DataFrame选择行

Question

嗨我试图从我的数据框中选择一个特定的行，通过使用.loc从基因名称中取出行及其相应数据，我无法通过数字进行拼接，因为每个数据框都有不同的行。我得到的错误是一个关键的错误，虽然我不知道为什么。我已经使用type（df3）来确认我的数据仍然是一个数据框。

for name in filenames: 
      df1 = pd.read_table(name, index_col=None, header = None, names =["Composite_Element_REF", "Beta_value", "Gene_Symbol", "Chromosome", "Genomic_Coordinate"], sep = "\t", skiprows= 2) # sep will seperate anything with space so it goes to the appropriate column, names is to lable column 
      df1 = df1.dropna() 
      df2 = df1[["Composite_Element_REF", "Beta_value","Gene_Symbol"]] 
      df3 = df2.rename(columns= {"Composite_Element_REF": "Composite_Element_REF "+ str(index), "Beta_value":"Beta_value "+str(index),"Gene_Symbol":"Gene_Symbol " +str(index)}, inplace=False) 
      #NR5A2 = df3.iloc[[2]] 
      NR5A2 = df3.loc[("Gene_Symbol " +str(index), "NR5A2"), : ] 

      index = index +1 

KeyError         Traceback (most recent call last) 
<ipython-input-78-5075955b850d> in <module>() 
    12  df3 = df2.rename(columns= {"Composite_Element_REF": "Composite_Element_REF "+ str(index), "Beta_value":"Beta_value "+str(index),"Gene_Symbol":"Gene_Symbol " +str(index)}, inplace=False) 
    13  #NR5A2 = df3.iloc[[2]] 
---> 14  NR5A2[str(index)] = df3.loc[("Gene_Symbol " +str(index), "NR5A2"), : ] 
    15 
    16  index = index +1 

Answer 1

要使用.loc你必须设置一些列作为索引。

但我认为你会被罚款与

df3[df3["Gene_Symbol " +str(index)] == 'NR5A2']