problems uploading csv

[avilella]

Hi,

I am trying to upload an example csv of raw input data. I took the one in the example, but complained it didn't have enough lines. So I modified it and added more lines like so:

Gene ID, name, control rep1, control rep2, treatment A rep1, treatment A rep2, EC Number
gene001, flavodoxin, 60, 40, 200, 220, 3.1.-.-
gene002, p53, 0, 4, 20, 30,
gene003, potassium uptake protein, 600, 633, 200, 220, 2.7.8.-
gene004, potassium uptake protein, 600, 633, 200, 220, 2.7.8.-
gene005, potassium uptake protein, 600, 633, 200, 220, 2.7.8.-
gene006, potassium uptake protein, 600, 633, 200, 220, 2.7.8.-
gene007, potassium uptake protein, 600, 633, 200, 220, 2.7.8.-
gene008, potassium uptake protein, 600, 633, 200, 220, 2.7.8.-

I then uploaded it to the system. I assign Gene ID and name as info columns, EC number column, select the treatment and controls, click 'Save' then click 'View'. See below.

The result is an empty page with no values/rows. See below.

Am I loading this data correctly?

[avilella]

I tried it with combination of only 'Gene ID' as label, or only 'name' as label, no luck.

I tried it with a larger file with about 25000 rows, same result.

[colindaven]

I'm not a developer here, but have used extensively.

To me I would avoid having a space in the column names try control rep1 as control_rep1

Colin

[avilella]

Tried that, same result. Does anybody have a working example?

On Jan 26, 2017 07:30, "Colin Davenport" notifications@github.com wrote:

I'm not a developer here, but have used extensively.

To me I would avoid having a space in the column names try control rep1 as control_rep1

Colin

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Victorian-Bioinformatics-Consortium/degust/issues/45#issuecomment-275324090, or mute the thread https://github.com/notifications/unsubscribe-auth/AAJpNxC-RE7nObcTH63RJEPbHR8AuOR9ks5rWEuhgaJpZM4LthgF .

[avilella]

I see there are examples on the website for the analysed values. In my case I want to load a csv with the raw counts, but I can't get it to work. Any ideas?

[colindaven]

Thats a shame. Your data all look ok to me. We installed locally (very hard) and used extensively.

Chief problems were: spaces in any field, column names being the same as condition names (or similar), header being ignored, etc.

You definitely need to put a CSV with the raw counts in.

[avilella]

Hi Colin,

Have you got a working example .csv that I can start with? If I have a working one, I can reverse engineer the issue from it.

A.

On Fri, Jan 27, 2017 at 2:58 PM, Colin Davenport notifications@github.com wrote:

Thats a shame. Your data all look ok to me. We installed locally (very hard) and used extensively.

Chief problems were: spaces in any field, column names being the same as condition names (or similar), header being ignored, etc.

You definitely need to put a CSV with the raw counts in.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Victorian-Bioinformatics-Consortium/degust/issues/45#issuecomment-275684466, or mute the thread https://github.com/notifications/unsubscribe-auth/AAJpNyTqqN-g9zDI274vfWyZ4Snw1WV_ks5rWgYYgaJpZM4LthgF .

[drpowell]

Turns out the problem is that the space you have in the column names after the comma. Specifically, if it should be fine if the header row was this

Gene ID,name,control rep1,control rep2,treatment A rep1,treatment A rep2,EC Number

I'll raise a bug, and try and fix this properly soon. Also, there is newer beta version of degust with a few new features running here : http://dna.med.monash.edu:4000/

[drpowell]

Bug raised for this : https://github.com/Victorian-Bioinformatics-Consortium/degust/issues/46