cirseq: A Python repository from susinmotion

CirSeq
---------------------------------------------------------------------------
---------------------------------------------------------------------------

Description
---------------------------------------------------------------------------
CirSeq is an application that generates consensus sequences from tandem
repeat sequence reads. This application accepts sequence data as gzipped
FASTQ files (extension .fastq.gz) and requires a user-supplied reference
sequence in FASTA format (extension .fasta). The output generated by CirSeq
includes the following files:

1. consensus.fastq.gz

Contains all consensus sequences generated in FASTQ format. Per
base quality scores can be converted to estimated error
probabilities using the formula: 10**(3*Quality score/-10).

2. data.sam.gz

Contains all consensus sequences that map to the user-supplied
reference sequence. Data is in SAM format.

3. Q[positive integer <= 41]threshold.txt

Summarizes the counts of bases mapped to each position of the
reference sequence at and above a quality threshold. The default
quality threshold is 20 (estimated error probability of 10**-6).
The columns in this file are as follows: Numeric position
(according to reference sequence provided), the reference base
(wild type sequence according to reference sequence provided) and
four columns with the counts of each base, A, C, G and T,
respectively, observed at each reference position.

4. RepeatLengthDistribution.txt

Contains the counts of reads with repeats from 25 - 99 bases long.
This distribution can aid in diagnosis of issues with sequencing
library preparation. The columns in this file are as follows:
Repeat length (in bases), counts of reads with the indicated
length.

5. QualityMetrics.txt

Summarizes the observed mutation frequency and
transition:transversion rates with respect to quality scores. These
metrics can be used to define an appropriate quality threshold. The
columns in this file are as follows: Quality score, mismatches,
aligned bases, transition mutations and transversion mutations.

6. ProcessingStats.txt

Summarizes important statistics of the sequencing data processing
that may aid in evaluating the quality of sequencing libraries and
diagnosing problems with library preparation and sequencing.

System requirements
---------------------------------------------------------------------------
The following packages are prerequisites for running CirSeq:

1. Python (version 2.7.5) https://www.python.org/ftp/python/2.7.5/Python-2.7.5.tgz
2. Cython (version 0.19.1) http://cython.org/release/Cython-0.19.1.tar.gz
3. NumPy (version 1.7.1) http://sourceforge.net/projects/numpy/files/NumPy/1.7.1/numpy-1.7.1.tar.gz/download
4. SciPy (version 0.13.3) http://sourceforge.net/projects/scipy/files/scipy/0.13.3/scipy-0.13.3.tar.gz/download
5. Bowtie2 (version 2.1.0) http://sourceforge.net/projects/bowtie-bio/files/bowtie2/2.1.0/
6. R (version 3.0.1) OPTIONAL for generation of plotted outputs http://cran.cnr.berkeley.edu/

NOTE 1: Cython requires a compiler. For OSX this may require installation
of Xcode.

NOTE 2: Bowtie2 and R binaries must be in the PATH.

Setup
---------------------------------------------------------------------------
Execute the following command in the script directory to compile the
ConsensusGeneration module:

python setup.py build_ext -—inplace

Usage
---------------------------------------------------------------------------
.[Script directory]/run.sh [output directory] [reference file (FASTA format)] [Script directory] [quality score] [gzipped FASTQ file(s)]

Please cite:
Acevedo, A, Brodsky, L. & Andino, R. Mutational and fitness landscapes of an RNA virus revealed through population sequencing. Nature (2013). doi:10.1038/nature12861

susinmotion/cirseq