This is a description of the pipeline designed to do scaffolding 
of fragmented genomes using RNA-seq. The code should run 
on any Unix-like system supporting python 2.6 or better. The code 
is developed on MacOS X on python 2.6.

Note that RNAPATH is not currently optimized for running on machines with 
small or medium amounts of RAM. 32 Gb minimum is recommended for the current 
version.

1. COMMAND LINE OPTIONS
2. MAPPING THE READS AND BUILDING THE RDS FILES
3. GETTING THE SCAFFOLDING READS
4. RUNNING RNAPATH.py


1. COMMAND LINE OPTIONS

To find out more about the settings for each script, type:

python $ERANGEPATH/<scriptname> 

to see the command line options. Note that all ERANGE command-line 
options are case-sensitive & that the scripts typically ignore 
command-line arguments that they do not recognize!


2. MAPPING THE READS AND BUILDING THE RDS FILES

Before running the RNAPATH script on a genome (assumed to be in fasta format), 
you will need to first map the RNA-seq reads using BLAT and import those reads
into an RDS file, as described in README.build-rds . 

3. GETTING THE SCAFFOLDING READS

Once you have an indexed RDS file, use the scriipit distalPairs.py to output 
the list of paired reads that do not map to the same contig. This involves 
specifying a distance to distalPairs.py that is greater than the length of the 
largest existing genomic contig. For example:

python ../commoncode/distalPairs.py 20000 rna_on_genomic.rds rna_on_genomic.crosspairs --splices --cache 20000000

4. RUNNING RNAPATH.py

You can now run RNAPATH.py. I suggest optionally using the included script processvelvet.py to rename the contigs, before running blat and generating the crosspair data.

Example: $ERANGEPATH/rnapath/RNAPATH.py genomic_contigs.fa rna_on_genomic.crosspairs RNAPATH.log genome.RNAPATH.fa

version 3.3    November 2010 - updated command line options
version 3.2    May      2010 - first release