Roche soft Error: "At least one read sequence...."

[torbean]

Hello to everyone.

I am facing very strange mistake, that is:

"Error: At least one read sequence must be present in the project
in order to proceed with mapping or assembly."

in RefMapper and in DeNovo, too. I never had it before. I am trying to map two biggest contigs from de novo to a vector sequence... But the run simply does not start... I can't find the description for the mistake in SEQanswers, as well.

Can anyone help, please???!!!

[zjiexin]

About the "Roche soft Error"

Hi torbean,

Did you get this problem solved? I encountered the same problem now and have no clue how to proceed. Any suggestions?

Thanks

[sklages]

What about providing some info?

Which version of Roche Suite? OS?
What kind of input data? Dataset size?
How has the gsMapper been invoked? Command line? GUI?
More surrounding text with the error message?

Sven

[zjiexin]

I partially figured out why the error message: the reads are from Illumina, thus shorter than 50bp. But I read manual for Newbler 2.8, which is the one I am using under Linux x86_64, parameter '-short' is suppose to force the program to take in shorter reads. However, I still got the same error message even with
'-short'. Input data is in fastq file, pair-end, total 170M (each reads is 48bp). Command line, very simple: runAssembly -o outfolder -short fqfile1 fqfile2. I even played with -mi and -ml parameters, not helping. Any suggestions?

[sklages]

hmm, not really. Probably some clipping cuts down the reads below 20bp?
Maybe you can see more if you use '-verbose'?

Just curious, why do you want to use newbler for denovo assembly of quite short illumina reads?
There are other assemblers like SOAPdenovo, velvet among others ...