parisepigenetics/rna_feat_ext

Software tools to extract mRNA's features from a list of ENSEMBL gene IDs.

mRNA feature extraction tools

authors: Costas BOUYIOUKOS, Antoine LU and Arnold Franz AKE

A set of computational tools to extract user defined mRNA features from a list of ENSEMBL gene IDs by using the web API of ENSEMBL BioMart and custom computations. Conceived and developed by Costas Bouyioukos @cbouyio at Paris Epigenetics @parisepigenetics and Universite Paris Diderot. Development involved two bioinformatics master students: Antoine LU @antoinezl who started as part of a coding project during his second year in the degree and Franz-Arnold AKE @franzx5 a second year Master's degree student who mainly worked on the clustering part of the project.

Installation.

To install the tools in your local python environment (user $HOME directory) type:

./setup.py install --user

(the --user flag installs the software on your personal account (no root privileges required).

Requirements.

Python.

• Biopython
• Pandas
• Biomart

All are available for installation via pip install <package_name>

External.

• RNA Vienna package
• MEME Suite

For external tools please follow the installation guidelines in the provided links.

Main Usage.

geneIDs2fasta.py ENSEMBL_geneIDs_file fasta_output_file

and

fasta2table.py ENSEMBL_fasta_output_file features_table_file

geneIDs2fasta.py

This program takes a text file with a list of ENSEMBL gene IDs and returns a FASTA formatted file of the corresponding cDNA sequences. The header is formatted and contains various metadata ordered as:

>ENSEMBL_transcript_ID |Gene stable ID | Gene name | cDNA start | cDNA end | TSL | APRIS | HAVANA_ENSEMBL | gene description | Source:|

fasta2table.py

This program takes the fasta formatted file returned by the previous script geneIDs2fasta in input, and return a semicolon separated table with the following header:

ensembl_gene_id;gene_name;coding_len;5pUTR_len;5pUTR_GC;5pUTR_MFE;5pUTR_MfeBP;3pUTR_len;3pUTR_GC;3pUTR_MFE;3pUTR_MfeBP;TOP_localScore;CAI;Kozak_Sequence;Kozak_Context

Testing

Test directory contains two test files to test and demonstrate the functionality of the tools.

• test/testENSEMBLids.txt Contains 6 genes with their ENSEMBL IDs.

• test/testTransExpr.csv Contains the expression levels of each individual transcript of the above genes from a case study.

TODO add section for MEME suite integration.

rna_feat_ext's Issues

Library and script to retrieve data from ENSEMBL

We need to consolidate all the code we have so we end up with something as useful as possible.

1. We need a pure python library that will include all the functions (perhaps some classes too).
2. We well need a script to execute and retrieve a query from the Biomart database API.
   The script will get nothing else than a list of ENSEMBL identifiers (either gene or transcript) and will output a table with all the features we need. without doing any computations.
   So no computations of energy or distance or anything else for the moment.
   later we will include the libraries for computation. (MFE, RBPBS and miRNAs).

Produce a setup.py way to install the software, and its requirements.

A proper way to install the software

Introduce Makefile

A minimal Makefile to handle installation, cleaning and documentation.

import local_score error

in the script rnafeatureslib it want to import local_score but the module is no where to be found, is that a in-house module or third-party module to download

How to Deal with NA values for getting Features Tables

Hello !
the question is in the Object .
How to deal with the NA values in order to get feature table ?
for example several rows for the same transcript but with NA values at many features...

Thanks

Check min-max size of UTRs.

RNAfold does not process UTRs smaller that 8 and longer than 10000 bps.
Include this checks while extracting features and printing the UTR files.

Introduce an option to retrieve the 3' and 5' UTR sites.

Clean up the branches of the git repository!

Produce a meaningful README.
Provide a small HOW TO with a small working test case.

Transcript abundance Selection

Following up Issue #7

*Later we can also provide an isoform percentage expression table (from tools like RSEM or Stringtie).

Put a transcript selection option.

Select transcripts based on isoforms expression percentages.

-- Actually for the moment we will only consider the most well annotated transcript from ENSEMBL based on the database criteria, Havana, APRIS and TSL