A Manhattan plot visualizes genome-wide association study (GWAS) results by displaying -log10 transformed p-values across chromosomal positions. Points are arranged by genomic position along the x-axis with alternating colors for each chromosome, making it easy to identify significant associations. A horizontal threshold line indicates genome-wide significance (typically p < 5×10⁻⁸). This plot is essential for identifying genetic variants associated with traits or diseases.
#' anyplot.ai
#' manhattan-gwas: Manhattan Plot for GWAS
#' Library: ggplot2 3.5.1 | R 4.4.1
#' Quality: 93/100 | Created: 2026-09-05
library(ggplot2)
library(dplyr)
library(ragg)
set.seed(42)
# --- Theme tokens -----------------------------------------------------------
THEME <- Sys.getenv("ANYPLOT_THEME", "light")
PAGE_BG <- if (THEME == "light") "#FAF8F1" else "#1A1A17"
INK <- if (THEME == "light") "#1A1A17" else "#F0EFE8"
INK_SOFT <- if (THEME == "light") "#4A4A44" else "#B8B7B0"
IMPRINT_PALETTE <- c("#009E73", "#C475FD", "#4467A3", "#BD8233",
"#AE3030", "#2ABCCD", "#954477", "#99B314")
ANYPLOT_AMBER <- "#DDCC77"
# --- Data --------------------------------------------------------------------
# Approximate human chromosome lengths (Mb), chromosomes 1-22
chr_lengths <- c(
249, 243, 198, 191, 180, 171, 159, 145, 138, 133,
135, 133, 114, 107, 102, 90, 83, 80, 59, 63, 48, 51
)
n_chr <- length(chr_lengths)
chr_info <- tibble::tibble(
chromosome = 1:n_chr,
length_mb = chr_lengths
) %>%
mutate(offset = lag(cumsum(length_mb), default = 0) * 1e6)
n_snps_total <- 9000
snps_per_chr <- round(n_snps_total * chr_lengths / sum(chr_lengths))
snps <- lapply(1:n_chr, function(i) {
tibble::tibble(
chromosome = i,
position = sort(sample(seq_len(chr_lengths[i] * 1e6), snps_per_chr[i])),
p_value = runif(snps_per_chr[i])
)
}) %>% bind_rows()
# Simulated GWAS hits: clusters of low p-values around a few causal loci
peak_chr <- c(3, 8, 14, 19)
peak_lead_log <- c(12.0, 9.2, 15.4, 7.8)
peaks <- Map(function(chr, lead_log) {
cluster_size <- 24
spread_bp <- rnorm(cluster_size, 0, 1.5e6)
lead_pos <- sample(seq_len(chr_lengths[chr] * 1e6), 1)
cluster_pos <- pmin(pmax(lead_pos + spread_bp, 1), chr_lengths[chr] * 1e6)
cluster_log <- pmax(0.1, lead_log - abs(spread_bp) / 4e5 + rnorm(cluster_size, 0, 0.3))
tibble::tibble(
chromosome = chr,
position = round(cluster_pos),
p_value = 10^(-cluster_log)
)
}, peak_chr, peak_lead_log) %>% bind_rows()
gwas <- bind_rows(snps, peaks) %>%
left_join(chr_info, by = "chromosome") %>%
mutate(
bp_cum = position + offset,
neg_log_p = -log10(p_value),
chr_parity = ifelse(chromosome %% 2 == 0, "even", "odd")
)
chr_axis <- chr_info %>%
mutate(
center = offset + length_mb * 1e6 / 2,
# Chromosomes 19-22 are short and sit close together; thin the labels
# past 18 to give the remaining ones breathing room.
label = ifelse(chromosome > 18 & chromosome %% 2 == 0, "", chromosome)
)
genome_wide_line <- -log10(5e-8)
suggestive_line <- -log10(1e-5)
sig_hits <- gwas %>% filter(p_value < 5e-8)
# Lead SNP (highest -log10 p) per significant locus, for peak annotation
lead_snps <- sig_hits %>%
group_by(chromosome) %>%
slice_max(neg_log_p, n = 1, with_ties = FALSE) %>%
ungroup()
# --- Plot ----------------------------------------------------------------
p <- ggplot(gwas, aes(x = bp_cum, y = neg_log_p, color = chr_parity)) +
geom_hline(yintercept = suggestive_line, linetype = "dotted",
color = ANYPLOT_AMBER, linewidth = 0.5) +
geom_hline(yintercept = genome_wide_line, linetype = "dashed",
color = IMPRINT_PALETTE[5], linewidth = 0.6) +
geom_point(size = 0.6, alpha = 0.65) +
geom_point(data = sig_hits, aes(x = bp_cum, y = neg_log_p),
color = IMPRINT_PALETTE[5], size = 2.2, alpha = 0.9, inherit.aes = FALSE) +
geom_text(data = lead_snps, aes(x = bp_cum, y = neg_log_p, label = paste0("chr", chromosome)),
color = INK, size = 2.6, fontface = "bold", vjust = -0.9, inherit.aes = FALSE) +
scale_color_manual(values = c(odd = IMPRINT_PALETTE[1], even = IMPRINT_PALETTE[3])) +
scale_x_continuous(breaks = chr_axis$center, labels = chr_axis$label,
expand = expansion(mult = 0.01)) +
scale_y_continuous(expand = expansion(mult = c(0, 0.12))) +
labs(
title = "manhattan-gwas · r · ggplot2 · anyplot.ai",
x = "Chromosome",
y = expression(-log[10](italic(p) * "-value"))
) +
guides(color = "none") +
theme_minimal(base_size = 8) +
theme(
plot.background = element_rect(fill = PAGE_BG, color = PAGE_BG),
panel.background = element_rect(fill = PAGE_BG, color = NA),
panel.grid.major.x = element_blank(),
panel.grid.minor.x = element_blank(),
panel.grid.minor.y = element_blank(),
panel.grid.major.y = element_line(color = INK, linewidth = 0.2),
axis.title = element_text(color = INK, size = 10),
axis.text.y = element_text(color = INK_SOFT, size = 8),
axis.text.x = element_text(color = INK_SOFT, size = 6.5),
plot.title = element_text(color = INK, size = 12)
)
# --- Save --------------------------------------------------------------------
ggsave(
filename = sprintf("plot-%s.png", THEME),
plot = p,
device = ragg::agg_png,
width = 8,
height = 4.5,
units = "in",
dpi = 400
)
Runnable source as JSON, for any HTTP client: https://api.anyplot.ai/specs/manhattan-gwas/ggplot2/code. Any spec id and library id listed in llms-full.txt fit the same URL shape; every URL below is complete and callable.
{
"spec_id": "manhattan-gwas",
"language": "r",
"library": "ggplot2",
"page": "https://anyplot.ai/manhattan-gwas/r/ggplot2",
"hub": "https://anyplot.ai/manhattan-gwas",
"code_json": "https://api.anyplot.ai/specs/manhattan-gwas/ggplot2/code",
"spec_json": "https://api.anyplot.ai/specs/manhattan-gwas",
"render_light_png": "https://storage.googleapis.com/anyplot-images/plots/manhattan-gwas/r/ggplot2/plot-light.png",
"render_dark_png": "https://storage.googleapis.com/anyplot-images/plots/manhattan-gwas/r/ggplot2/plot-dark.png",
"quality_score": 93.0,
"license": "MIT",
"guide": "https://anyplot.ai/llms.txt"
}Part of Manhattan Plot for GWAS on anyplot.ai.