Alignment Filtering

Filter alignments. Available filtering options:

Alignment length
Parsimony informative sites
Percent parsimony informative sites
Taxon completeness
Missing data (upcoming)

Steps

Install PySEGUL using pip if you haven’t done it yet

pip install pysegul

Create a new Python script, import the library, and write python code. Example:

import pysegul

def filter_alignments():
    input_dir = 'tests/align-data'
    input_format = 'nexus'
    datatype = 'dna'
    output_format = 'nexus'
    is_concat = False
    output_dir = 'tests/output'
    filter = pysegul.AlignmentFiltering(
        input_format,
        datatype,
        output_dir,
        output_format,
        is_concat
        )
    filter.input_dir = input_dir
    # For filtering by minimal alignment length
    filter.minimal_length(8)
    # For filtering by minimal parsimony informative sites
    filter.minimal_parsimony_inf(4)
    # For filtering by percent parsimony informative sites
    filter.percent_informative(0.5)
    # For filtering by taxon completeness (proportion of all taxa)
    filter.minimal_taxa(0.5)
    # For keeping alignments that contain all listed taxa
    filter.contain_taxa(['taxon_1', 'taxon_2'])

To concatenate the filtered alignments, set is_concat to True and provide a prefix and a partition format:

filter = pysegul.AlignmentFiltering(
    'nexus',
    'dna',
    'tests/output',
    'nexus',
    True,
    prefix='filtered_concat',
    partition_fmt='raxml',
    )

Upcoming filtering methods

Method	Keep alignments that
`minimal_taxa_count(count)`	have at least `count` taxa
`maximal_length(length)`	are at most `length` sites long
`maximal_parsimony_inf(count)`	have at most `count` parsimony informative sites
`missing_data(proportion)`	have at most `proportion` (0.0–1.0) missing data (gaps and `?`)

filter.minimal_taxa_count(10)
filter.maximal_length(2000)
filter.maximal_parsimony_inf(500)
filter.missing_data(0.25)