subplots(figsize=(10, 8))
+ax.scatter(df['log2FoldChange'], -np.log10(df['pvalue']), c=colors, alpha=0.5, s=15)
+
+# Add labels with adjust_text to avoid overlaps
+texts = []
+for idx in top_idx:
+    texts.append(ax.text(df.loc[idx, 'log2FoldChange'],
+                         -np.log10(df.loc[idx, 'pvalue']),
+                         df.loc[idx, 'gene'],
+                         fontsize=8))
+
+adjust_text(texts, arrowprops=dict(arrowstyle='-', color='gray', lw=0.5))
+plt.tight_layout()
+```
+
+## Threshold Customization
+
+```r
+# Standard thresholds
+# FC > 1 (2-fold change): Common for RNA-seq, may miss subtle changes
+# FC > 0.58 (~1.5-fold): More sensitive, use for subtle effects
+# padj < 0.05: Standard FDR threshold
+# padj < 0.01: Stringent, fewer false positives
+# padj < 0.1: Relaxed, use for exploratory analysis
+
+# Adjust thresholds based on your data
+pval_threshold <- 0.05
+fc_threshold <- 1  # log2 scale
+
+df$significance <- case_when(
+    df$padj < pval_threshold & df$log2FoldChange > fc_threshold ~ 'Up',
+    df$padj < pval_threshold & df$log2FoldChange < -fc_threshold ~ 'Down',
+    TRUE ~ 'NS'
+)
+```
+
+## Save Publication-Ready Volcano
+
+```r
+# R - high resolution
+ggsave('volcano.pdf', width = 8, height = 6)
+ggsave('volcano.png', width = 8, height = 6, dpi = 300)
+
+# EnhancedVolcano returns ggplot object
+p <- EnhancedVolcano(df, lab = df$gene, x = 'log2FoldChange', y = 'pvalue')
+ggsave('volcano.pdf', p, width = 10, height = 8)
+```
+
+```python
+# Python
+plt.savefig('volcano.pdf', bbox_inches='tight')
+plt.savefig('volcano.png', dpi=300, bbox_inches='tight')
+```
+
+## Related Skills
+
+- differential-expression/de-visualization - DE-specific plots
+- data-visualization/ggplot2-fundamentals - General ggplot2
+- data-visualization/color-palettes - Color selection
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->