helpers.py 文件源码

def find_functional_units(gpr_str):
    """
    Return an iterator of gene IDs grouped by boolean rules from the gpr_str.

    The gpr_str is first transformed into an algebraic expression, replacing
    the boolean operators 'or' with '+' and 'and' with '*'. Treating the
    gene IDs as sympy.symbols this allows a mathematical expansion of the
    algebraic expression. The expanded form is then split again producing sets
    of gene IDs that in the gpr_str had an 'and' relationship.

    Parameters
    ----------
    gpr_str : string
            A string consisting of gene ids and the boolean expressions 'and'
            and 'or'

    """
    algebraic_form = re.sub('[Oo]r', '+', re.sub('[Aa]nd', '*', gpr_str))
    expanded = str(expand(algebraic_form))
    for unit in expanded.replace('+', ',').split(' , '):
        yield unit.split('*')