Nature 期刊图复刻｜带内嵌边缘密度的多组时序回归拟合图

py小王子2026-06-03 8:49

论文地址：

https://www.nature.com/articles/s41586-023-06861-4

论文题目：

Disproportionate declines of formerly abundant species underlie insect loss

复现图片:

配色方案：

python 复制代码 
PALETTES = {
    1:  {"name": "原图经典 (青/橙/黄)",    "colors": ["#10A5B5", "#E85D3B", "#D1AC22"]},
    2:  {"name": "Nature 杂志风 (蓝/红/绿)", "colors": ["#2F4B7C", "#A05195", "#F95D6A"]},
    3:  {"name": "Science 经典 (深蓝/桔/灰)", "colors": ["#118AB2", "#FFD166", "#06D6A0"]},
    4:  {"name": "优雅莫兰迪 (雾蓝/灰粉/雅绿)", "colors": ["#769FCD", "#D6E6F2", "#B9D7EA"]},
    5:  {"name": "高级学术复古 (藏青/赭红/复古金)", "colors": ["#1D3557", "#E63946", "#457B9D"]},
    6:  {"name": "马卡龙少女 (浅水蓝/柔粉/奶油黄)", "colors": ["#A8DADC", "#EC9A9A", "#F4A261"]},
    7:  {"name": "现代商务 (深灰蓝/亮橙/浅绿)",  "colors": ["#2A4D69", "#4B86B4", "#ADCBE3"]},
    8:  {"name": "大地色系 (墨绿/驼色/焦糖)",    "colors": ["#264653", "#2A9D8F", "#E76F51"]},
    9:  {"name": "冷调科技 (钛银/深邃蓝/霓虹紫)", "colors": ["#0F4C81", "#1FA59B", "#6C5CE7"]},
    10: {"name": "绚丽渐变 (紫罗兰/珊瑚粉/金黄)", "colors": ["#665191", "#D45087", "#FF7C43"]}
}

完整代码：

python 复制代码 
import os
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from scipy.stats import norm

# 1. 自动创建保存图表的文件夹
output_dir = "图表"
os.makedirs(output_dir, exist_ok=True)

# 2. 读取同目录下的 data.xlsx 文件
excel_path = "data.xlsx"
if not os.path.exists(excel_path):
    raise FileNotFoundError(f"未在当前目录下找到 {excel_path} 文件，请确保文件存在。")

df = pd.read_excel(excel_path, skiprows=3)

# 重命名列名以方便代码精确调用数据
df.columns = [
    "Year",
    "Total_Cyan", "Total_Orange", "Total_Yellow",
    "Species_Cyan", "Species_Orange", "Species_Yellow",
    "Simpson_Cyan", "Simpson_Orange", "Simpson_Yellow",
    "Evenness_Cyan", "Evenness_Orange", "Evenness_Yellow"
]

# ==========================================
# 10 种精选配色方案字典
# ==========================================
PALETTES = {
    1:  {"name": "原图经典 (青/橙/黄)",    "colors": ["#10A5B5", "#E85D3B", "#D1AC22"]},
    2:  {"name": "Nature 杂志风 (蓝/红/绿)", "colors": ["#2F4B7C", "#A05195", "#F95D6A"]},
    3:  {"name": "Science 经典 (深蓝/桔/灰)", "colors": ["#118AB2", "#FFD166", "#06D6A0"]},
    4:  {"name": "优雅莫兰迪 (雾蓝/灰粉/雅绿)", "colors": ["#769FCD", "#D6E6F2", "#B9D7EA"]},
    5:  {"name": "高级学术复古 (藏青/赭红/复古金)", "colors": ["#1D3557", "#E63946", "#457B9D"]},
    6:  {"name": "马卡龙少女 (浅水蓝/柔粉/奶油黄)", "colors": ["#A8DADC", "#EC9A9A", "#F4A261"]},
    7:  {"name": "现代商务 (深灰蓝/亮橙/浅绿)",  "colors": ["#2A4D69", "#4B86B4", "#ADCBE3"]},
    8:  {"name": "大地色系 (墨绿/驼色/焦糖)",    "colors": ["#264653", "#2A9D8F", "#E76F51"]},
    9:  {"name": "冷调科技 (钛银/深邃蓝/霓虹紫)", "colors": ["#0F4C81", "#1FA59B", "#6C5CE7"]},
    10: {"name": "绚丽渐变 (紫罗兰/珊瑚粉/金黄)", "colors": ["#665191", "#D45087", "#FF7C43"]}
}

# 在这里选择你想要的配色方案（当前默认为第 1 种）
SELECTED_PALETTE = 1

# 提取当前选择的颜色
current_palette = PALETTES[SELECTED_PALETTE]
c_cyan, c_orange, c_yellow = current_palette["colors"]
print(f"当前启用的配色方案: 方案 {SELECTED_PALETTE} - {current_palette['name']}")

# ==========================================
# 3. 设置 matplotlib 样式
# ==========================================
plt.rcParams['font.family'] = 'sans-serif'
plt.rcParams['font.sans-serif'] = ['DejaVu Sans', 'Arial', 'Helvetica']
plt.rcParams['axes.edgecolor'] = '#000000'
plt.rcParams['axes.linewidth'] = 1.2
plt.rcParams['xtick.color'] = '#000000'
plt.rcParams['ytick.color'] = '#000000'

# 趋势线与散点绘制辅助函数
def plot_with_trend(ax, x, y, color):
    # 绘制数据散点
    ax.scatter(x, y, color=color, s=40, alpha=0.9, edgecolors='none')
    # 基于 Excel 真实数据进行线性回归拟合 (y = mx + b)
    m, b = np.polyfit(x, y, 1)
    ax.plot(x, m * x + b, color=color, linewidth=1.5)

# 初始化 1行 4列 的横向画布
fig, axes = plt.subplots(1, 4, figsize=(15, 3.8))
fig.patch.set_facecolor('white')

# 核心修改：在画布最左上角添加大写粗体序号 "b"
fig.suptitle("b", x=0.01, y=0.96, fontsize=16, fontweight='bold', ha='left', va='top')

# 辅助生成内嵌子图分布的固定 X 轴
x_axis = np.linspace(-3, 3, 100)

# ==========================================
# 子图 1: Total number of individuals
# ==========================================
ax = axes[0]
plot_with_trend(ax, df["Year"], df["Total_Cyan"], c_cyan)
plot_with_trend(ax, df["Year"], df["Total_Orange"], c_orange)
plot_with_trend(ax, df["Year"], df["Total_Yellow"], c_yellow)
ax.set_ylabel("Total number of\nindividuals", fontsize=11, fontweight='bold')
ax.set_ylim(90, 220)
ax.set_yticks([120, 150, 180, 210])

# 左下角嵌入密度分布图 (Inset)
inset1 = ax.inset_axes([0.05, 0.05, 0.3, 0.25])
inset1.fill_between(x_axis - 1.5, norm.pdf(x_axis, 0, 0.4), color=c_cyan, alpha=0.6)
inset1.fill_between(x_axis - 0.5, norm.pdf(x_axis, 0, 0.4), color=c_orange, alpha=0.6)
inset1.fill_between(x_axis + 0.5, norm.pdf(x_axis, 0, 0.4), color=c_yellow, alpha=0.6)
inset1.axvline(x=1.8, color='black', linestyle='--', linewidth=1)
inset1.set_xticks([]); inset1.set_yticks([])
inset1.spines['top'].set_visible(False); inset1.spines['right'].set_visible(False)

# ==========================================
# 子图 2: Number of species
# ==========================================
ax = axes[1]
plot_with_trend(ax, df["Year"], df["Species_Cyan"], c_cyan)
plot_with_trend(ax, df["Year"], df["Species_Orange"], c_orange)
plot_with_trend(ax, df["Year"], df["Species_Yellow"], c_yellow)
ax.set_ylabel("Number of species", fontsize=11, fontweight='bold')
ax.set_ylim(25, 44)
ax.set_yticks([30, 35, 40])

# 左下角嵌入密度分布图 (Inset)
inset2 = ax.inset_axes([0.05, 0.05, 0.3, 0.25])
inset2.fill_between(x_axis - 1.0, norm.pdf(x_axis, 0, 0.5), color=c_yellow, alpha=0.6)
inset2.fill_between(x_axis + 0.5, norm.pdf(x_axis, 0, 0.4), color=c_orange, alpha=0.6)
inset2.fill_between(x_axis + 1.5, norm.pdf(x_axis, 0, 0.3), color=c_cyan, alpha=0.6)
inset2.axvline(x=1.2, color='black', linestyle='--', linewidth=1)
inset2.set_xticks([]); inset2.set_yticks([])
inset2.spines['top'].set_visible(False); inset2.spines['right'].set_visible(False)

# ==========================================
# 子图 3: Simpson diversity
# ==========================================
ax = axes[2]
plot_with_trend(ax, df["Year"], df["Simpson_Cyan"], c_cyan)
plot_with_trend(ax, df["Year"], df["Simpson_Orange"], c_orange)
plot_with_trend(ax, df["Year"], df["Simpson_Yellow"], c_yellow)
ax.set_ylabel("Simpson diversity\n(effective number\nof species)", fontsize=11, fontweight='bold')
ax.set_ylim(8, 31)
ax.set_yticks([10, 15, 20, 25, 30])

# 左上角嵌入密度分布图 (Inset)
inset3 = ax.inset_axes([0.05, 0.65, 0.3, 0.28])
inset3.fill_between(x_axis - 1.8, norm.pdf(x_axis, 0, 0.3), color=c_yellow, alpha=0.6)
inset3.fill_between(x_axis, norm.pdf(x_axis, 0, 0.4), color=c_orange, alpha=0.6)
inset3.fill_between(x_axis + 1.8, norm.pdf(x_axis, 0, 0.3), color=c_cyan, alpha=0.6)
inset3.axvline(x=-0.3, color='black', linestyle='--', linewidth=1)
inset3.set_xticks([]); inset3.set_yticks([])
inset3.spines['top'].set_visible(False); inset3.spines['right'].set_visible(False)

# ==========================================
# 子图 4: Evenness
# ==========================================
ax = axes[3]
plot_with_trend(ax, df["Year"], df["Evenness_Cyan"], c_cyan)
plot_with_trend(ax, df["Year"], df["Evenness_Orange"], c_orange)
plot_with_trend(ax, df["Year"], df["Evenness_Yellow"], c_yellow)
ax.set_ylabel("Evenness", fontsize=11, fontweight='bold')
ax.set_ylim(0.28, 0.75)
ax.set_yticks([0.3, 0.4, 0.5, 0.6, 0.7])

# 左上角嵌入密度分布图 (Inset)
inset4 = ax.inset_axes([0.05, 0.65, 0.3, 0.28])
inset4.fill_between(x_axis - 1.5, norm.pdf(x_axis, 0, 0.4), color=c_yellow, alpha=0.6)
inset4.fill_between(x_axis, norm.pdf(x_axis, 0, 0.5), color=c_orange, alpha=0.6)
inset4.fill_between(x_axis + 1.5, norm.pdf(x_axis, 0, 0.4), color=c_cyan, alpha=0.6)
inset4.axvline(x=-0.4, color='black', linestyle='--', linewidth=1)
inset4.set_xticks([]); inset4.set_yticks([])
inset4.spines['top'].set_visible(False); inset4.spines['right'].set_visible(False)

# ==========================================
# 全局坐标轴统一美化调整
# ==========================================
for ax in axes:
    ax.set_xlabel("Year", fontsize=11)
    ax.set_xlim(-1, 21)
    ax.set_xticks([0, 5, 10, 15, 20])
    ax.spines['top'].set_visible(False)       # 移除顶部边框
    ax.spines['right'].set_visible(False)     # 移除右侧边框
    ax.tick_params(direction='out', length=5, width=1.2) # 刻度线朝外

# 4. 紧凑布局并微调顶部空间（防止序号被裁切），导出高清 PNG 图片
plt.tight_layout()
fig.subplots_adjust(top=0.88)  # 为最上方的序号 "b" 预留出充足的空间
final_output_path = os.path.join(output_dir, "reproduced_chart.png")
plt.savefig(final_output_path, dpi=300, bbox_inches='tight')
plt.close()

print(f"绘图成功！已成功添加最左上角序号 'b'，图表保存为: {final_output_path}")

数据获取

评论+私信获取

VX获取：Nature 期刊图复刻｜带内嵌边缘密度的多组时序回归拟合图

相关推荐
zhangfeng11331 小时前
超算/曙光DCU集群 昆山站 htc /public 目录全解
人工智能·python·机器学习
Maydaycxc1 小时前
Excel/WPS 自动化实战：科学计数法、千张表格循环处理、打包交付的多工具对比
python·自动化·excel·wps·rpa
玫幽倩1 小时前
2026盘古石取证决赛（手机取证）
python·电子取证·计算机取证·聊天软件·手机取证·fic
TechWayfarer1 小时前
IP精准定位服务接入实战：广告投放如何用位置数据做定向策略
python·网络协议·tcp/ip·flask
开源量化GO1 小时前
2026年期货量化主流工具期货与期权程序化选型：统一维护能力对照
python
赶在日落之前1 小时前
使用conda-pack打包完整 Python 环境 + 依赖包，传到无网机器解压即用
开发语言·人工智能·python
HappyAcmen1 小时前
8.角色 Prompt 模板
开发语言·python·prompt
梦想的颜色1 小时前
Docker 入门指南：从零开始掌握容器化技术
运维·服务器·vscode·python·算法·docker·云原生
Soari1 小时前
GitHub 开源项目解析：D4Vinci/Scrapling —— Python 网页抓取与自动化处理工具
python·开源·github·python爬虫·网页抓取·异步抓取
热门推荐
01GitHub 镜像站点022026 年 AI 编程工具终极横评：Cursor vs Claude Code vs Copilot vs Windsurf03【AI】2026 年具身智能模型和世界模型总结04Codex 下载安装指南：Windows 和 macOS 官方版下载05【踩坑记录 | 第一篇】微软商店无法使用时，如何手动安装 OpenAI Codex？附`.msix`文件系统错误解决方法06裂开！ChatGPT 居然开始要手机号验证，附详细解决方法07CC-Switch 下载、安装与使用配置指南【2026.5.29】08CC-Switch & Claude 基于 Linux 服务器安装使用指南09Codex 接入 DeepSeek API 完整配置文档10几个好用的ip纯净度检测网站