A library of base classes for domain-level strand displacement (DSD) programming. If you are starting a new project that requires domains, complexes, secondary structures, nucleotide sequences, reactions, etc. then you might find this module useful. Note that this is the in-house library of the nuskell compiler framework; it handels the parsing of supported input file formats (mainly *.PIL) and their translation into the respective DSD objects.
All objects provided here are singletons. In other words, once you
intialize a Domain d1 = DomainS(name = 'a', length = 15), you will not be
allowed to initialize a new domain with the same name but different length,
unless you delete (all references to) the variable d1 first. If you try to
initialize the same domain a second time, then the new object is d1.
Generally, each (DomainS, ComplexS, MacrostateS, ReactionS) must be initialized
with parameters that define their canonical form and (optionally) a name. If
both are given, the library checks that there are no conflicts with existing
objects, if the name is not provided, the library tries to initialize a new
object with an automatic name, but raises a SingletonError which holds a
reference to the existing object. If only the name is given, the existing
object is returned.
This library is expected to evolve further, potentially breaking backward compatibility as new challenges are waiting in the nuskell compiler framework. Don't hesitate to contact the authors with questions about future plans. Inheritance from the provided objects is fully supported and encouraged.
To install this library use pip:
$ pip install dsdobjects
or the following command in the root directory:
$ python ./setup.py install
from dsdobjects import DomainS, ComplexS
# Define a few toy domains:
a = DomainS('a', length = 15)
b = DomainS('b', length = 9)
c = DomainS('c', length = 6)
# DomainS objects have exactly one complement, it can be initialized
# and/or accessed using the __invert__ operator. The singleton type
# ensures that there is only one object for each domain.
assert (a is ~(~a))
# Use the Domains to define a Complex ...
foo = ComplexS([a, b, c, ~b, '+', ~a], list('((.)+)'), name = 'foo')
# ... and test some of the built-in complex properties:
foo.kernel_string
foo.canonical_form
foo.size
foo.pair_table
for (se, ss) in foo.rotate():
print(se, ss)
# If you initialize a disconnected complex ...
bar = ComplexS([a, b, c, ~b, '+', ~a], list('.(.)+.'), name = 'bar')
assert bar.is_connected is False
# ... use split to get all indiviudal complexes:
cx1, cx2 = bar.split()Initialize prototype objects by loading a system (or a single line) of *.PIL file format:
from dsdobjects import DomainS
from dsdobjects.objectio import set_io_objects, clear_io_objects, read_pil, read_pil_line
# Use the builtin singleton obects form the dsdobjects library.
set_io_objects()
# The following dictionary contains references to all objects.
outdict = read_pil('filename', is_file = True)
# The following let's you quickly initialize single objects.
d5 = read_pil_line("length d5 = 7")
assert isinstance(d5, DomainS)
d6 = read_pil_line("sequence d6 = NNNNN")
assert isinstance(d6, DomainS)
assert d6.sequence == 'NNNNN'0.8 -- requires Python<=3.7
- complete rewrite of the library to use singleton objects with weakref
Stefan Badelt
This library contains adapted code from various related Python packages coded in the DNA and Natural Algorithms Group, Caltech:
- "DNAObjects" coded by Joseph Berleant and Joseph Schaeffer
- peppercornenumerator coded by Kathrik Sarma, Casey Grun and Erik Winfree
- nuskell coded by Seung Woo Shin
MIT