Clarification on the “orientation” of chromosomal rearrangements in mate-pair or paired-end sequencing

Question

[alec_djinn]

I need some clarifications on the concept of "orientation" in case of chromosomal rearrangements. Given a DELETION event on a chromosome for example, is the resulting DNA at the breakpoint always in TH (Tail to Head) orientation? Do all the DUPLICATIONS lead to HT orientation? And in case of TRANSLOCATIONS, INVERSIONS etc..?

Can someone give a clear definition of "orientation", and make some examples?

[Devon Ryan]

You can probably figure this out yourself with a basic understanding of genomics and some pieces of paper with 5' and 3' on either end like:

5'---------------3'
3'---------------5'

Since you can't have 3' next to 3' or 5' next to 5', you'll quickly find the possibilities highly limited.

[alec_djinn]

Well, I was interested in having some example related to mate-pair sequencing results and how best interpret that. For example which kind of alignment raise a TRANSLOCATION HH or TT to HT call?

[Devon Ryan]

Head of what and tail of what? I can only guess that you mean the alignments, but perhaps you mean the affected regions. Normally we talk about alignment orientation in terms of "point toward each other" or "point away from each other". So mate pairs will point away from each other (e.g., the more upstream alignment will map as the reverse complement while the more downstream read will map as originally sequenced). Given that, I'll term a standard mate-pair orientation as TT. So that will be the case for deletions. Invertions will be HT or TH. Duplications could have any orientation. Translocations could have any orientation, particularly since that has no meaning when looking across chromosomes.

But of course, that's all assuming what you mean by tail and head, which has no standard meaning in this context.

[Istvan Albert]

There are two different issues here, one is the genomic rearrangement the other is the orientation of the fragments of the sequencing technology.

The two together can be quite confusing. Typically you need to establish what the expected orientation of the fragment is supposed to be, then arrangement may change that in various ways. The challenge is to note that you are matching reality against the "wrong" genome that we happen to call "reference".

[alec_djinn]

Ok.. we are getting closer to my point! So, my confusion is exactly on the point Head and Tail of what? Of a gene (promoter->terminator) or simply of the DNA strain (5'-3')? If you can suggest some links where I can read more about I would really appreciate.

[Istvan Albert]

I am guessing what you mean by HH and HT read orientation is the following: A pair -----------> <---------- would be Tail to Tail (paired end sequencing) A pair <---------- -----------> would be Head to Head (mate pair sequencing)

Still in for both cases insertions, deletions and duplications would not change the read pair orientation. An inversion however would.

[Devon Ryan]

Funnily, I would term the first one head to head and the second tail to tail, since the tail of an arrow is opposite to where the point is...yet again demonstrating that we should just never use this terminology :)

[Istvan Albert]

I would agree, but my life scientist colleagues call it the other way - thus I adopted that world view.

Head refers to the property/start of the underlying feature.

[alec_djinn]

You see that it confusing!

[Devon Ryan]

We don't normally use the terms "head" and "tail" in this context for exactly this reason :)

The only time I normally see head and tail used are when talking about tandem duplications. In that case, people will say "gene foo was generated by a head to tail tandem duplication of gene bar...". In those cases, the terms are separate from the technology used to discover the duplication.

Perhaps you could give an example of where you're running into this terminology.

[alec_djinn]

This is an example of data I am working with..

X    148738748    148739352    X    148806326    148806798    DB151(42)    TT(24)tt(18)    42    100    -51    0.635055    inversion
2    104782844    104783297    5    157548072    157548507    DB151(33)    HT(16)ht(17)    33    100    -18    0.751756    translocation
12    75478730    75479062    12    75488089    75488588    DB151(27)    TT(13)tt(14)    27    100    -13    0.597651    inversion
15    72922456    72922889    15    74399400    74399838    DB151(25)    HH(11)hh(14)    25    100    -132    0.778531    inversion
9    32080984    32081286    9    32090035    32090474    DB151(24)    HT(17)ht(7)    24    100    -22    0.764102    evertion
2    44750437    44750764    2    44768241    44768689    DB151(24)    HT(15)ht(9)    24    100    -22    0.73682    evertion
X    154404121    154404453    X    154412595    154412986    DB151(22)    HT(12)ht(10)    22    100    -53    0.671301    evertion
19    21463508    21463868    19    21474996    21475301    DB151(22)    TH(14)th(8)    22    100    -77    0.809107    deletion
9    38992802    38993073    9    40105522    40105985    DB151(21)    TT(7)tt(14)    21    100    -30    0.735062    inversion
6    92033928    92034276    6    92096994    92097376    DB151(21)    TT(12)tt(9)    21    100    -33    0.744563    inversion

or this one

##REFERENCE ID    C1    START_C1    END_C1    C2    START_C2    END_C2    CNV_TYPE    ORIENTATION
s1227    1    564224    564224    1    8104812    8104812    DELETION    TH
s1884    1    564224    564224    1    813205    813205    DELETION    TH
s2953    1    746408    746408    1    3315099    3315099    DELETION    TH
s2957    1    746408    746408    1    2807254    2807254    DELETION    TH
s565    1    794395    794395    1    2577764    2577764    DELETION    TH
s210    1    812805    812805    1    10357126    10357126    DELETION    TH
s2209    1    1217784    1217784    1    1464670    1464670    DUPLICATION    HT
s2953    1    1235826    1235826    1    1565945    1565945    DELETION    TH
s2957    1    1235826    1235826    1    1666621    1666621    DELETION    TH
s1051    1    1585374    1585374    1    2121131    2121131    DUPLICATION    HT

[Istvan Albert]

what you need it to go to the software documentation is establish what does TH and HT mean from the software author's point of view. As we discussed above these terms are not standardized.

[alec_djinn]

Ok. Thank you for all your comments guys, it helped me to understand better the situation.