Coverage of the simulated alignments

[Javier Herrero]

Hi Dent et al

I have a quick question about the simulated alignments
(http://compbio.soe.ucsc.edu/alignathon/data/simMammals.ancestor.maf.gz). I
am finding that the simulated genomes are not fully covered in this MAF
file. I am guessing this is due to lineage specific insertions. Is this
correct?

Thanks

Javier

--
Javier Herrero, PhD
Ensembl Coordinator and Ensembl Compara Project Leader
European Bioinformatics Institute (EMBL-EBI)
Wellcome Trust Genome Campus, Hinxton
Cambridge - CB10 1SD - UK

[Dent Earl]

Hey Javier,

Without looking at an exact example, I'm 99.9% sure your guess is correct. Evolver allows for tandem duplication expansions and only keeps track a homology relationship between the first duplication and the ancestor (this is a limitation of the simulator). Evolver also allows for random DNA insertion events. Furthermore, inserted material from the mobile element library is not considered homologous to anything at the time of insertion. As a result you can get lineage specific DNA. In the case of the ME insertions you can even have lineage specific DNA (in the eyes of the simulation) that is nearly identical to DNA in another lineage (because they come from the same mobile element family).

We tried to prevent this over-homologization in as realistic a way possible by taking the ME library from near the human lineage and using it as an input to RepeatMasker, allowing most of the MEs to be masked out of the simulated genomes.

d

> --
> You received this message because you are subscribed to the Google Groups "Alignathon" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to alignathon+...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>