Visualizing non-business data

[Former Member]

I am a novice at visualization.  I am trying to create a visualization with non-business data that shows start and stops in a linear fashion.

I am using genome data that contains start/stop coordinates.  Each patient has multiple start/stops per chromosome.  I would like to show for each patient all the start and stops along a straight line/column/bar.

Example data:

Patient = 12

Chromo = chrX

Start = 10     Stop = 15

Start = 24     Stop = 32

Start = 45     Stop = 47

And so on...

Anybody got an ideas?

Thanks so much

Cindy

[varunanand · Accepted solution]

Hi Cindy,

Here are a few suggestions, where we are just looking at Chr x.

Start = Sales, Stop = Profit, Product Category = Patient, Year = Each count of start and stop.

The following would be for a bar graph,

You can replicate the same for a line chart,

The idea is to,

• use an analogy to replicate your scenario.
• show Start and Stop across various instances.
• leverage Trellis to show across multiple patients or chromosome.

Please let me know if there are any questions or comments.

Thank you,

Varun Anand

[Former Member · Accepted solution]

Thank you so much.  Unfortunately that isn't what I was looking for.  I've uploaded a drawing of what I was hoping to represent.  Thank you for your time.

[varunanand · Accepted solution]

Hi Cindy,

The closest that I can think is to use a Stacked Bar Chart. But you would need to model the data a bit in the prepare tab and change the aggregation of the measures to get the pattern you mentioned.

Thank you,

Varun Anand

[varunanand · Accepted solution]

Hi Cindy,

I also found this article on using a Timeline Chart .

But as Jay and I mentioned you would need to prepare your data accordingly.

Thank you,

Varun Anand

[Former Member]

Another possibility is to blend together two data sets - one dataset would have your start/stop data and the second dataset would basically be a trellis against which your first dataset is mapped.  the trellis set would look something like:

start   end

1     1

2     2

3     3

4     4

5     5

....

you would use that as an outer join against the dataset that has the actual start stop, joining on the start and end.  this way you would get the blanks where there are no values as well as the sections that have a value.  Mind you I am not sure that this will 100% work but I've seen this technique applied in other tools that provide data blending capability so it should be possible in Lumira.