UTRs task1 Mata poly A sites

[ValWood]

Conditions: 972 h-cells were grown in Edinburgh Minimal Medium (EMM) at 32 °C

I'm trying to establish the best way to get the cluster peak site into Artemis.

headings are

Systematic name	Common name	Chromosome	Strand	UTR start	UTR end	Cluster number	Cluster start	Cluster length	Peak position	Total reads in cluster	Fraction of reads in cluster
SPAC1002.02	mug31	1	+	1799818	1	1	244	45	248	135	0.97826087
SPAC1002.03c	gls2	1	-	1799715	1	1	84	121	164	266	0.84984026
SPAC1002.04c	taf11	1	-	1803424	1	1	252	54	288	111	0.77622378
SPAC1002.04c	taf11	1	-	1803424	2	2	145	12	149	31	0.21678322
SPAC1002.05c	jmj2	1	-	1804348	1	1	41	46	63	61	1

I can't find the information about precisely what the column headings mean? Specifically:

1. How UTR end is defined (this always seems to be a small integer and is exactly the same as the data in next column "cluster number" @juanmatacambridge is this correct?)
2. Do we just need to take UTR start and add or subtract 'peak position' (dependent on strand) to obtain the genome position of the cluster peak ?
3. Does anyone recall where the data files which provided the genome browser data came from, it does not seem to be the files provided in the manuscript as this dataset has multiple peaks but in the SUPP data spreadsheet there is a much smaller number of peaks

e.g.

SPAC12B10.01c |   | 1 | - | 4570105 | 1 | 1 | 28 | 85 | 108 | 110 | 0.96491228

has a single peak position in the spreadsheet, but in the browser there are many clusters called:

https://www.pombase.org/jbrowse/?loc=I%3A4558733..4578733&tracks=PomBase%20forward%20strand%20features%2CPomBase%20reverse%20strand%20features%2CPoly(A)%20sites%20during%20vegetative%20growth%20(forward%20strand)%20-%20Mata%20(2013)%2CPoly(A)%20sites%20during%20vegetative%20growth%20(reverse%20strand)%20-%20Mata%20(2013)&highlight=

(unless I am misunderstanding what I am looking at)?

[juanmatacambridge]

To compile a table of annotated CS (or TTS) with a single number you should use the data in table S7, which gives the length of the 3'-UTRs as annotated (at the time of writing), as defined by a cluster of CS, or as defined by the most distant CS. The second data set (cluster) is probably the best, as it gives the length of the 'typical' 3'-UTR, Simply add the length of the 3'-UTR to your coordinates (end of CDS). These data have been validated independently by the Crammer lab (https://www.ncbi.nlm.nih.gov/pubmed/26883383) - the median difference between their data and our dataset is 3 nt.

[ValWood]

For the 3' UTRs, @kimrutherford will create features using

the existing 3' end of the CDS and Dataset7 /columnD / Length of Longest UTR defined by cluster peaks from 2013RNABIOL0135R-Supdata.xlsx

This means I will only need to make decisions for the TSS ;)

Need feature type 3'UTR /SO=SO:0000205 /systematic_ID=xxxxxxx /db_xref=PMID:23900342 /note="feature created using major cleavage site peak, vegetative growth, EMM"

~We could even link to the tracks using the coordinates from the transcript section~

[kimrutherford]

We could even link to the tracks using the coordinates from the transcript section

I don't understand that bit.

[kimrutherford]

Out of curiosity, I pulled out the longest UTR from the dataset, which is:

SPCC320.06         |            | 5636            |               5161 | 5267  

It doesn't seem to match what's in JBrowse?: https://www.pombase.org/jbrowse/?loc=III%3A147551..163550&tracks=PomBase%20forward%20strand%20features%2CPomBase%20reverse%20strand%20features%2CPoly(A)%20sites%20during%20vegetative%20growth%20(forward%20strand)%20-%20Mata%20(2013)%2CPoly(A)%20sites%20during%20vegetative%20growth%20(reverse%20strand)%20-%20Mata%20(2013)&highlight=

I thought there should be a peak at 158577-5161 = 153416

[kimrutherford]

There were a few entries in the dataset that used synonyms. Those were easy to fix. There are four other IDs that I need help with:

I can't find these two in PomBase: SPAC4H3.12c SPAC1F12.03c

SPBC1685.12c is not a current ID but there is a similar ID: SPBC1685.12c-antisense-1

SPBC8E4.02c is the ID of an RNA (SPNCRNA.9001) so I can't make a UTR. But that RNA overlaps a coding gene (SPBC8E4.01c) so maybe there was a mix up when the data set was created?

Some gene IDs (eg. SPBC13E7.09) have multiple lines in the data file but the value in the "Length of Longest UTR defined by cluster peaks" column is the same for each line so not a problem.

[kimrutherford]

Here's a first attempt: https://curation.pombase.org/kmr44/mata-polyA-sites-utrs-for-artemis.tar.gz

Now I have a script it's easy to re-create these files if things need tweaking. I added a /color= qualifier to make them stand out.

For the record, here's the script: utr_process_ticket_819.pl.txt

[ValWood]

We could even link to the tracks using the coordinates from the transcript section

I don't understand that bit.

yeah ignore that.

[ValWood]

I thought there should be a peak at 158577-5161 = 153416

There is a peak at 153416. Are you looking at the reverse strand? It isn't the strongest peak of all, but it is the highest peak in the cluster that would give the longest UTR.

[kimrutherford]

There is a peak at 153416. Are you looking at the reverse strand?

Yep, I was looking at the wrong strand. :-)

[ValWood]

https://www.pombase.org/status/new-and-removed-genes SPAC4H3.12c | not protein-coding (PMID:24929437, PMID:26615217) | 2015-12-11 SPAC1F12.03c | removed; replaced by a nuclear mitochondrial pseudogene (NUMT) feature | 2012-07-16 SPBC1685.12c | not protein-coding (PMID:24929437, PMID:26615217) | 2015-12-11

[ValWood]

Yep, I was looking at the wrong strand. :-)

I knew that - because I was too for the first few minutes ;-)

This is good. Absolutely spot on!

[kimrutherford]

SPBC8E4.02c is the ID of an RNA ...

That should be: SPBC8E4.02c is a synonym of an RNA ...

SPBC8E4.02c is the ID of an RNA (SPNCRNA.9001) so I can't make a UTR. But that RNA overlaps a coding gene (SPBC8E4.01c) so maybe there was a mix up when the data set was created?

Now I look more closely, I'm not sure that's true as there is a line for SPBC8E4.01c as well:

SPBC8E4.01c             230     230     425

It's different to SPBC8E4.02c

SPBC8E4.02c             647     284     617

[ValWood]

closing, new tasks will have new tickets.