Package Bio :: Package Align :: Class MultipleSeqAlignment
[hide private]
[frames] | no frames]

Class MultipleSeqAlignment

source code

object --+
         |
        MultipleSeqAlignment
Known Subclasses:

Represents a classical multiple sequence alignment (MSA).

By this we mean a collection of sequences (usually shown as rows) which are all the same length (usually with gap characters for insertions or padding). The data can then be regarded as a matrix of letters, with well defined columns.

You would typically create an MSA by loading an alignment file with the AlignIO module:

>>> from Bio import AlignIO
>>> align = AlignIO.read("Clustalw/opuntia.aln", "clustal")
>>> print(align)
SingleLetterAlphabet() alignment with 7 rows and 156 columns
TATACATTAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273285|gb|AF191659.1|AF191
TATACATTAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273284|gb|AF191658.1|AF191
TATACATTAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273287|gb|AF191661.1|AF191
TATACATAAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273286|gb|AF191660.1|AF191
TATACATTAAAGGAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273290|gb|AF191664.1|AF191
TATACATTAAAGGAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273289|gb|AF191663.1|AF191
TATACATTAAAGGAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273291|gb|AF191665.1|AF191

In some respects you can treat these objects as lists of SeqRecord objects, each representing a row of the alignment. Iterating over an alignment gives the SeqRecord object for each row:

>>> len(align)
7
>>> for record in align:
...     print("%s %i" % (record.id, len(record)))
gi|6273285|gb|AF191659.1|AF191 156
gi|6273284|gb|AF191658.1|AF191 156
gi|6273287|gb|AF191661.1|AF191 156
gi|6273286|gb|AF191660.1|AF191 156
gi|6273290|gb|AF191664.1|AF191 156
gi|6273289|gb|AF191663.1|AF191 156
gi|6273291|gb|AF191665.1|AF191 156

You can also access individual rows as SeqRecord objects via their index:

>>> print(align[0].id)
gi|6273285|gb|AF191659.1|AF191
>>> print(align[-1].id)
gi|6273291|gb|AF191665.1|AF191

And extract columns as strings:

>>> print(align[:, 1])
AAAAAAA

Or, take just the first ten columns as a sub-alignment:

>>> print(align[:, :10])
SingleLetterAlphabet() alignment with 7 rows and 10 columns
TATACATTAA gi|6273285|gb|AF191659.1|AF191
TATACATTAA gi|6273284|gb|AF191658.1|AF191
TATACATTAA gi|6273287|gb|AF191661.1|AF191
TATACATAAA gi|6273286|gb|AF191660.1|AF191
TATACATTAA gi|6273290|gb|AF191664.1|AF191
TATACATTAA gi|6273289|gb|AF191663.1|AF191
TATACATTAA gi|6273291|gb|AF191665.1|AF191

Combining this alignment slicing with alignment addition allows you to remove a section of the alignment. For example, taking just the first and last ten columns:

>>> print(align[:, :10] + align[:, -10:])
SingleLetterAlphabet() alignment with 7 rows and 20 columns
TATACATTAAGTGTACCAGA gi|6273285|gb|AF191659.1|AF191
TATACATTAAGTGTACCAGA gi|6273284|gb|AF191658.1|AF191
TATACATTAAGTGTACCAGA gi|6273287|gb|AF191661.1|AF191
TATACATAAAGTGTACCAGA gi|6273286|gb|AF191660.1|AF191
TATACATTAAGTGTACCAGA gi|6273290|gb|AF191664.1|AF191
TATACATTAAGTATACCAGA gi|6273289|gb|AF191663.1|AF191
TATACATTAAGTGTACCAGA gi|6273291|gb|AF191665.1|AF191

Note - This object replaced the older Alignment object defined in module Bio.Align.Generic but is not fully backwards compatible with it.

Note - This object does NOT attempt to model the kind of alignments used in next generation sequencing with multiple sequencing reads which are much shorter than the alignment, and where there is usually a consensus or reference sequence with special status.

Instance Methods [hide private]
 
__init__(self, records, alphabet=None, annotations=None)
Initialize a new MultipleSeqAlignment object.
source code
 
_str_line(self, record, length=50)
Returns a truncated string representation of a SeqRecord (PRIVATE).
source code
 
__str__(self)
Returns a multi-line string summary of the alignment.
source code
 
__repr__(self)
Returns a representation of the object for debugging.
source code
 
format(self, format)
Returns the alignment as a string in the specified file format.
source code
 
__format__(self, format_spec)
Returns the alignment as a string in the specified file format.
source code
 
__iter__(self)
Iterate over alignment rows as SeqRecord objects.
source code
 
__len__(self)
Returns the number of sequences in the alignment.
source code
 
get_alignment_length(self)
Return the maximum length of the alignment.
source code
 
add_sequence(self, descriptor, sequence, start=None, end=None, weight=1.0)
Add a sequence to the alignment.
source code
 
extend(self, records)
Add more SeqRecord objects to the alignment as rows.
source code
 
append(self, record)
Add one more SeqRecord object to the alignment as a new row.
source code
 
_append(self, record, expected_length=None)
Helper function (PRIVATE).
source code
 
__add__(self, other)
Combines two alignments with the same number of rows by adding them.
source code
 
__getitem__(self, index)
Access part of the alignment.
source code
 
sort(self, key=None, reverse=False)
Sort the rows (SeqRecord objects) of the alignment in place.
source code

Inherited from object: __delattr__, __getattribute__, __hash__, __new__, __reduce__, __reduce_ex__, __setattr__, __sizeof__, __subclasshook__

Properties [hide private]

Inherited from object: __class__

Method Details [hide private]

__init__(self, records, alphabet=None, annotations=None)
(Constructor)

source code 

Initialize a new MultipleSeqAlignment object.

Arguments:
  • records - A list (or iterator) of SeqRecord objects, whose

    sequences are all the same length. This may be an be an empty list.

  • alphabet - The alphabet for the whole alignment, typically a gapped

    alphabet, which should be a super-set of the individual record alphabets. If omitted, a consensus alphabet is used.

  • annotations - Information about the whole alignment (dictionary).

You would normally load a MSA from a file using Bio.AlignIO, but you can do this from a list of SeqRecord objects too:

>>> from Bio.Alphabet import generic_dna
>>> from Bio.Seq import Seq
>>> from Bio.SeqRecord import SeqRecord
>>> from Bio.Align import MultipleSeqAlignment
>>> a = SeqRecord(Seq("AAAACGT", generic_dna), id="Alpha")
>>> b = SeqRecord(Seq("AAA-CGT", generic_dna), id="Beta")
>>> c = SeqRecord(Seq("AAAAGGT", generic_dna), id="Gamma")
>>> align = MultipleSeqAlignment([a, b, c], annotations={"tool": "demo"})
>>> print(align)
DNAAlphabet() alignment with 3 rows and 7 columns
AAAACGT Alpha
AAA-CGT Beta
AAAAGGT Gamma
>>> align.annotations
{'tool': 'demo'}

NOTE - The older Bio.Align.Generic.Alignment class only accepted a single argument, an alphabet. This is still supported via a backwards compatible "hack" so as not to disrupt existing scripts and users, but is deprecated and will be removed in a future release.

Overrides: object.__init__

_str_line(self, record, length=50)

source code 

Returns a truncated string representation of a SeqRecord (PRIVATE).

This is a PRIVATE function used by the __str__ method.

__str__(self)
(Informal representation operator)

source code 

Returns a multi-line string summary of the alignment.

This output is intended to be readable, but large alignments are shown truncated. A maximum of 20 rows (sequences) and 50 columns are shown, with the record identifiers. This should fit nicely on a single screen. e.g.

>>> from Bio.Alphabet import IUPAC, Gapped
>>> from Bio.Align import MultipleSeqAlignment
>>> align = MultipleSeqAlignment([], Gapped(IUPAC.unambiguous_dna, "-"))
>>> align.add_sequence("Alpha", "ACTGCTAGCTAG")
>>> align.add_sequence("Beta",  "ACT-CTAGCTAG")
>>> align.add_sequence("Gamma", "ACTGCTAGATAG")
>>> print(align)
Gapped(IUPACUnambiguousDNA(), '-') alignment with 3 rows and 12 columns
ACTGCTAGCTAG Alpha
ACT-CTAGCTAG Beta
ACTGCTAGATAG Gamma

See also the alignment's format method.

Overrides: object.__str__

__repr__(self)
(Representation operator)

source code 

Returns a representation of the object for debugging.

The representation cannot be used with eval() to recreate the object, which is usually possible with simple python ojects. For example:

<Bio.Align.MultipleSeqAlignment instance (2 records of length 14, SingleLetterAlphabet()) at a3c184c>

The hex string is the memory address of the object, see help(id). This provides a simple way to visually distinguish alignments of the same size.

Overrides: object.__repr__

format(self, format)

source code 

Returns the alignment as a string in the specified file format.

The format should be a lower case string supported as an output format by Bio.AlignIO (such as "fasta", "clustal", "phylip", "stockholm", etc), which is used to turn the alignment into a string.

e.g.

>>> from Bio.Alphabet import IUPAC, Gapped
>>> from Bio.Align import MultipleSeqAlignment
>>> align = MultipleSeqAlignment([], Gapped(IUPAC.unambiguous_dna, "-"))
>>> align.add_sequence("Alpha", "ACTGCTAGCTAG")
>>> align.add_sequence("Beta",  "ACT-CTAGCTAG")
>>> align.add_sequence("Gamma", "ACTGCTAGATAG")
>>> print(align.format("fasta"))
>Alpha
ACTGCTAGCTAG
>Beta
ACT-CTAGCTAG
>Gamma
ACTGCTAGATAG
<BLANKLINE>
>>> print(align.format("phylip"))
 3 12
Alpha      ACTGCTAGCT AG
Beta       ACT-CTAGCT AG
Gamma      ACTGCTAGAT AG
<BLANKLINE>

For Python 2.6, 3.0 or later see also the built in format() function.

__format__(self, format_spec)

source code 

Returns the alignment as a string in the specified file format.

This method supports the python format() function added in Python 2.6/3.0. The format_spec should be a lower case string supported by Bio.AlignIO as an output file format. See also the alignment's format() method.

Overrides: object.__format__

__iter__(self)

source code 

Iterate over alignment rows as SeqRecord objects.

e.g.

>>> from Bio.Alphabet import IUPAC, Gapped
>>> from Bio.Align import MultipleSeqAlignment
>>> align = MultipleSeqAlignment([], Gapped(IUPAC.unambiguous_dna, "-"))
>>> align.add_sequence("Alpha", "ACTGCTAGCTAG")
>>> align.add_sequence("Beta",  "ACT-CTAGCTAG")
>>> align.add_sequence("Gamma", "ACTGCTAGATAG")
>>> for record in align:
...    print(record.id)
...    print(record.seq)
Alpha
ACTGCTAGCTAG
Beta
ACT-CTAGCTAG
Gamma
ACTGCTAGATAG

__len__(self)
(Length operator)

source code 

Returns the number of sequences in the alignment.

Use len(alignment) to get the number of sequences (i.e. the number of rows), and alignment.get_alignment_length() to get the length of the longest sequence (i.e. the number of columns).

This is easy to remember if you think of the alignment as being like a list of SeqRecord objects.

get_alignment_length(self)

source code 

Return the maximum length of the alignment.

All objects in the alignment should (hopefully) have the same length. This function will go through and find this length by finding the maximum length of sequences in the alignment.

>>> from Bio.Alphabet import IUPAC, Gapped
>>> from Bio.Align import MultipleSeqAlignment
>>> align = MultipleSeqAlignment([], Gapped(IUPAC.unambiguous_dna, "-"))
>>> align.add_sequence("Alpha", "ACTGCTAGCTAG")
>>> align.add_sequence("Beta",  "ACT-CTAGCTAG")
>>> align.add_sequence("Gamma", "ACTGCTAGATAG")
>>> align.get_alignment_length()
12

If you want to know the number of sequences in the alignment, use len(align) instead:

>>> len(align)
3

add_sequence(self, descriptor, sequence, start=None, end=None, weight=1.0)

source code 

Add a sequence to the alignment.

This doesn't do any kind of alignment, it just adds in the sequence object, which is assumed to be prealigned with the existing sequences.

Arguments:
  • descriptor - The descriptive id of the sequence being added. This will be used as the resulting SeqRecord's .id property (and, for historical compatibility, also the .description property)
  • sequence - A string with sequence info.
  • start - You can explicitly set the start point of the sequence. This is useful (at least) for BLAST alignments, which can just be partial alignments of sequences.
  • end - Specify the end of the sequence, which is important for the same reason as the start.
  • weight - The weight to place on the sequence in the alignment. By default, all sequences have the same weight. (0.0 => no weight, 1.0 => highest weight)

In general providing a SeqRecord and calling .append is prefered.

extend(self, records)

source code 

Add more SeqRecord objects to the alignment as rows.

They must all have the same length as the original alignment, and have alphabets compatible with the alignment's alphabet. For example,

>>> from Bio.Alphabet import generic_dna
>>> from Bio.Seq import Seq
>>> from Bio.SeqRecord import SeqRecord
>>> from Bio.Align import MultipleSeqAlignment
>>> a = SeqRecord(Seq("AAAACGT", generic_dna), id="Alpha")
>>> b = SeqRecord(Seq("AAA-CGT", generic_dna), id="Beta")
>>> c = SeqRecord(Seq("AAAAGGT", generic_dna), id="Gamma")
>>> d = SeqRecord(Seq("AAAACGT", generic_dna), id="Delta")
>>> e = SeqRecord(Seq("AAA-GGT", generic_dna), id="Epsilon")

First we create a small alignment (three rows):

>>> align = MultipleSeqAlignment([a, b, c])
>>> print(align)
DNAAlphabet() alignment with 3 rows and 7 columns
AAAACGT Alpha
AAA-CGT Beta
AAAAGGT Gamma

Now we can extend this alignment with another two rows:

>>> align.extend([d, e])
>>> print(align)
DNAAlphabet() alignment with 5 rows and 7 columns
AAAACGT Alpha
AAA-CGT Beta
AAAAGGT Gamma
AAAACGT Delta
AAA-GGT Epsilon

Because the alignment object allows iteration over the rows as SeqRecords, you can use the extend method with a second alignment (provided its sequences have the same length as the original alignment).

append(self, record)

source code 

Add one more SeqRecord object to the alignment as a new row.

This must have the same length as the original alignment (unless this is the first record), and have an alphabet compatible with the alignment's alphabet.

>>> from Bio import AlignIO
>>> align = AlignIO.read("Clustalw/opuntia.aln", "clustal")
>>> print(align)
SingleLetterAlphabet() alignment with 7 rows and 156 columns
TATACATTAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273285|gb|AF191659.1|AF191
TATACATTAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273284|gb|AF191658.1|AF191
TATACATTAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273287|gb|AF191661.1|AF191
TATACATAAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273286|gb|AF191660.1|AF191
TATACATTAAAGGAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273290|gb|AF191664.1|AF191
TATACATTAAAGGAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273289|gb|AF191663.1|AF191
TATACATTAAAGGAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273291|gb|AF191665.1|AF191
>>> len(align)
7

We'll now construct a dummy record to append as an example:

>>> from Bio.Seq import Seq
>>> from Bio.SeqRecord import SeqRecord
>>> dummy = SeqRecord(Seq("N"*156), id="dummy")

Now append this to the alignment,

>>> align.append(dummy)
>>> print(align)
SingleLetterAlphabet() alignment with 8 rows and 156 columns
TATACATTAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273285|gb|AF191659.1|AF191
TATACATTAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273284|gb|AF191658.1|AF191
TATACATTAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273287|gb|AF191661.1|AF191
TATACATAAAAGAAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273286|gb|AF191660.1|AF191
TATACATTAAAGGAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273290|gb|AF191664.1|AF191
TATACATTAAAGGAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273289|gb|AF191663.1|AF191
TATACATTAAAGGAGGGGGATGCGGATAAATGGAAAGGCGAAAG...AGA gi|6273291|gb|AF191665.1|AF191
NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN...NNN dummy
>>> len(align)
8

__add__(self, other)
(Addition operator)

source code 

Combines two alignments with the same number of rows by adding them.

If you have two multiple sequence alignments (MSAs), there are two ways to think about adding them - by row or by column. Using the extend method adds by row. Using the addition operator adds by column. For example,

>>> from Bio.Alphabet import generic_dna
>>> from Bio.Seq import Seq
>>> from Bio.SeqRecord import SeqRecord
>>> from Bio.Align import MultipleSeqAlignment
>>> a1 = SeqRecord(Seq("AAAAC", generic_dna), id="Alpha")
>>> b1 = SeqRecord(Seq("AAA-C", generic_dna), id="Beta")
>>> c1 = SeqRecord(Seq("AAAAG", generic_dna), id="Gamma")
>>> a2 = SeqRecord(Seq("GT", generic_dna), id="Alpha")
>>> b2 = SeqRecord(Seq("GT", generic_dna), id="Beta")
>>> c2 = SeqRecord(Seq("GT", generic_dna), id="Gamma")
>>> left = MultipleSeqAlignment([a1, b1, c1],
...                             annotations={"tool": "demo", "name": "start"})
>>> right = MultipleSeqAlignment([a2, b2, c2],
...                             annotations={"tool": "demo", "name": "end"})

Now, let's look at these two alignments:

>>> print(left)
DNAAlphabet() alignment with 3 rows and 5 columns
AAAAC Alpha
AAA-C Beta
AAAAG Gamma
>>> print(right)
DNAAlphabet() alignment with 3 rows and 2 columns
GT Alpha
GT Beta
GT Gamma

And add them:

>>> combined = left + right
>>> print(combined)
DNAAlphabet() alignment with 3 rows and 7 columns
AAAACGT Alpha
AAA-CGT Beta
AAAAGGT Gamma

For this to work, both alignments must have the same number of records (here they both have 3 rows):

>>> len(left)
3
>>> len(right)
3
>>> len(combined)
3

The individual rows are SeqRecord objects, and these can be added together. Refer to the SeqRecord documentation for details of how the annotation is handled. This example is a special case in that both original alignments shared the same names, meaning when the rows are added they also get the same name.

Any common annotations are preserved, but differing annotation is lost. This is the same behaviour used in the SeqRecord annotations and is designed to prevent accidental propagation of inappropriate values:

>>> combined.annotations
{'tool': 'demo'}

__getitem__(self, index)
(Indexing operator)

source code 

Access part of the alignment.

Depending on the indices, you can get a SeqRecord object (representing a single row), a Seq object (for a single columns), a string (for a single characters) or another alignment (representing some part or all of the alignment).

align[r,c] gives a single character as a string align[r] gives a row as a SeqRecord align[r,:] gives a row as a SeqRecord align[:,c] gives a column as a Seq (using the alignment's alphabet)

align[:] and align[:,:] give a copy of the alignment

Anything else gives a sub alignment, e.g. align[0:2] or align[0:2,:] uses only row 0 and 1 align[:,1:3] uses only columns 1 and 2 align[0:2,1:3] uses only rows 0 & 1 and only cols 1 & 2

We'll use the following example alignment here for illustration:

>>> from Bio.Alphabet import generic_dna
>>> from Bio.Seq import Seq
>>> from Bio.SeqRecord import SeqRecord
>>> from Bio.Align import MultipleSeqAlignment
>>> a = SeqRecord(Seq("AAAACGT", generic_dna), id="Alpha")
>>> b = SeqRecord(Seq("AAA-CGT", generic_dna), id="Beta")
>>> c = SeqRecord(Seq("AAAAGGT", generic_dna), id="Gamma")
>>> d = SeqRecord(Seq("AAAACGT", generic_dna), id="Delta")
>>> e = SeqRecord(Seq("AAA-GGT", generic_dna), id="Epsilon")
>>> align = MultipleSeqAlignment([a, b, c, d, e], generic_dna)

You can access a row of the alignment as a SeqRecord using an integer index (think of the alignment as a list of SeqRecord objects here):

>>> first_record = align[0]
>>> print("%s %s" % (first_record.id, first_record.seq))
Alpha AAAACGT
>>> last_record = align[-1]
>>> print("%s %s" % (last_record.id, last_record.seq))
Epsilon AAA-GGT

You can also access use python's slice notation to create a sub-alignment containing only some of the SeqRecord objects:

>>> sub_alignment = align[2:5]
>>> print(sub_alignment)
DNAAlphabet() alignment with 3 rows and 7 columns
AAAAGGT Gamma
AAAACGT Delta
AAA-GGT Epsilon

This includes support for a step, i.e. align[start:end:step], which can be used to select every second sequence:

>>> sub_alignment = align[::2]
>>> print(sub_alignment)
DNAAlphabet() alignment with 3 rows and 7 columns
AAAACGT Alpha
AAAAGGT Gamma
AAA-GGT Epsilon

Or to get a copy of the alignment with the rows in reverse order:

>>> rev_alignment = align[::-1]
>>> print(rev_alignment)
DNAAlphabet() alignment with 5 rows and 7 columns
AAA-GGT Epsilon
AAAACGT Delta
AAAAGGT Gamma
AAA-CGT Beta
AAAACGT Alpha

You can also use two indices to specify both rows and columns. Using simple integers gives you the entry as a single character string. e.g.

>>> align[3, 4]
'C'

This is equivalent to:

>>> align[3][4]
'C'

or:

>>> align[3].seq[4]
'C'

To get a single column (as a string) use this syntax:

>>> align[:, 4]
'CCGCG'

Or, to get part of a column,

>>> align[1:3, 4]
'CG'

However, in general you get a sub-alignment,

>>> print(align[1:5, 3:6])
DNAAlphabet() alignment with 4 rows and 3 columns
-CG Beta
AGG Gamma
ACG Delta
-GG Epsilon

This should all seem familiar to anyone who has used the NumPy array or matrix objects.

sort(self, key=None, reverse=False)

source code 

Sort the rows (SeqRecord objects) of the alignment in place.

This sorts the rows alphabetically using the SeqRecord object id by default. The sorting can be controlled by supplying a key function which must map each SeqRecord to a sort value.

This is useful if you want to add two alignments which use the same record identifiers, but in a different order. For example,

>>> from Bio.Alphabet import generic_dna
>>> from Bio.Seq import Seq
>>> from Bio.SeqRecord import SeqRecord
>>> from Bio.Align import MultipleSeqAlignment
>>> align1 = MultipleSeqAlignment([
...              SeqRecord(Seq("ACGT", generic_dna), id="Human"),
...              SeqRecord(Seq("ACGG", generic_dna), id="Mouse"),
...              SeqRecord(Seq("ACGC", generic_dna), id="Chicken"),
...          ])
>>> align2 = MultipleSeqAlignment([
...              SeqRecord(Seq("CGGT", generic_dna), id="Mouse"),
...              SeqRecord(Seq("CGTT", generic_dna), id="Human"),
...              SeqRecord(Seq("CGCT", generic_dna), id="Chicken"),
...          ])

If you simple try and add these without sorting, you get this:

>>> print(align1 + align2)
DNAAlphabet() alignment with 3 rows and 8 columns
ACGTCGGT <unknown id>
ACGGCGTT <unknown id>
ACGCCGCT Chicken

Consult the SeqRecord documentation which explains why you get a default value when annotation like the identifier doesn't match up. However, if we sort the alignments first, then add them we get the desired result:

>>> align1.sort()
>>> align2.sort()
>>> print(align1 + align2)
DNAAlphabet() alignment with 3 rows and 8 columns
ACGCCGCT Chicken
ACGTCGTT Human
ACGGCGGT Mouse

As an example using a different sort order, you could sort on the GC content of each sequence.

>>> from Bio.SeqUtils import GC
>>> print(align1)
DNAAlphabet() alignment with 3 rows and 4 columns
ACGC Chicken
ACGT Human
ACGG Mouse
>>> align1.sort(key = lambda record: GC(record.seq))
>>> print(align1)
DNAAlphabet() alignment with 3 rows and 4 columns
ACGT Human
ACGC Chicken
ACGG Mouse

There is also a reverse argument, so if you wanted to sort by ID but backwards:

>>> align1.sort(reverse=True)
>>> print(align1)
DNAAlphabet() alignment with 3 rows and 4 columns
ACGG Mouse
ACGT Human
ACGC Chicken