Mosaik Alignment Problems

[rahul.m.dhodapkar]

Dear SEQanswers community,

I am trying to use MosaikAligner on paired end Illumina reads. I first ran MosaikBuild on my reference sequence, (to get hg18_combined.dat) and then on the paired fastq files
(to get mosaikpaired_test.dat) and then built a jump database from hg18_combined.dat with a hash size of 15. however, when I try to run MosaikAligner, the whole process
freezes up, and MosaikAligner tells me that it needs ~100 days to complete the alignment. The terminal output is documented below:

[rd389@c065 bin]$ ./MosaikAligner -ia hg18_combined.dat -in mosaikpaired_test.dat -out mosaikpaired_test.out.dat -a fast -hs 15 -p 8 -j hg18_combined.dat.jump
------------------------------------------------------------------------------
MosaikAligner 1.0.1388                                              2010-02-01
Michael Stromberg & Wan-Ping Lee  Marth Lab, Boston College Biology Department
------------------------------------------------------------------------------

- Using the following alignment algorithm: single position (fast)
- Using the following alignment mode: aligning reads to all possible locations
- Using a maximum mismatch threshold of 4
- Using a hash size of 15
- Using 8 processors
- Using a jump database for hashing. Storing keys & positions in memory.
- loading jump keys database into memory... finished.
- loading jump positions database into memory... finished.
- loading reference sequence... finished.

Aligning read library (1999937):
 0% [                                     ]    0.2410 reads/s ETA 96.0 days

Is there something about the formatting of my command that is wrong? Or is there another processing step before alignment that I have failed to complete? Hope to hear back from mosaik users.

thanks in advance
-Rahul Dhodapkar

[rahul.m.dhodapkar]

Also, if anyone has the email address of Dr. Stromberg or Dr. Lee, I have been trying to find them, but haven't been able to.

-Rahul

[epigen]

I think if your command was wrongly formated, MOSAIK would have exited with an error instead of starting to align. I've only used it on single end reads so I don't know how it performs on paired end. How long did you wait and did it produce any output?
You could try setting the -bw parameter to a higher value to speed up the mapping. The value must be an odd number and depends on the read length, see manual. And maybe try -ls <radius> to enable local alignment search for PE reads.

The manual says Michael Strömbergs email is mikaels ][at][ bc >.< edu
You could also take a look at http://code.google.com/p/mosaik-aligner/

Good luck

Barbara