View Single Post
Old 06-21-2010, 09:10 AM   #2
epigen
Senior Member
 
Location: Germany

Join Date: May 2010
Posts: 101
Default Do not convert color-space data to nucleotides before mapping!

Some useful threads about the color space to nucleotide issue:
http://seqanswers.com/forums/showthread.php?t=2312
http://seqanswers.com/forums/showthread.php?t=2226

and about miRNA mapping:
http://seqanswers.com/forums/showthread.php?t=2182

ABI's SOLID analysis toolbox, BioScope, provides an easy-to-use interface but installation is a bit tricky (does not work properly on a cluster with CPUs distributed over several nodes). Pre-BioScope tools are available open source.
Other software is command-line only but has better user support. The developers are active on this forum. If you have limited computational equipment, you might consider Bowtie. It can be run on a normal PC and is very fast, but it does not align with indels. Otherwise, if you want a fast mapping tool with indels, I'd recommend BWA. There are also BFAST and NovoalignCS, which are slower but may be more sensitive because they align in real color space. (For BWA, the reads have to be pre-processed with its solid2fastq script, converting color space codes 0123. into ACGTN for alignment purposes. This is somewhat confusing because unaligned reads are reported in this pseudo-nucleotide code whereas aligned ones are converted into real DNA sequences.)
epigen is offline   Reply With Quote