PCA plot colors

[sudeep71]

Hi

Great browser!!! I am having trouble when i try to select a category on my meta file such as treatment or condition or time from the browser to re-draw the PCA. I get the following error

Error in : Must request at least one colour from a hue palette

This does not happen with demo data?

few lines from my metafile are attached:

example_meta.xlsx

[nephantes]

You might need to define like below. R doesn't like "/"

sample	day	time	treatment	date	group_time	group_treat	batch
AB-1	AB_D1	2.5	R	10_23_18	A	A_R	1
AB-2	AB_D1	2.5	I	10_23_18	A	A_I	1
AB-3	AB_D1	3	R	10_23_18	B	B_R	1
AB-4	AB_D1	3	I	10_23_18	B	B_I	1
AB-5	AB_D1	3.5	R	10_23_18	C	C_R	1
AB-6	AB_D1	3.5	I	10_23_18	C	C_I	1

[sudeep71]

@nephantes ,

I tried using both underscore and even removing the complete date column, but i still cannot get it to work. The initial plot shows all my samples in spectrum of colors, but i go down to the PCA options/Colorfield and choose on any of those column headers as a option to color and it throws out
the same error!
Error in : Must request at least one colour from a hue palette

I have got the heatmap etc to work. Also where is the log file located? I could send it to you too, if it helps

[nephantes]

Could you send me a sample dataset that I can recreate the problem?

[sudeep71]

[count_data.xlsx](https://github.com/UMMS-Biocore/debrowser/files/2787747/count_data.xlsx) [AB_meta.xlsx](https://github.com/UMMS-Biocore/debrowser/files/2787748/AB_meta.xlsx) Hi Alper, Sorry for the late reply. Attached are both the count file and meta_data file. Count file is reduced to just 100 rows but has all the samples. I have also attached a screenshot of the error! Stock picture [image: image.png] When i change to highlight one of the metadata column through QC options/PCA options/color field i get the error below, attached [image: image.png] [image: image.png]

…

On Sat, Jan 19, 2019 at 1:55 AM Alper Kucukural ***@***.***> wrote: Could you send me a sample dataset that I can recreate the problem. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#200 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AGcbqK0-iKanspkmGMKkkcla7ZJ9eF1Wks5vEsFpgaJpZM4aGXlY> .

[nephantes]

Your sample names in the count_data and meta data file doesn't match. In count_data you have A_1, A_2... with underscore. In meta data file A-1, A-2,... with dash. If you use these two files it would work.
AB_meta.txt
count_data.txt

[sudeep71]

Thanks for your time. Simple error but frustrating!

…

On Wed, Jan 23, 2019 at 11:49 AM Alper Kucukural ***@***.***> wrote: Your sample names in the count_data and meta data file doesn't match. In count_data you have A_1, A_2... with underscore. In meta data file A-1, A-2,... with dash. If you use these two files it would work. AB_meta.txt <https://github.com/UMMS-Biocore/debrowser/files/2788290/AB_meta.txt> count_data.txt <https://github.com/UMMS-Biocore/debrowser/files/2788291/count_data.txt> — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#200 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AGcbqI4f42K_6vsa9X6fgmPgmo7X9Aooks5vGJKvgaJpZM4aGXlY> .

[nephantes]

We will work on better error reporting while uploading the data. We are reading the data with a straight forward way by making the first column to row names and matching the column names in count data file with the sample names in the meta data file. First, we will report if there are duplicates in the first column and second, we will report if sample names do not match. Hopefully, this might eliminate some frustration in the next version. Please don't hesitate to contact if you have any suggestions to make this tool better.

[nephantes]

I have just finalized error reporting in data upload. If there are duplicates in the first column of data file or un-matching column names in meta-data and count files or any missing columns in any rows will be reported. Hopefully, it will reduce the frustration in data upload. I pushed a new version 1.10.7 to bioconductor. It will be available in a couple of days. I also push it to github, if you don't want to wait for bioconductor.