IntMTQ

We provide the source code of paper Platform-integrated mRNA Isoform Quantification for bioinformatics submission.

Goal

• IntMTQ is an integrative method combining isoform expressions from NanoString/Exon-array platforms to provide better quantification of RNA-Seq based transcript abundances.
• The code takes aligned Bam files as input and preprocesses RNA-Seq paired-end reads. Then, it performs integrated penalized model and generates isoform expressions.

Dependencies

The code is written in python 3.6, the following environment is suggested:

• Python 3.6
• CVXPY 1.0.6
• Numpy
• Pandas
• Itertools

To install cvxpy with conda, run the following command.

$ conda config --add channels oxfordcontrol
$ conda install -c cvxgrp cvxpy

Prepare RNA-Seq data

TopHat is applied to do RNA-Seq short read alignment. With the aligned Bam files and hg19 annotation, we can run the below command to generate q matrix (read counts table).

$ python read_counts.py bamfile hg19_2018June18 

Run IntMTQ with NanoString platform integrated

$ python IntMTQ.py NanoStringData.xlsx GeneName_NanoString.txt 

IntMTQ.py will load RNA-Seq data (obtained from RNA-Seq data preparation), NanoString data (NanoStringData.xlsx and GeneName_NanoString.txt). An excel file E1_expression.xlsx will be generated as output which contains the isoform expressions estimated by IntMTQ.