Welcome to the MapMan Family of Software Forum

Please do not hesitate to register and post your question.

Don't forget to subscribe to your posted message so you get notified on updates.
Every question you post will help others and or enhance the software!

Post a question,   post a bug!

Welcome to the MapMen Family of Software Forum Welcome to the MapMen Family of Software Forum

Using MapMan

mappings and categories

mappings and categories
mapping file mappings mapping files arabidopsis bincode
Válasz
2016.08.29. 20:18
I downloaded the MapMan mappings and I noticed that they seem to go by Affymetrix ID. There is 1 Affymetrix ID for each mapping. However, a single Affymetrix ID can map to several AGI numbers. For example, 246764_s_at has three AGI numbers associated with it, at2g27780, at2g13640 and at5g27310.

I think that this is because the probes for 246764_s_at on the Affymetrix microarray chip have been made to probe for the gene that is the first of the three AGI numbers, at2g27780, but that due to common sequences in proteins the probes then also will pick up on RNA from the other two AGI numbers, at2g13640 and at5g27310.

The 246764_s_at gene is categoried in MapMan as being in bincode of 1.2.2 PS.photorespiration.glycolate oxydase, so this would characterize the function of 246764_s_at, which is probably equivalent to at2g27780, but what about at2g13640 and at5g27310 ? Are they also in 1.2.2 PS.photorespiration.glycolate oxydase, or should only the first AGI number be considered to be in this bincode ?


Matthew

RE: mappings and categories
Válasz
2016.08.30. 9:12 válaszként erre: Matthew McCormack.
Hi Matthew

which version of the mapping file is this?
Your interpretation of the Affymetrix codes are correct. Affymetrix marked them with _s_ and _x_ the former usually gene families the latter unrelated genes. However we reanalyzed the oligos as Affymetrix based their naming scheme on a so-called '"target sequence" within the gene on which they designed their oligos. So it can be that an _s_ is perfectly fine and (almost) all its oligos only hit really one gene or conversely that you have an unflagged probeset, but it hits multple genes, as the oligos selcted within the "target sequence" indeed are unspecific. (sometimes all oligos are overlapping).

In any case an example from the TAIR10 versioned file
In 1.2.2 we find e.g.
258359_s_at:
Description: MULTIPLE HITS: (at3g14415,at3g14420). at3g14415: (S)-2-hydroxy-acid oxidase, peroxisomal, putative / glycolate oxidase, putative / short chain alpha-hydroxy acid oxidase, putative | chr3:4818298-4821006 FORWARD at3g14420: (S)-2-hydroxy-acid oxidase, peroxisomal, putative / glycolate oxidase, putative / short chain alpha-hydroxy acid oxidase, putative | chr3:4821523-4824178 FORWARD

This has two gene hits. In this case it is easy, as both genes map to the same function.

In general though we tried to see if the function is "somewhat compatible" so if at least one non trivial function was overlapping between the two genes then all Bins were chosen (remember that MapMan tries to minimize redundancy so the rule should be only one function i.e. BIN).
assuming that maybe it could still be a gene family but getting the flag MULTIPLE HITS as above

For some of the older versions e.g. TAIR9 we moved the ones with no overlap into 35.3 disagreeing hits.

Now of course most people use NGS so at least this small problem goes away. (In which case you can get the AGI assigned mapping files).
Please also note, that we are currently re-annotating all genes and improving the BIN structure. You can get sneak peak (all of metabolism) here
http://www.plabipd.de/portal/mercator-ii-alpha-version-.
We will soon add transcription factors there. By the end of the year we should have all finished. (The updated pathway files will be labeled accordingly, none is available yet)
Given this development, if you would still need Affymetrix support for the novel annotations, please let us know.

Best Wishes
Björn

RE: mappings and categories
Válasz
2016.08.30. 18:49 válaszként erre: Björn Usadel.
Hi Björn,

I am using the August 2012 version. Thank you very much for your explanation and the links. I think I should download the AGI assigned mapping files.

Thank you for your work annotating Arabidopsis genes. We are finding MapMan annotation to be much more helpful than GO annotations. Our PI is always saying that she would be willing to help with annotation by supplying any information our lab has. ( If you are interested send me an email, mccormack@molbio.mgh.harvard.edu, and I will talk with Jen to see what we may have. )

MapMan has always been a very valuable resource for us. I look forward to using your new annotations.

Matthew