Class AffyPowerToolsProbesetSummarizeTest

java.lang.Object
ubic.gemma.core.loader.expression.AffyPowerToolsProbesetSummarizeTest
public class AffyPowerToolsProbesetSummarizeTest extends Object
Author:
paul

  • Constructor Details

    • AffyPowerToolsProbesetSummarizeTest

      public AffyPowerToolsProbesetSummarizeTest()

  • Method Details

    • testLoadMPSNames

      @Test public void testLoadMPSNames()

    • testClariomSMouseHTIsConfiguredAtEveryStepOfTheCelChain

      @Test public void testClariomSMouseHTIsConfiguredAtEveryStepOfTheCelChain()
      Reprocessing a CEL walks THREE separate tables, and a platform missing from any one of them fails at a different depth with a different message. GPL24242 was missing from all three and they surfaced one at a time over an afternoon (2026-09-22), so walk the whole chain here rather than one link:
      1. affy.celmappings.properties — the chip-type string in the CEL header to the ORIGINAL platform. Absent: Couldn't figure out the GPL for Clariom_S_Mouse_HT, thrown by DataUpdaterImpl.determinePlatformsFromCELs before anything else is consulted.
      2. affy.altmappings.txt — the original platform to the one whose library we summarize with. Absent: is not recognized as an Affymetrix platform, refused by the CLI. Covered by GeoPlatformTest.
      3. affy.mps.properties — that platform to its four library files. Absent: There is no CDF or MPS configuration for, thrown when the apt command is built.
      Clariom_S_Mouse_HT resolves to itself throughout: its CEL files carry affymetrix-array-type=Clariom_S_Mouse_HT, which apt matches against #%chip_type in the pgf, and the non-HT pgf describes a different probe layout.

    • testEveryKnownChipTypeIsConfiguredThroughToItsLibrary

      @Test public void testEveryKnownChipTypeIsConfiguredThroughToItsLibrary()
      Every chip type we claim to recognize has to be configured at all three steps, or reprocessing dies part way in with a message that names only the step it reached. Nothing checked that, so GPL24242's three gaps surfaced one run at a time.

      The four exceptions below are pre-existing and are listed rather than fixed: what they should resolve to is a question about those arrays, not about this invariant. A NEW name in one file and not the others fails here instead of on frink.

    • testCELnameregex

      @Test public void testCELnameregex()