Python

完成品

以下のようなVolcano plotをPythonで書いていきます。

Script

以下をコピペで上記のグラフが出てきます。

import pandas as pd

url="https://raw.githubusercontent.com/vappiah/bioinfoscripts/main/testvolcano.csv"
c=pd.read_csv(url)

def volcanoplot(de, foldchange, pvalue, featurename, label=10):
    import numpy as np
    import pandas as pd
    import matplotlib
    import matplotlib.pyplot as plt
    import seaborn as sns
    import joblib
    from adjustText import adjust_text

    topfeature = []
    de = de.sort_values(pvalue)
    de["logpvalue"] = np.log10(de[pvalue]) * -1
    conditionlist = [
        (de[foldchange] > 1) & (de[pvalue] < 0.1),
        (de[foldchange] < -1) & (de[pvalue] < 0.1),
    ]

    choicelist = ['UP', 'DOWN']
    de['diffexpressed'] = np.select(conditionlist, choicelist, default='NaN')
    up = de[de["diffexpressed"]=="UP"]
    up = up.iloc[0:label, ]
    topfeature = up[featurename]
    down = de[de["diffexpressed"]=="DOWN"]
    down = down.iloc[0:label, ]
    topfeature = topfeature.append(down[featurename])


    color_dict = dict({'UP':'red',
                       'DOWN':'blue',
                       'NaN':'black'
                      })

    ax = sns.scatterplot(data=de, x=foldchange, y="logpvalue", hue="diffexpressed", palette=color_dict, markers='o', s=10)

    de = de[de[featurename].isin(topfeature)]
    texts = []
    for x, y, s in zip(de[foldchange], de["logpvalue"], de[featurename]):
            texts.append(plt.text(x, y, s))
    adjust_text(texts)
    return (ax)

myvolcanoplot.volcanoplot(c, "log2FC", "p-value", "GeneNames", label=5)

Script説明

データフレームのインポート

Volcanoplotとは直接関係ないですが、データは以下を使用しました。
https://github.com/vappiah/bioinfoscripts/blob/main/testvolcano.csv
pandasでcsvをcに入れていきます。

import pandas as pd

url="https://raw.githubusercontent.com/vappiah/bioinfoscripts/main/testvolcano.csv"
c=pd.read_csv(url)

GeneNames	log2FC	p-value
0LOC_Os09g01000.1	-1.886539	1.250000e-55
…	…	…

Volcanoplotでは、以下のカラムが必要になります。
①遺伝子名
②Log Fold Change
③p-value またはq-value

必要なModuleをインポート

    import numpy as np
    import pandas as pd
    import matplotlib
    import matplotlib.pyplot as plt
    import seaborn as sns
    import joblib
    from adjustText import adjust_text