Hi Jan,
Is it possible that the true genotype of the sample that reads were
generated from is actually a mosaic of two (or more) genotypes, and your
subsets happen to be dominated slightly, by random chance, by one or the
other? What happens when you align the reads of each subset to each of the
assemblies? Is the rearrangement between contigs, or within contigs?
~Joe
On Tue, Oct 8, 2013 at 10:30 PM, Söderman Jan <jan.soder...@lj.se> wrote:
> Hi,
>
> In order to get acquainted with Mauve I have generated two different sets
> of reads from a single set of E. coli WGS data. My expectation was that
> Progressive Mauve (after applying Contig Mover relative a completed and
> related genome) should have been able to produce LCBs without
> rearrangements, since the two sets only differ with respect to quality
> trimming of the reads. However, using the default settings, Progressive
> Mauve generates a number of rearrangements and two LCBs in the reverse
> orientation. By significantly increasing the Min LCB weight (to >180k) I'm
> able to reduce the number of rearrangements. Nevertheless, one
> rearrangement persists. Neither does the use of different seeds solve this
> problem. Similar results are also obtained using the contigs of one set of
> reads as a reference (in Contig Mover) for the other set of contigs. How do
> I produce confident alignments of reads originating from different E. coli
> isolates, given this difficulty in aligning contigs originating from the
> same sequencing event?
>
> The two sets of reads contained a similar number of paired-end reads
> (2*945k vs. 2*904k) of high quality scores according to FastQC.
> VelvetOptimiser produced 146 contigs from the first set of reads (n50 =
> 169k, longest contig 422k) and 159 contigs from the second set (n50 = 162k,
> longest contig = 268k).
>
> I'm grateful for any input on this matter.
>
> Sincerely,
>
> Jan Söderman
>
>
> ------------------------------------------------------------------------------
> October Webinars: Code for Performance
> Free Intel webinars can help you accelerate application performance.
> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most
> from
> the latest Intel processors and coprocessors. See abstracts and register >
> http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
> _______________________________________________
> Mauve-users mailing list
> Mauve-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/mauve-users
>
--
Joseph Fass
Lead Data Analyst
UC Davis Genome Center - Bioinformatics Core
http://bioinformatics.ucdavis.edu/
jnf...@ucdavis.edu
phone ~ 530.752.2698
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Mauve-users mailing list
Mauve-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mauve-users
