生信图表大全（第一篇）：41 张图的坐标轴、参数与绘制语言

gene,log2FC,padj
DWF4,2.85,3.2e-06
BZR1,1.92,4.1e-05
CHS,3.41,8.8e-08
F3H,2.34,1.2e-05
ANS,2.66,5.5e-06
PAL,1.63,2.3e-04
SAUR32,1.28,3.5e-03
GA20ox,1.75,9.9e-04
EXP8,1.42,4.7e-03
CHI,2.11,6.8e-05
FLS,1.55,1.8e-04
PER12,1.19,2.1e-02
DET2,-1.87,1.1e-04
GA2ox,-1.44,2.9e-03
ANR,-2.22,8.3e-05
F3H5H,-1.62,7.4e-03
LAC4,-1.28,1.6e-02
SUS1,-1.09,3.9e-02
BRI1,-1.71,5.2e-04
XTH22,-1.35,2.6e-03
ACT7,0.21,0.61
EF1a,-0.12,0.77
TUB6,0.34,0.44
UBQ10,-0.05,0.91
RBCS,0.58,0.18
LHCB,-0.42,0.29
NCED3,0.87,0.061
PYL4,-0.66,0.23
ABF2,0.44,0.37
SOD,-0.28,0.52
APX1,0.73,0.12
CAT2,-0.51,0.31
HSP70,0.39,0.41
WRKY1,0.95,0.049
MYB12,1.05,0.058

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd

df = pd.read_csv("volcano.csv")          # 上面折叠块里的数据
y = -np.log10(df["padj"])
up = (df.log2FC >= 1) & (y > -np.log10(0.05))
dn = (df.log2FC <= -1) & (y > -np.log10(0.05))

fig, ax = plt.subplots(figsize=(7.2, 5.4))
ax.scatter(df.log2FC[up], y[up], c="#C0392B", label="Up", s=42)
ax.scatter(df.log2FC[dn], y[dn], c="#2471A3", label="Down", s=42)
ax.scatter(df.log2FC[~(up | dn)], y[~(up | dn)], c="#B0B0B0",
           label="Not significant", s=42)
for v in (1, -1):
    ax.axvline(v, color="grey", ls="--", lw=1)
ax.axhline(-np.log10(0.05), color="grey", ls="--", lw=1)
ax.set_xlabel("log2 (fold change, treated / control)")
ax.set_ylabel("-log10 (adjusted p-value)")
ax.legend()
fig.savefig("01-volcano.png", dpi=200, bbox_inches="tight")

gene,A_Z1,A_Z2,A_Z3,B_Z1,B_Z2,B_Z3
DWF4,-1.2,-0.1,1.6,-0.9,0.2,1.4
DET2,-0.8,-0.3,0.9,-0.6,-0.1,1.1
BRI1,0.9,1.2,0.4,-0.7,-0.2,-0.5
BZR1,-1.1,0.3,1.8,-1.0,0.1,1.2
PRE1,0.4,1.0,1.4,-0.3,0.6,0.9
SAUR19,-0.6,0.2,1.1,-0.8,-0.4,0.7
CHS,1.5,1.8,0.6,0.9,1.2,-0.3
CHI,1.1,1.4,0.3,0.6,0.9,-0.5
F3H,0.8,1.2,0.1,0.4,0.8,-0.6
FLS,-0.9,-0.4,0.7,-1.1,-0.6,0.4
ANS,1.3,1.6,0.2,0.7,1.0,-0.4
ANR,-1.4,-0.8,0.3,-1.2,-0.5,0.1

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 02-heatmap.csv
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("02-heatmap.csv", index_col="gene")
fig, ax = plt.subplots(figsize=(6.4, 6.0))
im = ax.imshow(df.values, cmap="YlGnBu", vmin=-2, vmax=2, aspect="auto")
ax.set_xticks(range(df.shape[1]), df.columns, rotation=40, ha="right")
ax.set_yticks(range(df.shape[0]), df.index)
ax.grid(False)
for i in range(df.shape[0]):
    for j in range(df.shape[1]):
        ax.text(j, i, "%.1f" % df.values[i, j], ha="center", va="center", fontsize=7.5)
fig.colorbar(im, ax=ax, shrink=0.75, label="row z-score of TPM")
fig.savefig("02-heatmap.png", dpi=200, bbox_inches="tight")

gene,DWF4,BZR1,SAUR19,CHS,CHI,F3H,ANS,FLS
DWF4,1.00,0.86,0.82,-0.21,-0.18,-0.15,-0.24,0.35
BZR1,0.86,1.00,0.78,-0.12,-0.16,-0.09,-0.19,0.42
SAUR19,0.82,0.78,1.00,-0.18,-0.22,-0.11,-0.27,0.30
CHS,-0.21,-0.12,-0.18,1.00,0.93,0.88,0.91,-0.46
CHI,-0.18,-0.16,-0.22,0.93,1.00,0.85,0.89,-0.41
F3H,-0.15,-0.09,-0.11,0.88,0.85,1.00,0.86,-0.38
ANS,-0.24,-0.19,-0.27,0.91,0.89,0.86,1.00,-0.44
FLS,0.35,0.42,0.30,-0.46,-0.41,-0.38,-0.44,1.00

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 03-corr.csv
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("03-corr.csv", index_col="gene")
M = df.values
fig, ax = plt.subplots(figsize=(6.6, 5.6))
im = ax.imshow(M, cmap="RdBu_r", vmin=-1, vmax=1)
ax.set_xticks(range(len(df)), df.columns, rotation=40, ha="right")
ax.set_yticks(range(len(df)), df.index)
ax.grid(False)
for i in range(M.shape[0]):
    for j in range(M.shape[1]):
        ax.text(j, i, "%.2f" % M[i, j], ha="center", va="center", fontsize=7.5,
                color="white" if abs(M[i, j]) > 0.6 else "#222222")
fig.colorbar(im, ax=ax, shrink=0.8, label="Pearson r")
fig.savefig("03-corr-heatmap.png", dpi=200, bbox_inches="tight")

sample,DWF4_TPM,BZR1_TPM
S1,12,15
S2,24,31
S3,31,38
S4,38,44
S5,45,41
S6,52,63
S7,58,57
S8,64,74
S9,71,70
S10,77,85
S11,84,80
S12,90,97

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 04-pearson.csv
import numpy as np
import pandas as pd
from scipy import stats
import matplotlib.pyplot as plt

df = pd.read_csv("04-pearson.csv")
x, y = df["DWF4_TPM"], df["BZR1_TPM"]
r, p = stats.pearsonr(x, y)
fig, ax = plt.subplots(figsize=(5.6, 5.0))
ax.scatter(x, y, s=55, color="#4C72B0", edgecolors="white", zorder=3)
k, b = np.polyfit(x, y, 1)
xs = np.array([x.min(), x.max()])
ax.plot(xs, k * xs + b, "--", color="#C44E52", lw=1.8, label="linear fit")
ax.text(0.05, 0.92, "Pearson r = %.2f\\np = %.1e" % (r, p), transform=ax.transAxes,
        va="top", bbox=dict(boxstyle="round,pad=0.35", fc="#F5F7FA", ec="0.7"))
ax.set_xlabel("DWF4 expression (TPM)")
ax.set_ylabel("BZR1 expression (TPM)")
ax.legend(loc="lower right")
fig.savefig("04-pearson.png", dpi=200, bbox_inches="tight")

panel,x,y
A,1,2
A,2,4
A,3,8
A,4,16
A,5,32
A,6,64
A,7,128
A,8,256
A,9,512
A,10,1024
A,11,2048
A,12,4096
B,7.97,16.49
B,4.95,8.87
B,8.73,17.90
B,7.28,13.40
B,1.85,4.75
B,9.78,19.50
B,7.85,15.48
B,8.07,15.33
B,2.15,5.77
B,5.05,9.92
B,4.34,8.16
B,9.34,18.26
B,6.79,14.23
B,8.40,17.25
B,11.00,1.00

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 05-spearman.csv
import pandas as pd
from scipy import stats
import matplotlib.pyplot as plt

df = pd.read_csv("05-spearman.csv")
fig, axes = plt.subplots(1, 2, figsize=(10.4, 4.6))
for ax, panel, title in [(axes[0], "A", "A. Monotone but nonlinear"),
                         (axes[1], "B", "B. One leverage outlier")]:
    sub = df[df.panel == panel]
    r, _ = stats.pearsonr(sub.x, sub.y)
    rho, _ = stats.spearmanr(sub.x, sub.y)
    ax.scatter(sub.x, sub.y, s=50, color="#4C72B0", edgecolors="white", zorder=3)
    ax.set_title(title)
    ax.text(0.05, 0.9, "Pearson r = %.2f\\nSpearman rho = %.2f" % (r, rho),
            transform=ax.transAxes, va="top", fontsize=10,
            bbox=dict(boxstyle="round,pad=0.35", fc="#F5F7FA", ec="0.7"))
# B 组末行 (11.00, 1.00) 即红色杠杆离群点，会把虚线回归线拽到几乎躺平
fig.savefig("05-spearman.png", dpi=200, bbox_inches="tight")

tissue,tpm
Root,2
Root,4
Root,5
Root,7
Root,9
Root,11
Root,13
Root,16
Root,19
Root,23
Root,28
Root,34
Stem,8
Stem,12
Stem,15
Stem,18
Stem,21
Stem,24
Stem,27
Stem,30
Stem,34
Stem,39
Stem,45
Stem,52
Leaf,15
Leaf,22
Leaf,28
Leaf,34
Leaf,40
Leaf,47
Leaf,55
Leaf,63
Leaf,72
Leaf,84
Leaf,97
Leaf,110
Seed,40
Seed,55
Seed,68
Seed,80
Seed,95
Seed,110
Seed,128
Seed,145
Seed,168
Seed,190
Seed,215
Seed,245

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 06-boxplot.csv
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("06-boxplot.csv")
groups = df.tissue.unique()
data = [df[df.tissue == g].tpm.values for g in groups]
CAT = ["#4C72B0", "#DD8452", "#55A868", "#C44E52"]
fig, ax = plt.subplots(figsize=(6.2, 4.8))
bp = ax.boxplot(data, tick_labels=groups, patch_artist=True, widths=0.55,
                medianprops=dict(color="black", lw=1.6))
for patch, col in zip(bp["boxes"], CAT):
    patch.set_facecolor(col)
    patch.set_alpha(0.65)
rng = np.random.default_rng(1)
for i, d in enumerate(data, 1):
    ax.scatter(rng.normal(i, 0.05, len(d)), d, s=10, color="0.25",
               alpha=0.6, zorder=3)
ax.set_ylabel("Expression (TPM)")
fig.savefig("06-boxplot.png", dpi=200, bbox_inches="tight")

# 数据：与 2.6 节的 06-boxplot.csv 完全相同
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("06-boxplot.csv")
groups = df.tissue.unique()
data = [df[df.tissue == g].tpm.values for g in groups]
CAT = ["#4C72B0", "#DD8452", "#55A868", "#C44E52"]
fig, ax = plt.subplots(figsize=(6.2, 4.8))
vp = ax.violinplot(data, showextrema=False)
for body, col in zip(vp["bodies"], CAT):
    body.set_facecolor(col)
    body.set_alpha(0.65)
for i, d in enumerate(data, 1):
    q1, med, q3 = np.percentile(d, [25, 50, 75])
    ax.vlines(i, q1, q3, color="black", lw=4)
    ax.scatter(i, med, s=28, color="white", edgecolors="black", zorder=3)
ax.set_xticks(range(1, 5), groups)
ax.set_ylabel("Expression (TPM)")
fig.savefig("07-violin.png", dpi=200, bbox_inches="tight")

set,size
drought_only,26
salt_only,20
cold_only,30
drought_and_salt,12
drought_and_cold,9
salt_and_cold,11
all_three,5
drought_total,42
salt_total,38
cold_total,45

绘图代码（Python，独立可运行）

# 数据：区间数字取自上方 CSV，几何为三个半透明圆 + 区间文字
import matplotlib.pyplot as plt
from matplotlib.patches import Circle

CAT = ["#4C72B0", "#DD8452", "#55A868"]
fig, ax = plt.subplots(figsize=(6.2, 5.6))
for (x0, y0), col in zip([(-0.1, 0), (0.95, 0), (0.42, 0.9)], CAT):
    ax.add_patch(Circle((x0, y0), 0.95, facecolor=col, alpha=0.35,
                        edgecolor=col, lw=1.6))
for t, x0, y0 in [("drought 26", -0.72, -0.35), ("salt 20", 1.58, -0.35),
                  ("cold 30", 0.42, 1.42), ("12", 0.43, -0.62),
                  ("9", -0.08, 0.45), ("11", 0.94, 0.45), ("5", 0.42, 0.12)]:
    ax.text(x0, y0, t, ha="center", va="center", fontsize=9.5)
ax.set_xlim(-1.5, 2.4)
ax.set_ylim(-1.6, 1.9)
ax.axis("off")
fig.savefig("08-venn-treatment.png", dpi=200, bbox_inches="tight")

species,total_genes,one2one_orthologs
Fagopyrum_tataricum,33400,14200
Arabidopsis_thaliana,27600,14200

绘图代码（Python，独立可运行）

# 数据：两物种基因数与一对一同源数，两个圆即可
import matplotlib.pyplot as plt
from matplotlib.patches import Circle

fig, ax = plt.subplots(figsize=(6.0, 4.6))
ax.add_patch(Circle((-0.55, 0), 1.15, facecolor="#4C72B0", alpha=0.35,
                    edgecolor="#4C72B0", lw=1.6))
ax.add_patch(Circle((0.55, 0), 1.15, facecolor="#55A868", alpha=0.35,
                    edgecolor="#55A868", lw=1.6))
ax.text(-1.35, 0, "19,200", ha="center", fontsize=12)
ax.text(0.0, 0, "14,200", ha="center", fontsize=12, fontweight="bold")
ax.text(1.35, 0, "13,400", ha="center", fontsize=12)
ax.text(-1.1, 1.35, "F. tataricum", ha="center", fontsize=10.5, color="#4C72B0")
ax.text(1.1, 1.35, "A. thaliana", ha="center", fontsize=10.5, color="#55A868")
ax.set_xlim(-2.1, 2.1)
ax.set_ylim(-1.5, 1.9)
ax.axis("off")
fig.savefig("09-venn-ortholog.png", dpi=200, bbox_inches="tight")

DEG,DEP,PHOS,DAM,size
1,0,0,0,402
0,1,0,0,121
0,0,0,1,108
1,1,0,0,96
0,0,1,0,74
1,0,0,1,58
1,0,1,0,41
0,1,0,1,27
0,1,1,0,33
1,1,1,0,18
1,1,1,1,9

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 10-upset.csv（0/1 成员矩阵 + 交集大小）
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("10-upset.csv").sort_values("size", ascending=False).reset_index(drop=True)
sets = ["DEG", "DEP", "PHOS", "DAM"]
fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(8.6, 5.2), sharex=True,
                               gridspec_kw=dict(height_ratios=[3, 1.15], hspace=0.06))
ax1.bar(df.index, df["size"], color="#4C72B0", width=0.62)
for i, v in df["size"].items():
    ax1.text(i, v + 8, str(int(v)), ha="center", fontsize=9)
ax1.set_ylabel("Intersection size")
for i, row in df.iterrows():
    on = [s for s in sets if row[s] == 1]
    ax2.plot([i, i], [sets.index(on[0]), sets.index(on[-1])], color="black", lw=1)
    for s in sets:
        ax2.scatter(i, sets.index(s), s=52 if s in on else 14,
                    color="black" if s in on else "0.85")
ax2.set_yticks(range(4), sets)
ax2.invert_yaxis()
ax2.set_xticks([])
ax2.grid(False)
fig.savefig("10-upset.png", dpi=200, bbox_inches="tight")

gene,L1,L2,L3,L4,L5,L6
DWF4,3.1,3.4,3.0,12.5,13.1,12.8
CHS,88.4,95.1,91.7,54.2,57.8,55.6
CHI,12.2,13.0,12.6,9.8,10.4,10.1
F3H,25.6,27.8,26.3,19.4,20.9,20.1
ANS,1.8,2.1,1.9,6.7,7.0,6.9
PAL,41.0,44.6,42.8,30.5,32.2,31.4
BZR1,8.5,9.1,8.8,11.2,10.8,11.5
DET2,5.2,5.0,5.4,8.9,9.4,9.1
FLS,15.9,17.2,16.4,12.1,12.9,12.5
ANR,0.9,1.1,1.0,3.2,3.5,3.3

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 11-tpm.csv（行=基因，列=文库）
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("11-tpm.csv", index_col="gene")
fig, ax = plt.subplots(figsize=(6.6, 4.8))
bp = ax.boxplot([df[c] for c in df.columns], tick_labels=df.columns,
                patch_artist=True, widths=0.55)
colors = ["#4C72B0", "#DD8452", "#55A868", "#C44E52", "#8172B3", "#937860"]
for patch, col in zip(bp["boxes"], colors):
    patch.set_facecolor(col)
    patch.set_alpha(0.6)
ax.set_yscale("log")
ax.set_ylabel("TPM (log scale)")
fig.savefig("11-tpm-dist.png", dpi=200, bbox_inches="tight")

sample,group,PC1,PC2
Z1_r1,Stage_I,-3.2,-0.8
Z1_r2,Stage_I,-2.8,0.4
Z1_r3,Stage_I,-3.5,0.1
Z1_r4,Stage_I,-2.4,-0.3
Z2_r1,Stage_II,0.1,1.8
Z2_r2,Stage_II,0.6,2.3
Z2_r3,Stage_II,-0.4,1.2
Z2_r4,Stage_II,0.3,2.8
Z3_r1,Stage_III,3.1,-0.5
Z3_r2,Stage_III,3.6,0.6
Z3_r3,Stage_III,2.7,-1.1
Z3_r4,Stage_III,3.3,0.2

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 12-pca.csv
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("12-pca.csv")
CAT = ["#4C72B0", "#DD8452", "#55A868"]
fig, ax = plt.subplots(figsize=(6.6, 5.6))
for i, g in enumerate(df.group.unique()):
    sub = df[df.group == g]
    ax.scatter(sub.PC1, sub.PC2, s=60, color=CAT[i], edgecolors="white", label=g)
for gene, lx, ly in [("DWF4", 0.85, 0.31), ("CHS", -0.72, 0.45),
                     ("FLS", 0.81, -0.38), ("ANS", -0.66, 0.52)]:
    ax.annotate("", xy=(lx * 3.4, ly * 3.4), xytext=(0, 0),
                arrowprops=dict(arrowstyle="->", color="#8172B3", lw=1.4))
    ax.text(lx * 3.75, ly * 3.75, gene, color="#8172B3", fontsize=9,
            ha="center", fontweight="bold")
ax.axhline(0, color="0.8", lw=0.8)
ax.axvline(0, color="0.8", lw=0.8)
ax.set_xlabel("PC1 (52.3% variance)")
ax.set_ylabel("PC2 (21.7% variance)")
ax.legend()
fig.savefig("12-pca.png", dpi=200, bbox_inches="tight")

cell,cluster,UMAP1,UMAP2
c1,C1,-4.00,2.28
c2,C1,-4.26,1.15
c3,C1,-4.43,1.06
c4,C1,-3.94,3.27
c5,C1,-4.47,1.41
c6,C1,-3.53,2.34
c7,C1,-3.90,1.12
c8,C1,-4.03,2.66
c9,C1,-5.28,1.57
c10,C1,-5.81,0.77
c11,C1,-5.75,1.78
c12,C1,-5.20,2.26
c13,C2,3.15,2.82
c14,C2,0.61,2.49
c15,C2,2.95,3.11
c16,C2,1.55,2.55
c17,C2,2.07,2.23
c18,C2,4.01,2.23
c19,C2,2.97,3.84
c20,C2,2.45,2.89
c21,C2,3.10,3.06
c22,C2,1.84,3.07
c23,C2,4.29,1.53
c24,C2,3.82,3.11
c25,C3,-0.61,-2.10
c26,C3,0.72,-5.14
c27,C3,0.07,-3.45
c28,C3,-0.18,-3.35
c29,C3,-0.06,-3.37
c30,C3,1.37,-4.64
c31,C3,0.19,-4.44
c32,C3,0.12,-5.13
c33,C3,-0.55,-4.19
c34,C3,0.85,-2.91
c35,C3,-1.26,-4.75
c36,C3,0.61,-5.89

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 13-umap.csv
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("13-umap.csv")
CAT = ["#4C72B0", "#DD8452", "#55A868"]
fig, ax = plt.subplots(figsize=(6.2, 5.4))
for i, c in enumerate(df.cluster.unique()):
    sub = df[df.cluster == c]
    ax.scatter(sub.UMAP1, sub.UMAP2, s=46, color=CAT[i], edgecolors="white", label=c)
ax.set_xlabel("UMAP 1 (arbitrary units)")
ax.set_ylabel("UMAP 2 (arbitrary units)")
ax.legend(title="cluster")
fig.savefig("13-umap.png", dpi=200, bbox_inches="tight")

# 数据：与 2.2 节的 02-heatmap.csv 完全相同
import pandas as pd
import matplotlib.pyplot as plt
from scipy.cluster.hierarchy import linkage, dendrogram

df = pd.read_csv("02-heatmap.csv", index_col="gene")
Z = linkage(df.values, method="average", metric="euclidean")
fig = plt.figure(figsize=(6.8, 7.0))
gs = fig.add_gridspec(2, 1, height_ratios=[1, 2.6], hspace=0.04)
axd = fig.add_subplot(gs[0])
dd = dendrogram(Z, ax=axd, no_labels=True, color_threshold=0,
                link_color_func=lambda k: "0.4")
axd.set_yticks([])
axd.grid(False)
axh = fig.add_subplot(gs[1])
order = dd["leaves"]
im = axh.imshow(df.values[order], cmap="YlGnBu", vmin=-2, vmax=2, aspect="auto")
axh.set_xticks(range(df.shape[1]), df.columns, rotation=40, ha="right")
axh.set_yticks(range(len(order)), [df.index[i] for i in order], fontsize=8.5)
axh.grid(False)
fig.colorbar(im, ax=axh, shrink=0.7, label="row z-score of TPM")
fig.savefig("14-dendro-heatmap.png", dpi=200, bbox_inches="tight")

pathway,enrich_factor,count,padj
Flavonoid biosynthesis,0.082,24,0.00012
Brassinosteroid biosynthesis,0.096,11,0.0031
Phenylpropanoid biosynthesis,0.064,31,0.00040
Plant hormone signal transduction,0.041,38,0.0021
Cutin suberine and wax biosynthesis,0.075,9,0.012
Starch and sucrose metabolism,0.037,22,0.0068
Circadian rhythm - plant,0.052,8,0.038
Amino sugar metabolism,0.028,15,0.041

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd

df = pd.read_csv("kegg.csv")
y = -np.log10(df.padj)

fig, ax = plt.subplots(figsize=(8.8, 5.8))
sc = ax.scatter(df.enrich_factor, y, s=df["count"] * 28, c=y,
                cmap="viridis", alpha=0.8, edgecolors="0.3")
for name, xe, ye in zip(df.pathway, df.enrich_factor, y):
    ax.annotate(name[:24], (xe, ye), (xe + 0.004, ye + 0.14), fontsize=8)
ax.set_xlabel("enrichment factor (DEG ratio / background ratio)")
ax.set_ylabel("-log10 (padj)")
fig.colorbar(sc, label="-log10(padj)")
fig.savefig("16-kegg-bubble.png", dpi=200, bbox_inches="tight")

pathway,enrich_factor,count,padj
Flavonoid biosynthesis,0.082,24,0.00012
Brassinosteroid biosynthesis,0.096,11,0.0031
Phenylpropanoid biosynthesis,0.064,31,0.00040
Plant hormone signal transduction,0.041,38,0.0021
Cutin suberine and wax biosynthesis,0.075,9,0.012
Starch and sucrose metabolism,0.037,22,0.0068
Circadian rhythm - plant,0.052,8,0.038
Amino sugar metabolism,0.028,15,0.041

# 数据：将上方 CSV 保存为 17-kegg.csv
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("17-kegg.csv")
neglog = -np.log10(df.padj)
order = df["count"].sort_values(ascending=False).index
d = df.loc[order]
fig, ax = plt.subplots(figsize=(8.4, 5.2))
ax.barh(range(len(d)), d["count"], color=plt.cm.viridis(neglog[order] / neglog.max()),
        edgecolor="0.3", linewidth=0.5)
ax.set_yticks(range(len(d)), d.pathway, fontsize=9)
for i, v in enumerate(d["count"]):
    ax.text(v + 0.4, i, str(v), va="center", fontsize=8.5)
ax.set_xlabel("DEG count mapped to pathway")
sm = plt.cm.ScalarMappable(cmap="viridis",
                           norm=plt.Normalize(0, neglog.max()))
fig.colorbar(sm, ax=ax, shrink=0.8, label="-log10(padj)")
fig.savefig("17-kegg-bar.png", dpi=200, bbox_inches="tight")

gene,go_term
DWF4,BR_biosynthesis
BZR1,BR_signaling
BZR1,cell_elongation
SAUR19,cell_elongation
CHS,flavonoid_biosynthesis
F3H,flavonoid_biosynthesis
FLS,flavonoid_biosynthesis
ANS,flavonoid_biosynthesis
ANS,anthocyanin_biosynthesis
ANR,anthocyanin_biosynthesis

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 15-chord.csv（基因-功能关联表）
# 没有现成弦图包时用贝塞尔带手绘：左弧基因、右弧功能、带宽正比关联数
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("15-chord.csv")
genes = list(dict.fromkeys(df.gene))
terms = list(dict.fromkeys(df.go_term))
rel = list(zip(df.gene, df.go_term))
CAT = ["#4C72B0", "#DD8452", "#55A868", "#C44E52", "#8172B3", "#937860",
       "#DA8BC3", "#8C8C8C"]
gcol = {g: CAT[i] for i, g in enumerate(genes)}

def slots(order, a0, a1, gap):
    deg = {n: sum(1 for g, t in rel if n in (g, t)) for n in order}
    avail = (a1 - a0) - gap * (len(order) - 1)
    out, cur = {}, a0
    for n in order:
        w = avail * deg[n] / sum(deg.values())
        out[n] = (cur, cur + w)
        cur += w + gap
    return out, deg

gs, gd = slots(genes, np.deg2rad(95), np.deg2rad(265), np.deg2rad(6))
ts, td = slots(terms, np.deg2rad(-85), np.deg2rad(85), np.deg2rad(6))
tsl = np.linspace(0, 1, 40)
fig, ax = plt.subplots(figsize=(8.6, 6.0))
ax.set_aspect("equal")
ax.axis("off")
gu = {g: 0.0 for g in genes}
tu = {t: 0.0 for t in terms}
for g, t in rel:
    gw = (gs[g][1] - gs[g][0]) / gd[g]
    tw = (ts[t][1] - ts[t][0]) / td[t]
    a0 = gs[g][0] + gu[g] * gw
    a1 = a0 + gw
    b0 = ts[t][0] + tu[t] * tw
    b1 = b0 + tw
    gu[g] += gw
    tu[t] += tw
    P = lambda a: np.array([np.cos(a), np.sin(a)])
    p0, p1, q0, q1 = P(a0), P(a1), P(b1), P(b0)
    top = ((1 - tsl) ** 3)[:, None] * p0 + (3 * (1 - tsl) ** 2 * tsl)[:, None] * p0 * 0.25 \\
        + (3 * (1 - tsl) * tsl ** 2)[:, None] * q0 * 0.25 + (tsl ** 3)[:, None] * q0
    bot = ((1 - tsl) ** 3)[:, None] * p1 + (3 * (1 - tsl) ** 2 * tsl)[:, None] * p1 * 0.25 \\
        + (3 * (1 - tsl) * tsl ** 2)[:, None] * q1 * 0.25 + (tsl ** 3)[:, None] * q1
    poly = np.vstack([top, bot[::-1]])
    ax.fill(poly[:, 0], poly[:, 1], color=gcol[g], alpha=0.35, lw=0)
for g in genes:
    arc = np.linspace(gs[g][0], gs[g][1], 30)
    ax.plot(np.cos(arc), np.sin(arc), color=gcol[g], lw=7, solid_capstyle="butt")
    am = gs[g][0] / 2 + gs[g][1] / 2
    ax.text(1.12 * np.cos(am), 1.12 * np.sin(am), g, ha="right", va="center",
            fontsize=10, color=gcol[g], fontweight="bold")
for t in terms:
    arc = np.linspace(ts[t][0], ts[t][1], 30)
    ax.plot(np.cos(arc), np.sin(arc), color="0.25", lw=7, solid_capstyle="butt")
    am = ts[t][0] / 2 + ts[t][1] / 2
    ax.text(1.12 * np.cos(am), 1.12 * np.sin(am), t.replace("_", " "),
            ha="left", va="center", fontsize=10)
fig.savefig("15-go-chord.png", dpi=200, bbox_inches="tight")

gene,score,in_set
CHS,3.42,1
F3H,3.10,1
CHI,2.87,1
ANS,2.55,1
FLS,2.31,1
ANR,1.98,1
PAL,1.76,1
4CL,1.52,1
ACT7,1.31,0
EF1a,1.10,0
TUB6,0.92,0
RBCS,0.74,0
LHCB,0.55,0
SOD,-0.32,0
APX1,-0.51,0
CAT2,-0.77,0
NCED3,-0.98,0
PYL4,-1.24,0
ABF2,-1.47,0
WRKY1,-1.83,0

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 18-gsea.csv（score 降序，in_set=1 为成员）
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("18-gsea.csv")
s = df.score.values
hit = df.in_set.values.astype(bool)
w = np.abs(s) * hit / np.abs(s[hit]).sum()
run = np.cumsum(np.where(hit, w, -1.0 / (len(s) - hit.sum())))
peak = int(np.argmax(run))
ranks = np.arange(1, len(s) + 1)
fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(8.6, 5.4), sharex=True,
                               gridspec_kw=dict(height_ratios=[2.6, 1], hspace=0.08))
ax1.plot(ranks, run, color="#4C72B0", lw=2)
ax1.fill_between(ranks, run, 0, where=run > 0, color="#4C72B0", alpha=0.15)
ax1.axhline(0, color="0.6", lw=0.8)
ax1.axvline(peak, color="0.4", ls="--", lw=1)
ax1.annotate("peak ES = %.2f at rank %d" % (run[peak - 1], peak),
             (peak, run[peak - 1]), (peak - 7.5, run[peak - 1] - 0.02), fontsize=9)
ax1.set_ylabel("enrichment score (running sum)")
ax2.bar(ranks[hit], 1, color="black", width=0.7)
ax2.bar(ranks[~hit], 1, color="0.85", width=0.7)
ax2.set_xlabel("genes ranked by score")
fig.savefig("18-gsea.png", dpi=200, bbox_inches="tight")

sample,label,model_A,model_B,model_C
P1,1,0.97,0.91,0.58
P2,1,0.94,0.83,0.22
P3,1,0.91,0.88,0.71
P4,1,0.88,0.64,0.35
P5,1,0.85,0.79,0.62
P6,1,0.81,0.72,0.15
P7,1,0.76,0.58,0.83
P8,1,0.71,0.45,0.48
P9,1,0.65,0.81,0.29
P10,1,0.60,0.52,0.66
N1,0,0.42,0.49,0.77
N2,0,0.38,0.35,0.11
N3,0,0.33,0.61,0.52
N4,0,0.29,0.28,0.05
N5,0,0.25,0.40,0.90
N6,0,0.21,0.22,0.38
N7,0,0.17,0.31,0.44
N8,0,0.13,0.18,0.59
N9,0,0.09,0.12,0.03
N10,0,0.05,0.26,0.74

绘图代码（Python，独立可运行）

# 数据：将上方 CSV 保存为 19-roc.csv（label=1 阳性，列为各模型打分）
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("19-roc.csv")
label = df.label.values
fig, ax = plt.subplots(figsize=(5.8, 5.6))
for col in ["model_A", "model_B", "model_C"]:
    order = np.argsort(-df[col].values)
    lab = label[order]
    tpr = np.r_[0, np.cumsum(lab) / lab.sum()]
    fpr = np.r_[0, np.cumsum(1 - lab) / (1 - lab).sum()]
    auc = np.trapezoid(tpr, fpr)
    ax.plot(fpr, tpr, lw=2, label="%s  AUC = %.2f" % (col, auc))
ax.plot([0, 1], [0, 1], color="0.6", ls="--", lw=1.2, label="chance")
ax.set_xlabel("false positive rate")
ax.set_ylabel("true positive rate")
ax.legend(loc="lower right")
fig.savefig("19-roc.png", dpi=200, bbox_inches="tight")

patient,group,time_weeks,event
T1,treated,4,1
T2,treated,7,1
T3,treated,10,0
T4,treated,13,1
T5,treated,16,0
T6,treated,19,1
T7,treated,22,1
T8,treated,25,0
T9,treated,28,0
T10,treated,30,0
C1,control,2,1
C2,control,4,1
C3,control,5,0
C4,control,7,1
C5,control,9,1
C6,control,11,1
C7,control,12,0
C8,control,14,1
C9,control,16,1
C10,control,18,0