Karma-colorspace
Overview
KARMA(K-tuple Alignment with Rapid Matching Algorithm) is able to map 35bp single end color space reads at the speed of approximately reads per hour using Intel Xeon X760 2.66GHz and 128G memory.
We summarize software requirments as following:
- Binary reference genome in nucleotide space (see #sec4.1})
- Binary reference genome and word index in color space (see \ref{sec:2})
- Color space reads in valid color space FASTQ format (see \ref{sec:4.1} for file specification)
- Color space reads are longer than minimum length requirement. (see \ref{sec:4.2})
- Specify color space parameter when starting KARMA (see \ref{sec:3})
Please note the hardwares requirment for KARMA are:
- 20G memory. Due to memory share mechanism, running multiple processes of KARMA on the same machine consumes about the same amount of memory asrunning one process.
- 30G disk space
We listed a complete example reviewing the whole procedure from building word index to mapping color space reads in \ref{sec:5}.
Build Binary Reference Genome and Word Index
First, we need to build binary reference genome (option: --createReference)
\footnote{To let KARMA map nucleotide space reads, you need to use ``--createIndexto create the word index file.}
in nucleotide space. Assume NCBI36.fa is a FASTA file contains sequences of all chromosomes.
The command to invoke is:
karma --createReference --reference NCBI36.fa
Second, we need to build binary reference genome (option: --createReference) and word index (option: --createIndex) in color space. The same FASTA file is needed. However, to avoid naming conflicts, we suggest using word "CS" appending to the base file name for clarity. The command to invoke is:
ln -s NCBI36.fa NCBI36CS.fa karma --colorSpace --createReference --createIndex --reference NCBI36CS.fa
An important parameter is the size of words for indexing.
We recommand 15 (default value) for human reference genome.
Specifiy ``--wordSize N`` if you like to use $N$ as word size.
Typically you will observe performance change (see \ref{sec:4.4} for more discussion).
Note, multiple chromosomes are supported.
In current version, KARMA can take one FASTA file which contains sequences of all chromosomes.
Map Color Space Reads
KARMA takes valid color space FASTQ files inputs.
We usually use suffix .csfastq to distinguish it from nucleotide space reads.
For single end color space read, we can invoke command:
karma --reference NCBI36.fa --csReference NCBI36CS.fa --colorSpace single.csfastq
Mapping results are store in a SAM file named "single.sam".
Multiple input files are also acceptable, e.g.
karma --reference NCBI36.fa --csReference NCBI36CS.fa --colorSpace \ single1.csfastq single2.csfastq single3.csfastq
For paired end color space reads, option "--pairedReads" is requires.
Suppose the paired end reads are stored in file, pair1.csfastq and pair2.csfastq.
The command to invoke is:
karma --reference NCBI36.fa --csReference NCBI36CS.fa --colorSpace \ --pairedReads pair1.csfastq pair2.csfastq
Mapping results are store in a SAM file named "pair1.sam", which contains reads from both files.
Similarly multiple paired end reads files can be specified in command line, and KARMA will pair 1st and 2rd file, 3rd and 4th file and etc.
karma --reference NCBI36.fa --csReference NCBI36CS.fa --colorSpace \ --pairedReads pair1.csfastq pair2.csfastq pair3.csfastq pair4.csfastq
Additional Information
<a name="sec4.1"></a>Input file requirement
KARMA require input files in valid color space FASTQ format.
We require the length of reads(including leading primer) should equal to the length of its quality string.
A valid example of color space FASTQ file:
@Chromosome_20_048435095_Genome_2757096147
A02232200222021320012102212311002212
+
!!1111111111111111111111111111111111
Template:AnchorMinimum read length requirement
Keep in mind that the requirement of minimum color space read length for KARMA is
twice the size of word plus two (including leading primer) \footnote{For nucleotide space,
the minimum length requirement is twice the word size.}.
For example, KARMA use word size of 15 by default, so it will try to map color space
reads that are longer than 32 base pairs.
Template:Anchor Auxiliary tools
ABI SOLiD platform generated FASTA file (e.g. XXX.csfasta) and quality file (e.g. XXX\_QV.qual) separately. We wrote a script, solid2csfastq.py, to convert it to color space FASTQ file(e.g. XXX.csfastq). We believe a single color space FASTQ file will simplify post processing.
Template:Anchor Choose an appropriate size for word index
Size for word index is sensitive to mapping performance. A small size of word index will increase the number of calculation cycles for a single read and duplications of a single word. On the other side, a big size will require much larger memory. Please also keep in mind that appropriate size is related to your hardware architecture. For practically purpose, we found size of 15 is optimal.
A Complete Example
A wrap-up message for quick start mapping color space reads.
Building binary genome reference and word index:
karma --createReference --reference NCBI36.fa ln -s NCBI36.fa NCBI36CS.fa karma --colorSpace --createReference --createIndex --reference NCBI36CS.fa
Mapping color space reads:
karma --reference NCBI36.fa --csReference NCBI36CS.fa --colorSpace single.csfastq karma --reference NCBI36.fa --csReference NCBI36CS.fa --colorSpace \ --pairedReads pair1.csfastq pair2.csfastq
The output files are single.sam and pair1.sam and they conform SAM specification.
