Customization

You can create your customized biochemical constraint filter (as the biochemical constraints set) by inheriting DefaultBioFilter. For example:

from dsw import DefaultBioFilter

class RegionalizedGCFilter(DefaultBioFilter):

    def __init__(self, window_length, gc_bias):
        super().__init__(screen_name="Regionalized GC content constraint")
        self._window_length = window_length
        self._gc_bias = gc_bias  # bias based on 0.5 (or 50%)

    def valid(self, dna_string):
        if len(dna_string) >= self._window_length:
            for index in range(len(dna_string) - self._window_length + 1):  # judge in a window
                regional_dna_string = dna_string[index: index + self._window_length]
                gc_count = regional_dna_string.count("C") + regional_dna_string.count("G")
                if gc_count > (0.5 + self._gc_bias) * self._window_length:
                    return False
                if gc_count < (0.5 - self._gc_bias) * self._window_length:
                    return False

        else:
            gc_count = dna_string.count("C") + dna_string.count("G")
            if gc_count > (0.5 + self._gc_bias) * self._window_length:
                return False
            at_count = dna_string.count("A") + dna_string.count("T")
            if at_count > (0.5 + self._gc_bias) * self._window_length:
                return False

        return True

Here is an investigated example in this work named LocalBioFilter.

Besides, according to SPIDER-WEB, you can obtain the corresponding variable-length algorithms based on the customized biochemical constraints, for example:

from dsw import LocalBioFilter, find_vertices, connect_coding_graph

bio_filter = LocalBioFilter(max_homopolymer_runs=4, gc_range=[0.1, 0.3], undesired_motifs=["GCC"])
vertices = find_vertices(observed_length=10, bio_filter=bio_filter)
accessor = connect_coding_graph(observed_length=10, vertices=vertices, threshold=1)