#31 alternative transcripts

[Valerie Wood]

Need to curate some alternative transcript examples for Chado loading

can do some from the gene expression list:

gene expression, alternative splicing (3)
gene expression, alternative transcript, alternative start (1)
gene expression, alternative transcripts (7)
gene expression, alternative transcripts detected (1)
gene expression, alternative transcripts, alternative initiation site (1)
gene expression, alternative transcripts, alternative polyadenylation (2)

[Valerie Wood]

tco1 associated transcript SPNCRNA.600 is another good special case example

[Valerie Wood]

People might like to refer to them by a name if they work on specific transcripts in a paper

SPAC1556.06.1 are a bit clunky.

we can think about this.....e.g meu1.1

[Valerie Wood]

also do apn1. This is an odd example. The major transcript is pseudo, but there are minor transcripts which code for full length protein

[Valerie Wood]

see also spd1
https://sourceforge.net/tracker/?func=detail&aid=3459352&group_id=65526&atid=2096421

[Valerie Wood]

See GeneDB listed

also
/ID="SPAC31G5.09c"
/note="alternative UTRs for this feature are represented by different polyA sites"

[Valerie Wood]

check list from trt1 page
under gene expression
(add all to this list to collect?)

[Valerie Wood]

Another example:

Going in the Right Direction: Mating-type Switching of Schizosaccharomyces pombe is Controlled by Judicious Expression of Two Different swi2 Transcripts.
Yu C, Bonaduce MJ, Klar AJ.
PMID: 22209903

[Valerie Wood]

example
https://rt.sanger.ac.uk/Ticket/Display.html?id=211679

[Valerie Wood]

from arnaud
looks like the EST alignments show evidences for splice variants,
e.g. for PTF1, it seems there is an alternative donor.
http://demo4.ensemblgenomes.org/Schizosaccharomyces_pombe/Location/View?db=core;g=SPAC823.14;r=I:2605757-2607414;t=SPAC823.14.1;time=1319813407502.502

(others could be identified with a script)

[Valerie Wood]

Submission of alternative transcripts to EMBL:
JQ308183 JQ308183 1962 17-JAN-2012 18:46:03 S.pombe Swi2S (swi2) mRNA, complete cds.
JQ308182 JQ308182 2476 17-JAN-2012 18:46:03 S.pombe Swi2L (swi2) mRNA, complete cds.

[Valerie Wood]

need to sort this
From antonia:
should meu-1-1 really be called this and not meu1? (meu1-1 sounds like an allele)

[Valerie Wood]

spd1 also has alternative transcripts

[Valerie Wood]

A couple more alternative transcript examples came out of the PHAF FILE creation.
SPAC1556.06 is meu-1-1 above

n 25/01/2014 08:41, Kim Rutherford wrote:

There are 6 phenotype annotations that aren't being written out. 4 are
from alleles of SPCC622.17, which is a pseudogene. Should they be
included in the PHAF file?

Ah right I want to change this. This gene has 2 transcripts. The major transcript produces a non functional protein, and the minor transcript produces the correct protein, but the activity is negligible. The author seemed to think it was best to represent the major form. I don't think that was a good plan so we will change this.
We can use this as an example of isoforms when we start curating them. This is a case where the GO annotations will need to be represented on the correct transcript using column 17

The other two are missing because of a bug. Both annotations are from
SPAC1556.06, which has two transcripts. Some part of the loader is
connecting the allele with one of the transcripts instead of the gene,
which is screwing things up. Also the PomBase page for SPAC1556.06 is
not showing the phenotype annotations either, probably for the same
reason - it's stored incorrectly in Chado. I'll make a ticket about
that. It's good that we caught it.

This is similar, but in this case I think the allele should be on the gene, or at least we should only describe alleles consistently with respect to the 'canonical gene' (transcript 1)?

I'll put a ticket on the curation tracker... we will have a stab at creating the transcripts required, and annotating these. Then we'll create tickets for any changes required by the loaders etc

[Antonia Lock]

SPAC1A6.03c from duncan mata paper
Annotated intron retained in most reads (but a few trans-reads detected); when retained truncated protein is produced

we have annotated the smaller version

[Valerie Wood]

PMID:7909513

The cyc]7+ transcript
induced in h90 cells formed a broad band and actually
comprised two bands with the sizes of 3.2 and 3.0 kb. In
h- haploid cells, however, the 3 kb transcript was
undetectable under any conditions we examined. Clearly,
the two transcripts were derived from the cycl7+ gene
since both bands were totally undetectable in the cyc7
disruptant (data not shown). In h90 resl- cells, the basal
level of the 3.2 kb transcript was lower than that in h90 wild
type cells, but the 3 kb transcript was properly induced upon
nitrogen starvation though the maximum level of induction
was slightly lower than in h90 wild cells. However, in h90
res2- cells, although the level of the 3.2 kb transcript was
not significantly changed, the induction of the 3 kb transcript
was totally abolished. These results indicate that the cycl 7+
gene, expressed as two major transcripts, is regulated by
the resl + and res2+ cell cycle 'start' genes.

Extesions will need to be updated for specific transcript ID, details are in comment field for this session

[Valerie Wood]

uvi15

pombase has 791 bp primary transcript annotated, but that's probably not right
from PMID:7898433 - txn start mainly -50, some -53; polyA at 980, 1175
from PMID:10954610 - 54 bp 1186-1239 (inclusive) important for reg, so initial transcript must extend at least that far (but note this is beyond polyA, so not part of mature mRNA)
also, primary transcript goes at least to 1287 (fig 7)

[Valerie Wood]

Add nup98 alternative transcript based on
DDJB:LC043101
Makes a N-terminally truncated version of the protein.