def test_single_X(self):
p = Seq("GYTXTRS")
r = x_to_ggsg(p)
assert r == Seq("GYTGTRS")
p = Seq("XGYTTRS")
r = x_to_ggsg(p)
assert r == Seq("GGYTTRS")
p = Seq("GYTTRSX")
r = x_to_ggsg(p)
assert r == Seq("GYTTRSG")
def x2ggsg(input, output):
"""Replace stretches of Xs with Serine-Glycine linker (in a GGSG pattern)
INPUT and OUTPUT are paths to fasta files or "-" to specify STDIN/STDOUT.
"""
for (name, seq, qual) in readfq(input):
replacement = x_to_ggsg(seq)
if replacement != seq:
output_title = f"{name}|withGSlinker"
else:
output_title = name
print(f">{output_title}\n{replacement}", file=output)
def test_many_Xs(self):
p = Seq("GYTXXXXXXXXXTRS")
r = x_to_ggsg(p)
assert r == Seq("GYTGGSGGGSGGTRS")
def test_many_single_Xs(self):
p = Seq("GXYTXTXRXS")
r = x_to_ggsg(p)
assert r == Seq("GGYTGTGRGS")
def test_double_X(self):
p = Seq("GYTXXTRS")
r = x_to_ggsg(p)
assert r == Seq("GYTGGTRS")
def test_multiple_stretches(self):
p = Seq("XGYTXXXTRXXS")
r = x_to_ggsg(p)
assert r == Seq("GGYTGGSTRGGS")
def test_Xs_infix(self):
p = Seq("GYTXXXXXTRS")
r = x_to_ggsg(p)
assert r == Seq("GYTGGSGGTRS")
def test_Xs_suffix(self):
p = Seq("GYTTRSXXXX")
r = x_to_ggsg(p)
assert r == Seq("GYTTRSGGSG")
def test_Xs_prefix(self):
p = Seq("XXXGYTTRS")
r = x_to_ggsg(p)
assert r == Seq("GGSGYTTRS")
def test_no_Xs(self):
p = Seq("GYTTRS")
r = x_to_ggsg(p)
assert p == r
def test_null_seq(self):
p = Seq("")
r = x_to_ggsg(p)
assert p == r
