GC content for RNA sequences

[Lucandia]

Setup

I am reporting a problem with:

• Biopython version: 1.81
• Python version: 3.9.0
• System: macOS Sonoma

Expected behaviour

from Bio.SeqUtils import gc_fraction
gr_fraction("GGAUCUUCGGAUCU") 
# expected result: 0.5

Actual behaviour

from Bio.SeqUtils import gc_fraction
gr_fraction("GGAUCUUCGGAUCU") 
# result: 0.7777777777777779

What to fix

Please add U to the non-ambiguous nucleotides to support RNA sequences (currently, the non-ambiguous nucleotides are ATCGSW). The previous GC content function (now deprecated) was supporting RNA sequences too

[peterjc]

The old function didn't explicitly handle U either:

def GC(seq):
    """Calculate G+C content (DEPRECATED).
    Use Bio.SeqUtils.gc_fraction instead.
    """
    warnings.warn(
        "GC is deprecated; please use gc_fraction instead.",
        BiopythonDeprecationWarning,
    )

    gc = sum(seq.count(x) for x in ["G", "C", "g", "c", "S", "s"])
    try:
        return gc * 100.0 / len(seq)
    except ZeroDivisionError:
        return 0.0

(See 960a5e2 which removed it)

To mimic the old behaviour, use the ignore mode rather than the default remove mode:

>>> from Bio.SeqUtils import gc_fraction
>>> gc_fraction("GGAUCUUCGGAUCU", "ignore") 
0.5

That said, explicitly handling U for RNA support makes sense. I think it would be a one line change to add U to the dictionary here:

https://github.com/biopython/biopython/blob/biopython-183/Bio/SeqUtils/__init__.py#L32

This would also need a new test case (e.g. your example). Would you like to make a pull request?

[Lucandia]

Sure, I'm on it!