Splitting header and content using regex

Question

I have the following sequence text in a file

every header starts with ">" content number of lines is random

>XM_024446048.1 PREDICTED: Homo sapiens mannosidase alpha class 2A member 1 (MAN2A1), transcript variant X2, mRNA
CAGCCCCC
TGAGCGAC
TCCCTAATGTG
ACAGTAAAGAA
>NM_001308028.1 Homo sapiens FER tyrosine kinase (FER), transcript variant 2, mRNA
CAGCCC
CCGTGACGC
GGGGTGGTGACT
GGCTC
GGTGGT
GTGAC
>NM_0013082323028.1 H STZ mRSN1A
CAGCCC
CCGTGACGC
GGG
GTGGTGA
CTGGCTCCGGAGT
CTGAGGGGTTCGG

I want to create a nested dict in the following format:

nested_dict = { 'sequence1': {

'header': 'XM_024446048.1 PREDICTED: Homo sapiens mannosidase alpha class 2A member 1 (MAN2A1), transcript variant X2, mRNA',
'content': 'CAGCCCCCTGAGCGACTCCCTAATGTGACAGTAAAGAA'


},
               'sequence2': {

'header': 'NM_001308028.1 Homo sapiens FER tyrosine kinase (FER), transcript variant 2, mRNA',
'content': 'CAGCCCCCGTGACGCGGGGTGGTGACTGGCTCGGTGGTGTGAC'


},

               'sequence3': {

'header': 'NM_0013082323028.1 H STZ mRSN1A',
'content': 'CAGCCCCCGTGACGCGGGGTGGTGACTGGCTCCGGAGTCTGAGGGGTTCGG'


}

I'm stuck with regex expressions, can anyone help me?

Answer 1

You could use

^>
(?P<header>
[^\n()]*
    (?:\((?P<name>[^\n()]+)\))?
[^()\n]*
)
[\n\r]
(?P<content>[\s\S]+?)(?=^>|\Z)

with the multiline and verbose flag, see a demo on regex101.com.

---

In Python:

rx = re.compile(r'''
    ^>
    (?P<header>
    [^\n()]*
        (?:\((?P<name>[^\n()]+)\))?
    [^()\n]*
    )
    [\n\r]
    (?P<content>[\s\S]+?)(?=^>|\Z)''', re.M | re.X)

sequences = {"sequence{}".format(idx):
                 {"header": m.group('header'),
                  "content": m.group("content"),
                  "name": m.group('name')
                  }
             for idx, m in enumerate(rx.finditer(data), 1)}
print(sequences)

This yields

{'sequence1': {'header': 'XM_024446048.1 PREDICTED: Homo sapiens mannosidase alpha class 2A member 1 (MAN2A1), transcript variant X2, mRNA', 'content': 'CAGCCCCC\nTGAGCGAC\nTCCCTAATGTG\nACAGTAAAGAA\n', 'name': 'MAN2A1'}, 'sequence2': {'header': 'NM_001308028.1 Homo sapiens FER tyrosine kinase (FER), transcript variant 2, mRNA', 'content': 'CAGCCC\nCCGTGACGC\nGGGGTGGTGACT\nGGCTC\nGGTGGT\nGTGAC\n', 'name': 'FER'}, 'sequence3': {'header': 'NM_0013082323028.1 H STZ mRSN1A', 'content': 'CAGCCC\nCCGTGACGC\nGGG\nGTGGTGA\nCTGGCTCCGGAGT\nCTGAGGGGTTCGG', 'name': None}}

See a demo on ideone.com.