edinburgh-genome-foundry / codon-usage-tables Goto Github PK

I've used pandas df.to_csv() to write out a custom codon table
codon_table.csv

reading this table gives an error when reading the last line. adding the length check ( if len(line)>3:) solves this problem.

def csv_string_to_codons_dict(csv_string):
    """Transform a CSV string of a codon table to a dict."""
    result = {}
    for line in csv_string.split("\n")[1:]:
        if len(line)>3: # new check
            aa, codon, freq = line.split(",")
            if aa not in result:
                result[aa] = {}
            result[aa][codon] = float(freq)
    return result

Can we get this into a Python package and/or put it on PyPI?

When installing via pip pip install python_codon_tables the following error is produced

Traceback (most recent call last):
      File "<string>", line 1, in <module>
      File "/private/var/folders/_f/5kqpr8kx5zl9qjtkzzkb31xw0000gn/T/pip-install-a0_5fa8a/python-codon-tables/setup.py", line 16, in <module>
        with open(os.path.join('data', 'version.txt'), 'r') as f:
    FileNotFoundError: [Errno 2] No such file or directory: 'data/version.txt'

Seems that the issue is that the data directory is one up from the packaged directory.

Please include a license file for the repository and the PyPI package python-codon-tables. On PyPI it only mentions about a Public Domain license, but I don't see a physical license document.

https://pypi.org/project/python-codon-tables/

https://www.kazusa.or.jp/codon/ is down (been down for a few days now). Wondering if you've got any thoughts for implementing a failover?

(PS Hello Zulko and EGF, we love the stuff you're working on!)