Re: [wgs-assembler-users] degenerate contigs

[Paul C. <pca...@gm...>]

Hi Brian,

First, I'd like to thank you and the development team at your institution
for making cabog public. I am finding it a very valuable tool to use.

On Sun, Apr 15, 2012 at 2:56 PM, Walenz, Brian <bw...@jc...> wrote:

> Without getting into precise definitions:  Scaffolder (cgw) promotes
> unitigs that looks like unique sequence (based on coverage, length and a
> few other signals) to contigs.


What command line options govern this? Your answer probably depends on what
I'm trying to do. I usually do two types of assemblies:

1) metagenomic (therefore, a complex mixed sample containing sequences from
many species)

2) targeted smaller assemblies with reads that are similar to one species.
Here, I'm trying to make the assembly quicker and hopefully more accurate
by only selecting reads that are similar to one species in hopes to
assembly a complete genome.

Thank you again for you help,

Paul


>  The left over unitigs are available for gap filling as repeats or
> singletons.  The unique contigs are then promoted almost immediately to
> single-contig scaffolds.  With no mates, that's all scaffolder will do.
>  The scaffolds/contigs are output as is, and the left over unitigs are
> output as degenerate contigs.
>
> bri
> --
> Brian Walenz
> Senior Software Engineer
> J. Craig Venter Institute
>
> ________________________________________
> From: Paul Cantalupo [pca...@gm...]
> Sent: Sunday, April 15, 2012 2:53 PM
> To: wgs-assembler-users
> Subject: [wgs-assembler-users] degenerate contigs
>
> Hi
>
> I work with non-paired end 454 sequences. When I perform an assembly, I
> always get a set of regular contigs and degenerate contigs. The celera
> assembler glossary says that degenerate contigs are those unitigs that
> cannot be placed into scaffolds. Well, with my non-paired end data, how can
> *any* contig be placed into a scaffold. Scaffolds cannot be built without
> paired-end data, right?. So, can somebody tell me the difference between a
> "regular" contig and a degenerate contig?
>
> Thank you for your help,
>
> Paul
>
> University of Pittsburgh
> Pittsburgh, PA 15260
>
>
>