The pink bars are the probes. These are getting read from the link.psl file. Note that IGB translates probe-to-target coordinates into genomic coordinates using the genometry data models.

Note that pink bars (probes) for minus strand probe sets appear on the bottom of the blue boxes.

Also, see this Web page for description of link.psl format:

http://www.affymetrix.com/support/technical/manual/alignments_psl_manual.affx

I found a copy of a link.psl file for ATH1 array here:

http://www.affymetrix.com/analysis/downloads/na21/ivt/ATH1-121501.ath1.link.psl.zip

The linked psl file shows up properly in 6.6. None of the files in the 'affy' folders which are labeled as 'probes' (see Homo sapiens or Rattus norveticus) show the pink bars though.

Do you mean under the data source NetAffx?

To test the probe sets visualization for mouse or human, you need to access the NetAffx DAS source, which does not appear to be available at the moment.

Opening and viewing the link.psl file worked fine for me.

Is there any reason to expect that it would not work if coming from a DAS2 server?

I ask because the NetAffx DAS source is currently down - at least for me.

Fixed. Works with psl from DAS2 too.

The tracks are showing in the main image correctly, with the pink probes overlaying the gene models BUT... there are now 2 tracks in the track table, and they behave oddly...

the track that shows up as 'empty' (ie no load icon, no load mode, no data set, no delete) is the track that controls the image that is seen (ie color changes in this track are seen; yellow in the attached picture). The track in which all of the values are present is the track that is empty (dark blue in attached picture).

Another issue is that if you 'remove data from selected tracks' on the probe sets, they never re-load, even if you move to a new area of the same chromosome.

Multi track bugs with these files

Multi track issue

Done in r9540.

Please test it with other probeset files too.

There is still a bug/some 'weirdness' with these files. When you check mark them to get them into the data management table, the show up with 2 track checked. If I uncheck 2 track before loading, they still load up as 2 track, with the two track box suddenly rechecked. Then I uncheck and it makes 1 track. However, the pink probes are NOT properly offset to the top/bottom indicating 'side'. They completely overlap the cyan glyph. I have to activate 2 track again and then uncheck it to get them to show the offset.

If the track is permitted to load as two track and then made 1 track, it loads and displays fine.

However, both human and plant are properly loading a 'single' track for each probe file.

Did this happened again?

Passed V.13601, Mac, Java 1.06.0_37

Tested with A. thaliana and H. sapiens. Probes appear in pink, offset from the read but in the same track, as depicted in last two photos attached to this bug.