This document intends to provide usage information for the scripts 'use_express.py' and 'filter_contigs.py'
ln addition to the mailing list, please email me ("Matthew MacManes" <macmanes@gmail.com>) with comments or suggestion.


#############
use_express.py ##
###########

This script takes the assembly Trinity.fasta, and calculates expression (FPKM) using Bowtie2 and eXpress, which need to be installed and in your $PATH.

Bowtie2 download and documentation can be found at: http://bowtie-bio.sourceforge.net/bowtie2/index.shtml
eXpress is documented and can be downloaded at: http://bio.math.berkeley.edu/eXpress/

Usage: python use_express4.py -b input.fa  -t index_name -p [num threads] -X [insert size] -l left.fq -r right.fq -n file.sam]

Required:  

-b file.fa, --b2base=file.fa          fasta file for B2 index (default: Trinity.fasta)
-l file.fq, --left=file.fq                 comma sep list of left reads
-r file.fq, --right=file.fq              comma sep list of right reads

and/or

-U unpaired reads, --unpaired=unpaired reads

Optional

-t index, --target=index             Name of bowtie2 index (default=index)
-p [INT], --threads=[INT]           Number of threads to use (default=2)
-X [INT], --insert=[INT]              Max inner distance (default=500)
-o path to output directory,       --outdir=path to output directory (default=CWD)
-n file.sam,                               full path filename for SAM file from bowtie2 (default=hits.sam)

Running this script should return the file 'results.xprs', which can be used for filtering contigs, as well as for input to edgeR or DESeq.


#############
filter_contigs.py ##
###########

This script takes the assembly Trinity.fasta, and the results.xprs file generated in the 'use_express.py' script, and filters out rare/poorly supported contigs. The resultant file 'High.cov.*.fasta' will contain only those well supported contigs. 

Required: Biopython

Usage: python filter_contigs.py -b Trinity.fasta -F .8

-b Name of Multifasta file. Important, this must be the same file that you used on the above, or the names wont match up properly.
-F The minimum FPKM that you want to retail. Default 1. e.g. transcripts <1 FPMK are filtered out.
