Computational Genomics Gene prediction pipeline

This pipeline is designed by team2-group1, to predict genes of the samples from team1 using a number of Gene prediction tools. This pipeline is used to generate a merged result from several tools.

Gene Prediction Pipeline

Requirements

python3
Latest Perl
bedtools
samtools
Latest Prodigal
Latest GeneMarkS-2
Latest Aragorn
Latest Barrnap
Latest Biopython (If running Bedtools)

bedtools and samtools are required for the union of GeneMarkS-2 and Prodigal results
All required tools need to be installed properly and added to $PATH

Quick Start

-f :Path to file input directory (Required)
-p :Run Prodigal prokaryotic mRNA gene prediction tool
-g :Run GeneMarkS-2 prokaryotic mRNA gene prediction tool
-nc :Run Aragorn and Barrnap to predict tRNA/tmRNA and rRNA (respectively) (optional)
-ncs :Separate Aragorn and Barrnap results into two distinct sets of nucleotide fasta files

Default behavior will still require -f and will run both Prodigal and GeneMarkS-2 with Bedtools
bedtools will run if both Prodigal and GeneMarkS-2 are run, and includes a union folder of both tools
Example usage: ./geneprediction_pipeline_t1.py -f <input_dir>

Output Description

Prodigal and GeneMarkS-2 run individually will be found in their respective folders, ./prodigalresults or ./gms2results
Output files are split into three folders. One for GFF format, fna and faa.
If Prodigal and GeneMarkS-2 are run in tandem, then the combined output will also be in ./prodigal-genemark

Aragorn and Barrnap results are joined by default into single .fna files by assembly, located in ./arabarr

Nucleotide and Amino acid fasta formats may be used with BLAST homology validation as described below.

Blast ( For validation )

Requirements

Version_5 database (required)
taxonomic_id list (required)
EDirect command-line utility (required)
Latest Perl (required)
python3 (required)
blast+ (required)

The validation part assume that all the requirements are installed and the tools should be added to $PATH.

Quick start

For downloading database:
Use ./update_blastdb.pl --blastdb_version 5 --showall to see the option.
Use ./update_blastdb.pl --blastdb_version 5 [Database] --decompress to download.

For getting the taxonomy_idlist:
Use get_species_taxids.sh -n [organism]

For blastp (amino acid):
./blastp.py -d [queried_fold] -t [taxonomy_idlist] -o [outputfolder]
or blastx (DNA seqs):
./blastx.py -d [queried_fold] -t [taxonomy_idlist] -o [outputfolder]

Description of argument

For blastp.py or blastx.py:
-d :the folder that contains only fasta files you want to validate.
-t :the taxonomy_idlist for specific organism.
-o :the output folder for your outputs.

For validationP.py or validationX.py:
-s :the folder that contains only fasta files you want to validate.
-b :the folder that contains only blast results for your fasta files.
-o :the output folder for your outputs.

Output Description

There will be two folders in your output folder:
knownprotein/ : The fasta files in this folder have got rid of the sequences that do not have hit in blast.
novelgene/ : The fasta files in this folder do not have hit in blast.