Pandas Drop 布尔值

Question

我们得到了一个 file.tsv，我们需要构建一个函数。其中之一是如果一列（此处称为“low_confidence_variant”）= True，则删除每一行。我在某种程度上为这部分而奋斗。另外，有什么优化建议吗？根据结果，我们需要制作一个迈阿密图。这是我到目前为止所做的。任何提示都会有用；

import numpy as np

import pandas as pd

import matplotlib.pyplot as plt

def read_file(file, chromosome):

df = pd.read_csv(file, sep='\t', usecols=['chromosome', 'position', 'pval', 'low_confidence_variant'])

df.drop(['low_confidence_variant'], True)

df.dropna()

sub_data = df.replace({'pval': 0}, 1e-274)

sub_data['log10'] = -np.log10(sub_data['pval'])

chr_group = sub_data.groupby(['chromosome'])

chromosome = chr_group.get_group(chromosome)

return chromosome

df1 = read_file('vitamin_d.females.tsv.gz', 1)

df2 = read_file('vitamin_d.males.tsv.gz', 1)

xa = df2['position']

ya = df2['log10']

xb = df1['position']

yb = df1['log10'] * -1

fig, (ax1, ax2) = plt.subplots(nrows=2, ncols=1, sharex=True, figsize=(12, 4))

ax1.scatter(xa, ya, s=1, c="tab:blue")

ax1.set_ylabel('males $\it{-log_{10}(pval)}$')

ax1.set_title('vitamin D (nmol/L)', fontweight='bold')

ax1.axhline(-np.log10(5*10**-8), c ='darkgray', ls='--')

ax2.scatter(xb, yb, s=1, c="tab:blue")

ax2.set_ylabel('females $\it{log_{10}(pval)}$')

ax2.axhline(np.log10(5*10**-8), c ='darkgray', ls='--')

plt.xlabel('Chromosome 1 positions')

plt.subplots_adjust(hspace=.0)

plt.show()

fig.savefig(fname='miami.png', dpi=300, bbox_inches='tight', format='png')

Answer 1

我有点不确定你的意思。Say Df = A   B   low_confidence_variant10  20    True2    4    False6    0    FalseSo after deleting the rows with low_confidence_variant = True, you should havedf = A   B     low_confidence_variant2   4      False6   0      False正确的？如果这就是您的意思：### Add below linedf = df[df['low_confidence_variant'] != True]并删除这一行### Delete this line from the codedf.drop(['low_confidence_variant'], True)您所做的就是删除整个列本身。