synbio

synbio is a library for designing and assembling DNA. Users can design plasmids or libraries and export multi-step build protocols. Input SeqRecords. Output assembly SeqRecords, protocols, plate maps, and robotic picklists.

Documentation is available at https://lattice-automation.github.io/synbio

Installation

# with pip
pip install synbio

# OR with conda
conda install -c jtimmons synbio

Models

synbio only expects the user to define their Design and Protocol. Several protocols are pre-defined.

Designs

All are in synbio.designs:

  • Combinatorial - list of SeqRecords to combinatorially anneal into all valid assemblies

  • CombinatorialBins - list of bins of SeqRecords for combinatorial assembly of records between bins

  • Plasmid - single list of SeqRecords to combine into a plasmid

  • PlasmidLibrary - list of list of SeqRecords to combine into plasmids

Protocols

All are in synbio.protocols:

Example

In the example below, the user specifies a combinatorial library design. All SeqRecords are tested for circularization with other SeqRecords. New and valid plasmids are assembled.

Behind the scenes, synbio is filtering all combinations of SeqRecords from the design that will circularize into valid plasmids (via circuits in a graph). After running the protocol, users can export plate maps (to_csv()), composite plasmids (to_fasta(), to_genbank()), and assembly instructions (to_txt(), to_picklists()).

"""Example of a Combinatorial Golden Gate assembly with human and robot output protocols."""

import os
from Bio.SeqIO import parse
from synbio.designs import Combinatorial
from synbio.protocols import GoldenGate

def read_all_records():
    gg_dir = os.path.join(".", "data", "goldengate")
    records = []
    for file in os.listdir(gg_dir):
        if file.endswith(".gb"):
            records.extend(parse(os.path.join(gg_dir, file), "genbank"))
    return records

# create a combinatorial library design from all valid combinations
design = Combinatorial(read_all_records())

# create a protocol using Golden Gate as the sole composite step and run
protocol = GoldenGate(
    name="Combinatorial Golden Gate",
    design=design,
    include=["KanR"],  # only keep circularized plasmids with a KanR SeqFeature
    min_count=5,  # only keep circularized plasmids from >=5 SeqRecords
)
protocol.to_fasta("plasmids.fasta")  # export multi-FASTA
protocol.to_csv("plates.csv")  # export plate layouts
protocol.to_txt("protocol.txt")  # export human protocol
protocol.to_picklists("picklist", platform="hamilton")  # export a hamilton picklist

plasmids.fasta:

>J23100_AB+B0032m_BC+C0012m_CD+B0015_DE+DVK_AE
GGAGTTGACGGCTAGCTCAGTCCTAGGTACAGTGCTAGCTACTAGAGTCACACAGGAAAG
TACTAAATGATGGTGAATGTGAAACCAGTAACGTTATACGATGTCGCAGAGTATGCCGGT
...

plates.csv:

Setup Wells with volumes (uL) shown:
Plate:1,1,2,3,4,5,6,7,8,9,10,11,12
A,B0015_DE(4),C0080_CD(18),R0010_AB(54),water(36)
B,B0015_DE(160),DVK_AE(160),cre_CD(18),water(156)
...

protocol.txt:

Combinatorial GoldenGate:
1. Setup PCR plate with (volumes) shown:
    1.1. Dilute plasmid DNA to 75 ng/µL in 'water'
    1.2. Create 'assembly-mix' from 1:1 T4 Ligase Buffer (10X) and NEB Golden Gate Assembly Mix
...

picklist.gwl:

A;Plate:2;;;15;;2.0;;;
D;Plate:3;;;80;;2.0;;;
W;;;;;;;;;
...

Annotation

In addition to DNA assembly, synbio exposes a plasmid annotation function in synbio.features. An example below shows a SeqRecord being augmented additional SeqFeatures from a curated database of common plasmid features.

from Bio.SeqIO import parse
from synbio.features import annotate

record = next(parse("plasmid.fa", "fasta"))
record_with_features = annotate(record, identity=0.96)

Alternatives

This is a non-exhaustive list. Contact me for a comparison of these libraries/platforms and synbio.

  • Aquarium is an extensive library/application for LIMS, protocol definition/execution, and workflow design. A lab operating system.

  • Autoprotocol is a specification standard for experiments in the life sciences.

  • BioBricks is a general focus, web-based editor for describing experiments in Biology.

  • Biocoder is a C++ library with extensive protocol step definition capabilities.

  • Plateo is a python library for planning, running and checking laboratory experiments. Great for parsing and exporting plates and picklists form multiple formats.

  • pydna is a python DNA assembly simulation library with a human-readable description of clone and assembly strategies.