将每一个OTU拆分成不同的分类级别

import pandas as pd

# 读取OTU表
otu_df = pd.read_csv('C:\\Users\\Administrator\\Desktop\\1.txt', header=None)

# 创建一个空的DataFrame来存储结果
df = pd.DataFrame(columns=["OTUID", "Kingdom", "Phylum", "Class", "Order", "Family", "Genus", "Species"])

for i, otu in enumerate(otu_df[0]):
    # 使用"|"拆分每个OTU
    split_otu = otu.split("|")
    
    # 初始化一个字典来存储每个分类级别
    tax_dict = {"Kingdom": "Unassigned", "Phylum": "Unassigned", "Class": "Unassigned",
                "Order": "Unassigned", "Family": "Unassigned", "Genus": "Unassigned", 
                "Species": "Unassigned"}
    
    # 遍历拆分后的列表，并根据前缀将每个部分分配到相应的分类级别
    for part in split_otu:
        if part.startswith("k__"):
            tax_dict["Kingdom"] = part
        elif part.startswith("p__"):
            tax_dict["Phylum"] = part
        elif part.startswith("c__"):
            tax_dict["Class"] = part
        elif part.startswith("o__"):
            tax_dict["Order"] = part
        elif part.startswith("f__"):
            tax_dict["Family"] = part
        elif part.startswith("g__"):
            tax_dict["Genus"] = part
        elif part.startswith("s__"):
            tax_dict["Species"] = part
    
    # 删除分类级别前缀（如"k__"）
    tax_dict = {key: value.split("__")[1] if "__" in value else value for key, value in tax_dict.items()}
    
    # 将结果添加到DataFrame中
    df.loc[i] = ["ASV_" + str(i+1)] + list(tax_dict.values())

# 将结果保存为新的txt文件
df.to_csv('C:\\Users\\Administrator\\Desktop\\annotated_otu_table.txt', sep='\t', index=False)