[Genome] Conservation track again

Archana Thakkapallayil archanat at soe.ucsc.edu
Fri Jul 20 10:22:12 PDT 2007


Hello Stephan,

Here is the response from the person who developed the phastCons algorithm:

The immediate drop in conservation score is caused by the elephant and 
tenrec insertion, which you can see from the alignment below is not very 
conserved between these two species.  The conservation in the following 
11-base block is moderately high but not sufficient to bring the score 
back up.  There are substitutions in the 1st, 3rd, 6th, 7th (minimum 3 
subst), and 11th positions of this block, and some are low probability 
substitutions (e.g., on the branch to macaque).

Attached is the alignment block with insertions (I assume you are 
looking at the multiz17way on hg17) :
Alignment block 1 of 2 in window, 31247796 - 31247872, 77 bps
B D     Human  
ctaaccctaaacaagtgctcaaccct-tgaatgggcc-tggatg-gctcccctggggactgcttcctgc-
B D     Chimp  
ctaaccctaaacaagtgctcaaccct-tgaatgggcc-tggatg-gctccactggggactgcttcctgc-
B D    Rhesus  
ctaaccctaaacaagtgctcaaccct-tgaatgggcc-tggatg-gctcccctggggactacttcctgc-
B D     Mouse  
ctaaccctaaacaagtactcaaccct-tgaatgggcc-aggatg-gctcccctggggacaacttcctgc-
B D       Rat  
ttaaccctaaacaactactcaaccct-tgaatgggcc-tggatg-gctcccctggggggaacttcctgc-
      Rabbit  
ctaaccctaaacaagtgctcaaccctctgtatgggccttggatg-gctcccctggggacagcttcctgc-
B D       Dog  
======================================================================
B D       Cow  
ctaaccctaaacaagtgctcaaccct-tgaatgggcc-tggatg-gctcccctggggacagcttcctgc-
    Elephant  
ct-accctaaacaagtgctcaaccct-tgaatgggcc-tggatg-gctcccctgaggacagcttcctgc-
      Tenrec  
ctaaccctaaacaagtgctcaaccct-tgaatgagcc-tggatgagctcccctgagaccagcttcctgcc
     Opossum  
======================================================================

       Human  ----tccccaacccc----------
       Chimp  ----tccccaacccc----------
      Rhesus  ----tccccagcccc----------
       Mouse  ----tccccaacccc----------
         Rat  ----tccccaacccc----------
      Rabbit  ----ccccccacccc----------
         Dog  =========================
         Cow  ----cccccaacccc----------
    Elephant  -tcccccccagccctc--cagccct
      Tenrec  accaccaccaccccacctcagccct
     Opossum  =========================

I hope this information helps you. Please let us know if you have 
further questions. I apologize for the delay in answering your question.

Regards,

Archana
UCSC Genome Bioinformatics Group

Stephan Struckmann wrote:
> Hello Archana,
>
> I resent the request that I had sent some weeks ago and which had 
> obviously been lost. In the meanwhile I cannot reproduce my problems 
> with BLAT any more, so maybe it has been solved, sorry that I did not 
> check that first. But the other question with the Multiz17 track is 
> still reproducible, maybe the phastCons creator can help.
>
> Thank you very much,
>
> Greetings,
>
> Stephan
>
> --On Dienstag, 17. Juli 2007 17:46 -0700 Archana Thakkapallayil 
> <archanat at soe.ucsc.edu> wrote:
>
>> Hello Stephan,
>>
>> One of our developers have forwarded your question on the Conservation
>> track, to the person who created the phastCons algorithm. I will get 
>> back
>> to you when I have more information.
>>
>> Regarding your second question on BLAT, could you please give us some
>> more information like :
>>
>>  - An example of the sequence
>>  - Are you using hgBlat or standalone BLAT ?
>>  - If you are doing it on your own machine, what parameters and command
>> lines are you using ?
>>  - What do you remember it as doing before?
>>
>> Regards,
>>
>> Archana
>> UCSC Genome Bioinformatics Group.
>>
>>> Hello,
>>>
>>> I have two questions about the conservation track. For example on
>>> <http://genome.ucsc.edu/cgi-bin/hgTracks?hgsid=93864701&hgt.right2=%3E%3 
>>>
>>> E+&position=chr6%3A31247759-31247836>
>>>
>>> Conservation (Multiz17) shows really low values from position 31247861
>>> although most species of that Multiz17 are shown below and obviously 
>>> have
>>> conservation until at least 31247871?
>>>
>>> The second question is about BLAT. If I search for a sequence, that I
>>> received from UCSC before, the hits seem systematically shorter than 
>>> the
>>> query. Is the prolongation part of BLAT that restrictive?
>>>
>>> Thank You for Your always prompt and good answers,
>>>
>>> Stephan Struckmann
>>> ---------------------------------------
>>> Institut für Mathematik und Informatik
>>> E.-M.-A.-Universität Greifswald
>>
>
>
>
> Institut für Mathematik und Informatik
> E.-M.-A.-Universität Greifswald



More information about the Genome mailing list