biobricks 0.0.32

Biobricks automates bioinformatics data.

BioBricks

BioBricks is a tool to load data from biological datasets in an easy and automated way.

Installation

To install BioBricks package use the python packagge manager pip.

$ pip install biobricks

Usage

To import BioBricks package use the following state:

import biobricks as bb

The first time you use BioBricks package, an initialization is required with the following process.

1. Set an environment variable BBLIB in your shell with the desired location path of BioBricks library. Be sure you have enough available space to save locally your bricks.

export BBLIB=/opt/biobricks
echo $BBLIB

2. Obtain a BioBricks account from the following link: https://members.biobricks.ai/register. Go to your email an validate your account. Once you succefully login to BioBricks website go to the link https://members.biobricks.ai/token to obtain a token. Copy your token and use it in the following statement:

bb.initialize(<YOUR TOKEN>)

To download a brick and save it locally in your library use the function bb.pull. As an example, Tox21 data dataset can be downlaoded with the following statement.

bb.pull('tox21')

Once the dataset is stored locally in the BioBricks library, the data can be loaded with bb.load function.

tox21 = bb.load('tox21')

This function returns a SimpleNamespace object with all the tables available in the dataset. The tables are store in pyarrow.Table type. To list the available table the in the SimpleNamespace object use:

tablenames = sorted(list(vars(tox21).keys()))
for tablename in tablenames:
    print(tablename)

The list of avaible tables can be visulized with the autocomplete functionality in your python editor.

To convert a pyarrow table to a pandas data frame use the following statement:

ache_p4 = tox21.tox21_ache_p4.to_pandas()

To list the bricks currently available in the BioBricks GitHub repository you can visit:

https://github.com/biobricks-ai

Complete example

import biobricks as bb

bb.pull('tox21')
tox21 = bb.load('tox21')
tablenames = sorted(list(vars(tox21).keys()))
for tablename in tablenames:
    print(tablename)
ache_p4 = tox21.tox21_ache_p4.to_pandas()
print(ache_p4.head())