feat: function to generate poly(A) tail sequence

src/mk_polyAtail.py

+"""Generate a poly(A) tail."""
+
+import random as rn
+
+
+def mk_polyA(length=100, a=0.914, c=0.028, g=0.025, u=0.033):
+    """Generate a poly(A) tail of specified length and composition.
+
+    This function generates a nucleotide sequence
+    that has compositional statistics resembling
+    those of poly(A) tails.
+
+    Parameters
+    ----------
+    length : int
+        Length of the desired tail
+    a : float
+        Frequency of 'A' nucleotides in the tail
+    c : float
+        Frequency of 'C' nucleotides in the tail
+    g : float
+        Frequency of 'G' nucleotides in the tail
+    u : float
+        Frequency of 'U' nucleotides in the tail
+
+    Returns
+    -------
+    string
+        The generated polyA tail
+
+    Raises
+    -------
+    ValueError
+        Provided length is not a positive int
+    ValueError
+        Provided length is too large (max allowed = 200)
+    ValueError
+        Provided proportions are not positive numbers
+    """
+    max_len = 200
+
+    # check parameters
+    length = int(length)
+    a = float(a)
+    c = float(c)
+    g = float(g)
+    u = float(u)
+    if length < 0:
+        raise ValueError(f'{length} is not a positive integer')
+    if length > 200:
+        raise ValueError(f'{length} is larger than the maximum {max_len}')
+    if a < 0 or c < 0 or g < 0 or u < 0:
+        raise ValueError(f'One of the base frequencies is negative {a}, {c}, {g}, {u}')
+
+    bases = ['A', 'C', 'G', 'U']
+    weights = [a, c, g, u]
+
+    # ensure that the values are normalized
+    s = sum(weights)
+    weights = [w/s for w in weights]
+    tail_bases = rn.choices(bases, weights=weights, k=length)
+    return "".join(tail_bases)