Hi Sri,
One of our engineers commented:
"The "strand" is a misnomer for that column, and indeed is the relative
orientation. By -A, we are denoting the reverse-complement of sequence
A, i.e. both reversing the order of the bases and complementing A -> T,
C -> G etc.
Your diagram C is correct for '-' (opposite orientation). For 2a, I'm
not quite sure how to interpret that arrow under Block_B... does it mean
that we read + strand bases in the opposite order? In that case, the "+"
strand for Block_A and Block_B have the same bases in the same order, so
I would call that a "+". This is how I would expand diagram C to show a
relative-orientation="-":
Block_A Block_B
5' 3' 5' 3'
--------------> -------------->
(+) ACGTTGACAATGTCA (+) TGACATTGTCAACGT
(-) TGCAACTGTTACAGT (-) ACTGTAACAGTTGCA
<--------------<--------------
3' 5' 3' 5'
Block_A and Block_B are double-stranded sequences. Block_A's (+)
sequence, read left-to-right, matches Block_B's (-) sequence read right
to left. It is also true that Block_A's (-) sequence read right-to-left
matches Block_B's (+) sequence read left to right."
Please let us know if you have any additional questions: [email protected]
-
Greg Roe
UCSC Genome Bioinformatics Group
On 8/15/11 1:58 PM, Sampath, Srirangan wrote:
> Hi katrina,
> Thank you for your response. The Seg-dups strand info is
> still not clear. When the relative orientation is '+', then it is always +A ~
> +B. One never sees -A ~ -B, as this is equivalent to +A ~ +B. And similarly I
> never saw -A ~ +B (it was always +A ~ -B). Now the question is, is the 'Other
> Position Relative Orientation' based on 'strand' or the orientation of the
> sequence with respect to each other?
>
> Consider the following scenarios..and tell me which one is appropriate. My
> impression is that relative orientation should be a reverse orientation of
> the sequence and has noting to do with strand (situation 2a). But please
> clarify.
>
>
> 1) +A ~ +B<==> -A ~ -B<==> relative orientation = +
> The sequences are in the same orientation and strand (<-<- / -> -> )
>
>
> Block_A Block_B
> --------------> -------------->
> (+) ACGTTGACAATGTCA (+) ACGTTGACAATGTCA
> (-) TGCAACTGTTACAGT (-) TGCAACTGTTACAGT
>
>
> 2) -A ~ +B<==> +A ~ -B<==> relative orientation = -
>
> (a) This is simply a reverse orientation of the sequence (<- -> / -> <- )
>
> Block_A Block_B
> --------------> <--------------
> (+) ACGTTGACAATGTCA (+) ACTGTAACAGTTGCA
> (-) TGCAACTGTTACAGT (-) TGACATTGTCAACGT
>
>
> or
>
> (b) This is what you implied, simple flipping of the strands, although in the
> same orientation. Similar to 1 (<-<- / -> -> )
>
> Block_A Block_B
> -------------->
> (+) ACGTTGACAATGTCA (+) TGCAACTGTTACAGT
> (-) TGCAACTGTTACAGT (-) ACGTTGACAATGTCA
> -------------->
> or
>
> (C) There is reverse orientation and strand flipping
>
> Block_A Block_B
>
> -------------->
> (+) ACGTTGACAATGTCA (+) TGACATTGTCAACGT
> (-) TGCAACTGTTACAGT (-) ACTGTAACAGTTGCA
> <--------------
>
>
> Thank you,
> Sri.
>
> On Aug 15, 2011, at 9:48 AM, Katrina Learned wrote:
>
> Hi Sri,
>
> 1) The one base pair difference in the start positions is due to our
> 0-based coordinate system. Please see this FAQ for more information:
> http://genome.ucsc.edu/FAQ/FAQtracks.html#tracks1
>
> 2) The "strand:" information that you see on the item's details page is
> there in error; one of our engineers has removed it. Sorry for the
> confusion and thank you for bringing it to our attention! The 'strand'
> column in the genomicSuperDups table is represented in the item details
> by "Other Position Relative Orientation:." The items in this track are
> spans of double-stranded genomic DNA -- they don't have an inherent
> strand like genes etc. However, if the forward strand sequence of the
> region aligns to the reverse strand of the other region, then the
> relative orientation is '-'. And that's a reciprocal relationship. So
> say region A has other-region B, + and - are the stranded sequences, and
> ~ means "aligns to":
>
> +A ~ +B<==> -A ~ -B<==> relative orientation = +
>
> -A ~ +B<==> +A ~ -B<==> relative orientation = -
>
> 3) We suggest checking the references listed in the description page for
> more details
> (http://genome.ucsc.edu/cgi-bin/hgTrackUi?db=hg19&g=genomicSuperDups) or
> contacting the authors (the email is on the description page also), but
> on of our engineers thinks that based on this statement in the
> description page's Methods section (emphasis added):
>
> The repeats were then reinserted into the pairwise alignments, the
> ends of alignments trimmed, and global alignments were generated.
>
> that the trimming of alignment ends might explain the shorter length of
> the "optimal global alignment."
>
> I hope this information is helpful! Please contact the mail list
> ([email protected]) again if you have any further questions.
>
> Katrina Learned
> UCSC Genome Bioinformatics Group
>
>
>
> On 8/10/11 9:38 AM, Sampath, Srirangan wrote:
> Below are some questions regarding hg18 Seg-dup tracks.
>
> 1) I am using Table browser to download the 'genomicSuperDups' and using 'all
> fields from selected table' to extract all the fields in the schema. There is
> 1bp shift in the 'chromStart' and 'otherStart' columns between the data
> downloaded from Table Browser and the corresponding information from the
> Genome Browser. The start positions of the intervals in Genome Browser is
> shifted one base ahead of the Table Browser. The End positions do match. Why
> is that and which one is correct?
>
> Below I have pasted the both the Table Browser rows and the corresponding
> link to the Genome Browser
>
> Table Browser:
> bin chrom chromStart chromEnd name score strand otherChrom otherStart
> otherEnd otherSize uid posBasesHit testResult verdict chits ccov alignfile
> alignL indelN indelS alignB matchB mismatchB transitionsB transversionsB
> fracMatch fracMatchIndel jcK k2K
> 118 chr10 47833296 47862414 chr10:45491750 419130 + chr10 45491750 45520834
> 135374737 0 1000 N/A N/A N/A N/A align_both//0004/both020825 28953 63 298
> 15276 28292 661 457 204 0.97717 0.975048 0.0231848 0.0232378
>
> Genome Browser:
> http://www.genome.ucsc.edu/cgi-bin/hgc?hgsid=206552079&o=47833296&t=47862414&g=genomicSuperDups&i=chr10%3A45491750
>
>
> 2) There is a column with 'strand' information on the 'genomicSuperDups'
> schema (Table Browser). But there is no 'Other Position Relative Orientation'
> data. Is the 'strand' column from Table Browser the actual strand information
> of the sequence, or relative orientation of the 'other position'? From the
> Table Browser output when I manually checked a few, all '+' seem to have '+'
> relative orientation of the other position , and all '-' seem to have '-'
> relative orientation of the other position.
>
>
> 3) Sometimes the link to the 'Optimal Global Alignment' does not provide the
> Alignment for the entire interval. For example, the track for
> Item: chr10:48284311
> is 428kb long. But the 'Optimal Global Alignment' provides alignment for only
> 103 kb. Below are the links
>
> http://www.genome.ucsc.edu/cgi-bin/hgc?hgsid=206563651&o=45896971&t=46325469&g=genomicSuperDups&i=chr10%3A48284311
> http://humanparalogy.gs.washington.edu/build36/align_both//0004/both020916
>
>
>
> Thank you,
> Sri.
>
>
>
> Dr. Srirangan Sampath Ph.D.,
> ABMG Clinical Cytogenetics Fellow
> Medical Genetics Laboratories
> Department of Molecular& Human Genetics
> John P. McGovern Campus NABS-O250
> Baylor College of Medicine,
> Houston, TX-77021
> Cell: 504-390-5512
>
> _______________________________________________
> Genome maillist [email protected]
> https://lists.soe.ucsc.edu/mailman/listinfo/genome
>
> Dr. Srirangan Sampath Ph.D.,
> ABMG Clinical Cytogenetics Fellow
> Medical Genetics Laboratories
> Department of Molecular& Human Genetics
> John P. McGovern Campus NABS-O250
> Baylor College of Medicine,
> Houston, TX-77021
> Cell: 504-390-5512
>
> _______________________________________________
> Genome maillist - [email protected]
> https://lists.soe.ucsc.edu/mailman/listinfo/genome
_______________________________________________
Genome maillist - [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome
