Python 对齐匹配序列
我试图根据前5个字符和后5个字符对齐匹配序列。例如:Python 对齐匹配序列,python,bioinformatics,Python,Bioinformatics,我试图根据前5个字符和后5个字符对齐匹配序列。例如: AAATGCEGAIRPVOGJKKK KKKTGKAFKEJWKLJFFFF FFFKEJFWKFJWEKFJIGK 将对齐并连接以创建: AAATGCEGAIRPVOGJKKKTGKAFKEJWKLJFFFFKEJFWKFJWEKFJIGK 请注意,映射区域不应重复。实际上,我有3条以上的线,它们没有排序,因此我尝试创建一个循环,将所有线对齐在一起。我不确定最好的方法 我想你可以通过制作前缀词典来解决这个问题。一旦你有了它,从一个任
AAATGCEGAIRPVOGJKKK
KKKTGKAFKEJWKLJFFFF
FFFKEJFWKFJWEKFJIGK
将对齐并连接以创建:
AAATGCEGAIRPVOGJKKKTGKAFKEJWKLJFFFFKEJFWKFJWEKFJIGK
请注意,映射区域不应重复。实际上,我有3条以上的线,它们没有排序,因此我尝试创建一个循环,将所有线对齐在一起。我不确定最好的方法 我想你可以通过制作前缀词典来解决这个问题。一旦你有了它,从一个任意的序列开始,找到它的后缀在前缀列表中。然后转到具有该前缀的序列,构建链 下面是一些代码:
def make_circular_overlapping_sequence(sequences, min_overlap=1, max_overlap=5):
# start by mapping prefixes to full sequences
prefixes = {}
for seq in sequences:
for length in range(min_overlap, max_overlap+1):
prefixes[seq[:length]] = seq
# pick arbitrary a start sequence
start = current = sequences[0]
# build a chain of sequences with overlapping suffixes and prefixes
chain = [start]
while True:
# try longest suffixes first
for length in range(max_overlap, min_overlap-1, -1):
suffix = current[-1-length:]
if suffix in prefixes:
current = prefixes[suffix]
if current == start: # looped around, so we're done
return "".join(chain)
chain.append(current[length+1:]) # don't duplicate the prefix
break
else: # for loop ended without breaking
raise ValueError("No match found for sequence {!r}"
.format(current))
测试输出:
>>> sequences = '''AAATGCEGAIRPVOGJKKK
KKKTGKAFKEJWKLJFFFF
FFFKEJFWKFJWEKFJIGK
GKXYZ1234AAAT'''.split()
>>> make_circular_overlapping_sequence(sequences)
'AAATGCEGAIRPVOGJKKKTGKAFKEJWKLJFFFFKEJFWKFJWEKFJIGKXYZ1234AAAT'
注:
“…ABC”
,后面可能跟有“ABC…”
或“BC…”
),则可能无法按需工作。目前,它总是选择最长的重叠[“a…B”,“B…C”,“C…B”]
),则代码可能会永远运行return”“。join(chain)[:-length]
我想你可以通过编一本前缀词典来解决这个问题。一旦你有了它,从一个任意的序列开始,找到它的后缀在前缀列表中。然后转到具有该前缀的序列,构建链 下面是一些代码:
def make_circular_overlapping_sequence(sequences, min_overlap=1, max_overlap=5):
# start by mapping prefixes to full sequences
prefixes = {}
for seq in sequences:
for length in range(min_overlap, max_overlap+1):
prefixes[seq[:length]] = seq
# pick arbitrary a start sequence
start = current = sequences[0]
# build a chain of sequences with overlapping suffixes and prefixes
chain = [start]
while True:
# try longest suffixes first
for length in range(max_overlap, min_overlap-1, -1):
suffix = current[-1-length:]
if suffix in prefixes:
current = prefixes[suffix]
if current == start: # looped around, so we're done
return "".join(chain)
chain.append(current[length+1:]) # don't duplicate the prefix
break
else: # for loop ended without breaking
raise ValueError("No match found for sequence {!r}"
.format(current))
测试输出:
>>> sequences = '''AAATGCEGAIRPVOGJKKK
KKKTGKAFKEJWKLJFFFF
FFFKEJFWKFJWEKFJIGK
GKXYZ1234AAAT'''.split()
>>> make_circular_overlapping_sequence(sequences)
'AAATGCEGAIRPVOGJKKKTGKAFKEJWKLJFFFFKEJFWKFJWEKFJIGKXYZ1234AAAT'
注:
“…ABC”
,后面可能跟有“ABC…”
或“BC…”
),则可能无法按需工作。目前,它总是选择最长的重叠[“a…B”,“B…C”,“C…B”]
),则代码可能会永远运行return”“。join(chain)[:-length]
你总是从第一行开始吗?如果没有,您如何选择从哪个序列开始?如果有多个匹配的可能性,该怎么办?如果没有呢?还有,你怎么处理那些不匹配的序列呢。或者,如果你有一个序列与其他10个序列相匹配呢?有什么方法可以确定你的序列偏好吗?我认为如果你使用暴力的话,
re
就太过分了。使用startswith
和endswith
字符串方法会更有效。您总是从第一行开始吗?如果没有,您如何选择从哪个序列开始?如果有多个匹配的可能性,该怎么办?如果没有呢?还有,你怎么处理那些不匹配的序列呢。或者,如果你有一个序列与其他10个序列相匹配呢?有什么方法可以确定你的序列偏好吗?我认为如果你使用暴力的话,re
就太过分了。使用startswith
和endswith
字符串方法会更有效。