I use the Brassica Affy data from for transcriptomics data of B.juneca. However, when i run mapman the ids did not match. The ids I used were unigene ids. The part of the error message was the ids are repeated in the table. But, if i go for the unique records only then i will loose quite a lot of genes.

Suggestions....

Hi Prateek,

MapMan requires the input file to contain only one value (or row of values if multiple comparisons
are imported as one single table) per gene identifier. If your input file has several rows for one
unigene ID i would suggest that you summarize the values for each ID first (e.g. by calculating
the mean or median value) and then import the resulting (shorter) table. This way you will not
loose any data but get more robust expression estimates for each unigene.

I hope this helps,
bests,
Marc

Hello Marc,

Thanks for the post. I do not have multiple rows for single Unigene IDs. For your consideration I am attaching the data files so that u can understand what i wan to say.

Thanks again
Prateek

Hi Prateek,

i had a quick look at the attached file and found out that it actually does contain
multiple rows that have the same identifier: There are numerous rows that have
the ID "X". Since MapMan is not imposing any format requirement for the identifier,
it is not checking whether the IDs look like genuine unigene IDs. A single character
like "X" is acceptable. The only obligatory requirement for MapMan to work is that
the used IDs are unique in the input file. And this is not the case in yours.

I do not fully understand what the "X" rows in your file mean. Are they measurements
that cannot be attributed to one ID? Or are they replicate measurements of the last
preceding "Bna.NN" ID?

cheers,
Marc

Marc,
Those "X' in the IDs column just are meant to represent the Blanks. So that Mapman reads that the column is blank...(no uingene Ids for a particular probe).

Hope that helps.
Prateek

In case the "X"s are just blanks you can delete them from your input file. They are
not needed in MapMan - MapMan was designed to display log2-fold change data -
so the numerical data that you are importing should contain e.g. the log2-fold changes
in expression between to conditions for each measured transcript.

cheers,
Marc

Marc,
Values represented here are log2 fold induction between control and treated samples for respective transcripts.

Prateek

Hi Prateek,

well, in that case everything should be fine after deleting the unneeded blank X rows -
bests,
Marc