[Genome] related to conservation track

[Rachel Harte]

Dear Xiao,

One of our engineers gave this explanation for what you are seeing:

"The scores are generated by a hidden markov model, which has internal
state.  So instead of considering each base in isolation, the neighborhood
is considered and the model does include a (tunable) smoothing parameter.
In that tiny (11bp) region, the score seems to be slowly decaying from a
previous high.  Gaps (the --'s and ==='s) are treated as missing
information, not as harshly penalized as mismatches I believe.  Also,
there are some matching bases in the most distant species (the run of A's
in t_castaneum) that are probably softening the fall (some precipitous
drops can be seen in the neighborhood, but in those cases most everything
seems to be mismatching at the boundaries)."

I hope that this helps you. Please let us know if you have further
questions.

Rachel

Rachel Harte
UCSC Genome Bioinformatics Group
http://genome.ucsc.edu


On Thu, 31 May 2007 alicexiao.li at utoronto.ca wrote:

>
> Dear UCSC Genome browser colleages:
>
>
> This maybe a very silly question but it confused me alot. :)
>
> When I used ucsc genome brower to identify conserved region for CG4040
> 3'UTR between 12 drospholia species. I found in the region chrX
> 8898540-8898550, the figure sures these region are relatively
> conserved. However, it is obvious that at base level, it is not
> conserved(lots of - and =). Detailed information can be seen  in the
> following website:
>
> http://genome.ucsc.edu/cgi-bin/hgTracks?hgsid=93161117&hgt.out1=1.5x&position=chrX%3A8898530-8898549
>
> So is that actually correct?
>
> cheers
>
> Xiao
>
> _______________________________________________
> Genome maillist  -  Genome at soe.ucsc.edu
> http://www.soe.ucsc.edu/mailman/listinfo/genome
>