This documentation is for version 2.0.dev, which is not released yet.
Load the library:
>>> import lingpy
Retrieve basic information about the library::
>>> print(lingpy.__doc__)
LingPy package for quantitative tasks in historical linguistics.
Documentation is available in the docstrings. Online documentation is available
at http://lingpy.org
Subpackages
-----------
algorithm --- Basic Algorithms for Sequence Comparison
align --- Specific Algorithms Alignment Analyses
basic --- Basic Classes for Language Comparison
check --- Classes for Exceptions, Warnings, and Check
compare --- Basic Modules for Language Comparison
convert --- Functions for Format Conversion
data --- Data Handling
evaluate --- Basic Classes and Functions for Algorithm Evaluation
read --- Basic Functions for Data Input
sequence --- Basic Functions for Sequence Modeling
thirdparty --- Temporary Forks of Third-Party-Modules
Load all important packages:
>>> from lingpy import *
Carry out a multiple alignment analysis of four sequences. First, define the sequences:
>>> seqs = ['woldemort','waldemar','wladimir','vladymyr']
Create an instance of the Multiple class:
>>> msa = Multiple(seqs)
Align the sequences, using simpole progressive alignment with default parameters:
>>> msa.prog_align()
Print out the results to the screen:
>>> print(msa)
w o l - d e m o r t
w a l - d e m a r -
w - l a d i m i r -
v - l a d y m y r -