从.txt fi创建多个嵌套字典

Question

我正在尝试创建一个由多个词典组成的词典。我是从一个.txt文件创建的：

chrY 6 8 +
chrY 3 5 +
chrX 10 11 +
chrX 13 15 -

我想要的结果是：

{'chrY': {'+' : {'start': [3 , 6], 'end': [5, 8]}}, 'chrX': {'+' : {'start': [10], 'end': [11]} , '-': {'start' : [13], 'end' : [15]}}}

到目前为止，我的代码包括：

import sys
first_dict = {}
intron_dict = {}
def main():
    with open(sys.argv[1], 'r') as intron:
        for line in intron.readlines():
            line = line.split()
            chromosome = line[0]
            start = line[1]
            end = line[2]
            strand = line[3]
            first_dict = {chromosome : (strand, start, end)}

            for k, v in first_dict.iteritems():
                intron_dict.setdefault(k, []).append(v)
        print (intron_dict)
if __name__=='__main__':
    main()

这段代码允许我对chrY和chrX键进行排序，而不会覆盖值。我在合并“+”和“-”键以及将数据转换为所需格式时遇到问题。到目前为止，我的输出看起来像：

{'chrY': [('+', '6', '8'), ('+', '3', '5')], 'chrX': [('+', '10', '11'), ('-', '13', '15')]}

Answer 1

这里是另一个没有defaultdict的方法。只是用if ... else

import sys
intron_dict = dict()
def main():
    with open(sys.argv[1], 'r') as intron:
        for line in intron.readlines():
            line = line.split()
            chromosome = line[0]
            start = int(line[1]) # converted to int to avoid quotes in result
            end = int(line[2])
            strand = line[3]
            first_dict = {strand : {'start' : [start], 'end' : [end]}}

            if intron_dict.has_key(chromosome):
                if intron_dict[chromosome].has_key(strand):
                    intron_dict[chromosome][strand]['start'].append(start)
                    intron_dict[chromosome][strand]['end'].append(end)
                else:
                    intron_dict[chromosome][strand] = first_dict[strand]
            else:
                intron_dict.setdefault(chromosome, first_dict)

        print (intron_dict)

if __name__=='__main__':
    main()

输出：

{'chrY': {'+': {'start': [6, 3], 'end': [8, 5]}}, 'chrX': {'+': {'start': [10], 'end': [11]}, '-': {'start': [13], 'end': [15]}}}

Answer 2

您可以使用嵌套的^{}来简化代码，其中第三级的值是列表：

from collections import defaultdict

result = defaultdict(lambda: defaultdict(lambda: defaultdict(list)))

with open('test.txt') as f:
    for row in f:
        ch, start, end, op = row.split()
        result[ch][op]['start'].append(start)
        result[ch][op]['end'].append(end)

import json
print(json.dumps(result, indent=4))

输出：

{
    "chrY": {
        "+": {
            "start": [
                "6", 
                "3"
            ], 
            "end": [
                "8", 
                "5"
            ]
        }
    }, 
    "chrX": {
        "+": {
            "start": [
                "10"
            ], 
            "end": [
                "11"
            ]
        }, 
        "-": {
            "start": [
                "13"
            ], 
            "end": [
                "15"
            ]
        }
    }
}

Answer 3

一种方法是使用defaultdict。对于exmaple：

import sys
from pprint import  pprint
from collections import defaultdict

first_dict = defaultdict(dict)
intron_dict = {}

d = dict()


def main():
    with open('test.csv', 'r') as intron:
        for line in intron.readlines():
            chromosome, start, end, strand, = line.split()

            if strand not in first_dict[chromosome]:
                first_dict[chromosome][strand] = defaultdict(list)

            first_dict[chromosome][strand]['start'].append(start)
            first_dict[chromosome][strand]['end'].append(end)

    pprint(first_dict)

if __name__=='__main__':
    main()

结果：

defaultdict(<class 'dict'>,
            {'chrX': {'+': defaultdict(<class 'list'>,
                                       {'end': ['11'],
                                        'start': ['10']}),
                      '-': defaultdict(<class 'list'>,
                                       {'end': ['15'],
                                        'start': ['13']})},
             'chrY': {'+': defaultdict(<class 'list'>,
                                       {'end': ['8', '5'],
                                        'start': ['6', '3']})}})