rnaseqdata 0.0.8

New Data type known as SeqData for RNA-Seq data analysis

\n# ernav2_seqdata A new data type known as SeqData is designed for RNA-seq data analysis. The data type is designed for data integration from various sources and dimensions:

Expression of RNA is measured by read counts of transcripts. A typical bioinformatics pipeline of mRNA-seq determines reads counts (RC) of transcripts. The RCs are typically 2-D table, of which samples are in rows, and transcripts (or genes) are in columns, or in the reverse. After that, the RC table would be normalized as FPM or FPKM or somewhere else by a certain normalized method. The next, co-founding factors among samples would be removed using a certain method namely DESEQ2 or EdgeR etc. Moreover, those data would be transformed into various table, namely log, or partitioned into some subset. Bioinformatician should manage all those data sets during statistical anlaysis.

Biological scientists may be more care about significance of mRNA-seq data analysis, and what those significance reveals. In this case, sample informations, or patient information, or features of samples (namely single cells) shall be considered. Moreover, aside from transcript ID or Gene ID, other annotations would be integrated, for example, genomic annoations namely chromosome locus, protein annotations namely domain identification would be integrated, too. Those annoation data may not be used in statistical process, but really needed for further study.

SeqData is tree structure. The root contains data of phenotypes and annotations. Each node contains various attributes including X in m x n, and var (statistical aggregations). Nodes inherite the attributes of the root nodes. Data of children nodes is determined by those of parent nodes.

installation

It is convenient to install the repository using pip. The package could be found at Pythone Package Index.

pip install --upgrade rnaseqdata

Development

git clone [e-mail]:Tiezhengyuan/ernav2_seqdata.git
cd ernav2_seqdata

create virtual environment

virtualenv venv
source venv/bin/activate
pip install -r requirements.txt

Unit testing

pytest tests/unittests

quick tourial

In Python3

from rnaseqdata import RootData, SeqData
import numpy as np
import pandas as pd

Create SeqData

root = RootData()
c = SeqData(root)
c.put_data('test', np.eye(3), root)
c.to_df('test)

      0    1    2
 0  1.0  0.0  0.0
 1  0.0  1.0  0.0
 2  0.0  0.0  1.0