Getting Started

Installing LoopDB

LoopDB is available through PyPi, so you can just

pip install loopdb

Setting up a database

LoopDB is based on SQLAlchemy, which allows to use most of the common database back-ends. We recommend using PostgreSQL. After you have installed PostgreSQL, or any other database server, create a new database. With PostgreSQL it can be done with

createdb loopdb

Initialising LoopDB

Create a loopDB instance by

from loopDB import LoopDB

loopDB = LoopDB( 'postgresql:///loopdb', clean = True )

where the first argument is the address of your database, and clean flag determines whether the database should be emptied before initialisation. Default is clean=False.

Setting up a schema

Now it's time to create a schema for the Loop Assembly. This involves setting up the restriction enzyme, restriction enzyme site, base sequence and backbone tables, which will be used for defying and storing parts. Read more about data structure in the Tables section. You have two options to crate a schema, either using a json file or manually. Please refer to the schema-example and manual-example.

Adding a part

You can define a new Level 0 part by using LoopDB.addPart() method, followed by the LoopDB.commit() method

myPart = loopDB.addPart(name = "My Part", seq = "ATG...", backbone = "L0-CDS")
loopDB.commit()

where backbone is a name of a backbone you want to use for your part. If you are adding more than one part, you can commit only onece after you issued several addPart() commands. Alternatively you can load the part from a GenBank file, whcih will preserve and store all the sequence annotations.

genBankPart = loopDB.addPart(name = "My GB Part",
                    gbFile = "part.gb", backbone = "L0-CDS")

Assembling a part

After you have defined several Level 0 parts you can use them to run an assembly using the same addPart() method. The only difference is that instead of specifying the sequence you should pass a children argument containing names of the parts for assembly

myPromoter      = loopDB.addPart(name = "My Promoter",
                        seq = "GGA...", backbone = "L0-Prom5")
myCDS           = loopDB.addPart(name = "My CDS",
                        seq = "ATG...", backbone = "L0-CDS")
myTerminator    = loopDB.addPart(name = "My Part",
                        seq = "TCT...", backbone = "L0-Term3")
myGene          = loopDB.addPart(name = "My Gene",
                        children = [myPromoter, myCDS, myTerminator], backbone = "Ly1")
loopDB.commit()

One you have several Level 1 parts you can use the same method to assemble Level 2 parts and so on. Furthermore, you can mix and match different levels, as long as their backbones have compatible restriction sites. Having said that, make sure that the parts you pass to addPart() method are in right order and are assembled in a compatible backbone, i.e. the 3' overhang of each successive part matches 5' overhang of the previous part; and 3' overhang of the first part and 5' overhang of the last part match the corresponding sequences of the template backbone.

Retrieving parts

You can retrieve an existing part by passing part name to LoopDB.getPart() method. You can also use, getBaseSeq(), getBackbone(), etc. to retrieve existing records for every LoopDB table.

myPart = loopDB.getPart("MyPart")

Further, you can access part's children or retrieve part sequence, by using the following properties

myPart.seq
>> 'ATGGT...'
myPart.fullSeq
>> 'GTAGCAT ATG... GCTGAT'
myPart.children
>> [<tables.Part object at 0x10d5c8b10>, <tables.Part object at 0x10d5c8d50>]

The difference between seq and fullSeq is that the first one will return only the actual sequence of the part, while the second one will return the complete sequence, including that of the backbone. Additionally you can use record and fullRecord properties to get partial or complete Biopython SeqRecord that will include all the annotations.