how to set BAR_GRAPH scale value?

Jalview Discussion
1

Hi everybody,

I’d kindly like to ask you: how could I set a vertical axis scale of the BAR_GRAPH in the JALVIEW_ANNOTATION?

I mean, I imported the output values from a sequence analysis using the defined BAR_GRAPH “number,text character|” format, as indicated here , but once plotted, the highest value of the prediction will be represented as the highest possible value, e.g. if the highest output value of my prediction is 3 in a 0 to 10 scale it will be plotted as if it was 10.
I know that I can use labels below each bar but a visual view of a histogram with a correct scale will be much more clear especially when comparing annotations of different sequences.


Thank you for your attention,

regards,

fabio |

···
  • |
2

Thanks for the interesting email, Fabio!

3

hi Jim,
ty for ur kind reply…i tried to set a threshold, btw if all the values r below of the threshold value this one will no be represented. ya, it sounds weird to have histogram of values that are below a threshold but when comparing different sequences even “negative” predictions can be useful…

well…to overcome the problem in a “very dirty” way i set the maximum value for a residue (the first or the last one), colored it in white and without any label/tooltip number to mask it.

cheers,
fabio

— Ven 6/5/11, Jim Procter foreveremain@gmail.com ha scritto:

···

Da: Jim Procter foreveremain@gmail.com
Oggetto: Re: [Jalview-discuss] how to set BAR_GRAPH scale value?
A: “Fabio di Mirabella” fiore1606@yahoo.it
Cc: jalview-discuss@jalview.org
Data: Venerdì 6 maggio 2011, 14:06

Thanks for the interesting email, Fabio!

On 06/05/2011 11:44, Fabio di Mirabella wrote:

Hi everybody,

I’d kindly like to ask you: how could I set a vertical axis scale of the BAR_GRAPH in the JALVIEW_ANNOTATION?

Unfortunately, you can’t - and as you say, this is actually a serious oversight, since the user can be easily mislead. I’ve just lodged a feature request on this (http://issues.jalview.org/browse/JAL-829) - and hopefully we’ll be able to fit it in to a release in the not-to-distant-future (not the 2.7 due imminently, but perhaps 2.7.1 or 2.7.2). Taking a close look at the annotation file and the annotation visualization model is high on my TODO list for this summer - and this particular feature is actually very easy to implement.

For the moment, I can think of a couple of workarounds that people have used in the past, which includes providing a threshold line which represents the upper or lower extreme of the data - this might work for your situation.

Jim.