gene track no arrowheads

[Ho, Eric]

Hi,

I used a bed file to specify genes along the chromosome. GB is fine to display them but I didn't see the arrowheads in the bars i.e. >>> or <<< even the bed specified strand to be either + or -

Below is a partial list of the ncbiGene table:

+-----+-------+------------+----------+--------------+-------+--------+------------+----------+----------+------------+------------+-------------+
| bin | chrom | chromStart | chromEnd | name | score | strand | thickStart | thickEnd | reserved | blockCount | blockSizes | chromStarts |
+-----+-------+------------+----------+--------------+-------+--------+------------+----------+----------+------------+------------+-------------+
| 585 | MSV | 149 | 455 | YP_009154761 | 0 | + | 149 | 455 | 65280 | 1 | 307, | 149, |
| 585 | MSV | 465 | 1200 | YP_009154762 | 0 | + | 465 | 1200 | 65280 | 1 | 736, | 465, |
| 585 | MSV | 1352 | 2527 | YP_009154763 | 0 | - | 1352 | 2527 | 65280 | 1 | 1176, | 1352, |
| 585 | MSV | 1708 | 2527 | YP_009154764 | 0 | - | 1708 | 2527 | 65280 | 1 | 820, | 1708, |
+-----+-------+------------+----------+--------------+-------+--------+------------+----------+----------+------------+------------+-------------+

Any ideas?


Eric.

[Matthew Speir]

Hi Eric,

Thank you for your question about displaying strand information for your
BED track.

In the trackDb stanza for the ncbiGene track, did you properly specify
the track type? You need to specify the number of columns your BED table
has, otherwise, it defaults to the display of BED3 data. For example,
your ncbiGene track would need the following type line to properly display:

type bed 12

You may also want to look into the "genePred" table format,
https://genome.ucsc.edu/FAQ/FAQformat.html#format9, which includes
options specialized for displaying gene annotations in the Genome
Browser. Information on the track settings available for genePred tracks
can be found here:
https://genome.ucsc.edu/goldenPath/help/trackDb/trackDbDoc.html#genePred.

I hope this is helpful. If you have any further questions, please reply
to genome...@soe.ucsc.edu. All messages sent to that address are
archived on a publicly-accessible Google Groups forum. If your question
includes sensitive data, you may send it instead to genom...@soe.ucsc.edu.

Matthew Speir
UCSC Genome Bioinformatics Group

[Matthew Speir]

Hi Eric,

Could you provide us with the full trackDb stanza for your ncbiGene
track? This should help us debug the issue further.

A starting point for genePred would be the utilities:
- bedToGenePred
- gff3ToGenePred
- gtfToGenePred

All of which allow you to convert your existing formats into the
genePred format. GenePred tables can be loaded with the utility
"hgLoadGenePred". You can look around the hg38 documentation for some
examples of genePred track loading:
http://genome-source.cse.ucsc.edu/gitweb/?p=kent.git;a=tree;f=src/hg/makeDb/doc/hg38;.

I hope this is helpful. If you have any further questions, please reply
to genome...@soe.ucsc.edu. All messages sent to that address are
archived on a publicly-accessible Google Groups forum. If your question
includes sensitive data, you may send it instead to genom...@soe.ucsc.edu.

Matthew Speir
UCSC Genome Bioinformatics Group

On 7/20/16 9:44 AM, Ho, Eric wrote:
> Matt,
>
> I did specify the line below in trackDb.ra
>
> type bed 12
>
> I'll definitely look into genePred format. It would be helpful if you could point me to the procedure (or kent utils) to generate such format.
>
> But I saw this example from http://genome.ucsc.edu/FAQ/FAQformat.html under BED format
>
> http://genome.ucsc.edu/cgi-bin/hgTracks?org=human&position=chr7&hgt.customText=http://genome.ucsc.edu/goldenPath/help/ItemRGBDemo.txt
>
> It is based on the simple BED and still can display the arrowheads. So why can't my BED?
>
> Eric.

[Ho, Eric]

Matt,

I did specify the line below in trackDb.ra

type bed 12

I'll definitely look into genePred format. It would be helpful if you could point me to the procedure (or kent utils) to generate such format.

But I saw this example from http://genome.ucsc.edu/FAQ/FAQformat.html under BED format

http://genome.ucsc.edu/cgi-bin/hgTracks?org=human&position=chr7&hgt.customText=http://genome.ucsc.edu/goldenPath/help/ItemRGBDemo.txt

It is based on the simple BED and still can display the arrowheads. So why can't my BED?

Eric.
----- Original Message -----
From: "Matthew Speir" <msp...@soe.ucsc.edu>
To: "Ho, Eric" <h...@lafayette.edu>, "genome-mirror" <genome...@soe.ucsc.edu>
Sent: Wednesday, July 20, 2016 12:25:30 PM
Subject: Re: [genome-mirror] gene track no arrowheads

[Ho, Eric]

track ncbiGene
shortLabel NCBI Genes
longLabel NCBI Genes from GenBank Record
group genes
visibility pack
priority 1
color 12,12,120
type bed 12
itemRgb On
baseColorUseCds given
baseColorDefault genomicCodons

[Ho, Eric]

Hey Matt,

Thanks for your support thus far. I managed to get genePred to work almost perfectly; it's very close. But I have two questions:

1. about displaying the "translated protein from predicted mRNA". how can I turn it on? The webpage displayed only "Non-protein coding gene or gene fragment, no protein prediction available", there was no link.

2. I still don't have the arrowheads! What might be wrong?

[Ho, Eric]

I want to supplement more information regarding #1 in my previous reply. Actually, the missing of 'translated protein from predicted mRNA' occurred only in certain genes.

I'd translated the predicted mRNA of a gene that didn't have the translated protein feature being showed up, the +1 frame is prefect. So, I don't what have irritated GB.

[Matthew Speir]

Hi Eric,

To go back to the original issue with the BED tracks not correctly
display the strand indicator arrows, we believe that this is a bug and
will look into it. Unfortunately, I can't offer a timeframe for when
this will be fixed.

I'm not sure why the strand arrows don't show up for your genePred
track. Perhaps you have the "codon coloring" (or codon translation)
option turned on? If you have this option turned on and there are no
introns in your annotations, then no strand arrows will be displayed.

One of our engineers notes that while it's possible to get this sequence
information into genePred tables, that it is much easier to just use the
new "bigGenePred" file format. These "predicted mRNA" and "translated
protein from predicted mRNA" links should be present on the details
pages of bigGenePred tracks. The bigGenePred file format does work for
native tracks as well. You can find instructions on creating your own
bigGenePred files here:
https://genome.ucsc.edu/goldenPath/help/bigGenePred.html. To use these
for native tracks, all you need to do is add a "bigDataUrl" line to the
stanza that points to the file. The information on this bigDataUrl line
can just be a relative path to the file on your machine.

I hope this is helpful. If you have any further questions, please reply
to genome...@soe.ucsc.edu. All messages sent to that address are
archived on a publicly-accessible Google Groups forum. If your question
includes sensitive data, you may send it instead to genom...@soe.ucsc.edu.

Matthew Speir
UCSC Genome Bioinformatics Group

[Ho, Eric]

Matt,

I affirm you that this part is true: "Perhaps you have the "codon coloring" (or codon translation)

option turned on? If you have this option turned on and there are no
introns in your annotations, then no strand arrows will be displayed."

When I turned the codon coloring off, the arrowheads appeared.

Thanks, Eric.