At present, when evidence is aligned (using SIM4?) there are cases where
portions of the evidence are excluded from the alignment, and there is no easy way to know (for example, just by looking at the graphical viewer) if this has occurred.
Two examples of when this might occur:
1) In an EST, if there is a short i.e. 8 bp sequence that matches the sequence
of an upstream exon, the alignment algorithm won't align this 8 bp sequence to
the appropriate exon (because the sequence is too short) but instead will just
extend the alignment to the 5' side of the down stream exon. When we look at
these sequences in the alignment viewer we see either a misalignment or dashes
indicating that there is other sequence but we don't know where it came from.
We can usually resolve this by doing blasts and such, however, this could be
very confusing for your average user.
2) If a piece of evidence extends from one scaffold to the next, there is no way
to know this currently, unless you retrieve and blast the actual EST or cDNA
sequence.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
A nice feature (that would help to alleviate potential confusion due to the
shortcomings in the alignment algorithm) would be some sort of visible flag to
indicate that the evidence did not match the underlying sequence. However, it
would be important to not flag every case of a mismatch, because the sequence
quality of ESTs is not perfect but instead to only flag those cases which had
ends of perhaps >4? bases that do not match.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Logged In: YES
user_id=1326210
Originator: YES
Two examples of when this might occur:
1) In an EST, if there is a short i.e. 8 bp sequence that matches the sequence
of an upstream exon, the alignment algorithm won't align this 8 bp sequence to
the appropriate exon (because the sequence is too short) but instead will just
extend the alignment to the 5' side of the down stream exon. When we look at
these sequences in the alignment viewer we see either a misalignment or dashes
indicating that there is other sequence but we don't know where it came from.
We can usually resolve this by doing blasts and such, however, this could be
very confusing for your average user.
2) If a piece of evidence extends from one scaffold to the next, there is no way
to know this currently, unless you retrieve and blast the actual EST or cDNA
sequence.
Logged In: YES
user_id=1326210
Originator: YES
A nice feature (that would help to alleviate potential confusion due to the
shortcomings in the alignment algorithm) would be some sort of visible flag to
indicate that the evidence did not match the underlying sequence. However, it
would be important to not flag every case of a mismatch, because the sequence
quality of ESTs is not perfect but instead to only flag those cases which had
ends of perhaps >4? bases that do not match.