Map Manager QT

Questions & Answers

Answers to some questions that users of Map Manager QT have asked. If these answers don't help, ask a question of your own .

Return to Map Manager QT home page

On-line Manual for Map Manager and Map Manager QT

Map Manager QT in general

• How do I cite Map Manager QT?
  • When you refer to Map Manager QT, please use the full name as written here, three words of mixed case with spaces. Although the program name is abbreviated on our Sun Web server, the Mac OS allows long file names with spaces, and there is no reason to use "MapManager", "MapMgr", or similar forms.
  • Two references for Map Manager QT are:
    • Manly KF and Olson JM (1999) Overview of QTL mapping software and introduction to Map Manager QT.
      Mamm Genome 10, 327-334.
    • K.F. Manly. (1998) User's Manual for Map Manager Classic and Map Manager QT. http://mapmgr.roswellpark.org/MMM/MMMBookTOC.html

QT Analysis in general...

• Why is the QT menu inactive?
  • The QT menu is inactive when the Linkage Analysis submenu (Option menu) is set for the analysis of radiation hybrids.

Interval Mapping...

• When I sort by trait value, the order of the progeny in the trait window does not change.
  • This is a problem of the display conventions of Map Manager QT. Progeny display orders are associated with chromosomes. Traits are not normally associated with a chromosome and therefore have no display orders. Our solution is the Align item in the QT menu. This aligns a trait window with a locus in an SDP window and causes the trait window to use the progeny order of the SDP window. To display trait values in sorted order, sort a chromosome by the trait value (use Apply Progeny Sort under the Chrom menu), then open an SDP window for that chromosome, make sure that it is using the trait order (use the popup menu at the right edge of the window), open the trait window and align it with a locus in the chromosome (use Align under the QT menu).
• What does NAN(x) mean?
  • NAN means "Not a number", and it is generated by the math routines if the program tries to divide by zero, take the log of a negative number, or perform some other mathematical atrocity. I am trying to identify and fix situations which produce NAN's. If you have a dataset which produces NAN's with the latest version of Map Manager QT, I would appreciate receiving that part of the file which demonstrates the problem.
• Why does interval mapping fail in the BXD recombinant inbred set?
  • See the first item under "Why does the Interval Mapping... process fail...", below.
• What does Map Manager QT do with missing data?
  • Missing trait values--As of Map Manager QTb9, progeny which have no trait value are completely excluded from mapping and permutation analysis. Progeny which have no trait value now have no effect on mapping or permutation tests. This should eliminate problems which were caused in previous versions by missing trait values which were included in the permutations of a permutation test.
  • Missing marker genotypes--During interval mapping or a permutation test, Map Manager QT excludes from the analysis progeny which are not typed for the trait or which are not typed for all of the marker loci included in the analysis. If there are missing marker genotypes, therefore, the number of informative progeny will differ for different loci or intervals. The QT Links report, the interval mapping report, and the permutation test report include a report of the minimum and maximum number of informative loci encountered during calculations.
• Why does the Interval Mapping... process fail?
  • During interval mapping or a permutation test, Map Manager QT excludes from the analysis any progeny which are not typed for the trait or which are not typed for any of the marker loci included in the analysis. This may create a situation in which the regression equations cannot be solved. For example, if the informative progeny at one locus are all of the same genotype, the regression will fail.
  • Check the Linkage Evaluation setting under the Option menu. You may have Linkage Evaluation set to "backcross" for an intercross or vice versa.
  • Interval mapping may fail if you have markers in the chromosome being analyzed which have identical or nearly identical phenotype distribution patterns with loci in the chromosome controlling for other QTLs. Map Manager QT will warn about identical loci, but it will not warn about nearly identical loci.
• Why does my backcross data give such strange results (or no results)?
  • It may be encoded incorrectly. In Map Manager, the convention for coding backcrosses has been to use maternal and paternal codes rather than heterozygote. This is described on page 65 of the manual. This convention is different from most other programs, including Mapmaker (Mapmaker 2.0 actually accepted both conventions). In this version I do not have an automatic way to change the coding, but you can do it by manually editing the data file. The data file is a text file; although it does not have a TEXT file type, it can be opened by BBEdit or other text editors. The seventh line of the file is a string of characters which give the abbreviations for phenotypes in the order unknown, maternal, paternal, heterozygote, not-maternal, not-paternal. Change the fourth (heterozygote) letter to a letter you don't use, and change the unused parental letter to the letter previously used for heterozygote. This problem has happened both with data imported from Mapmaker and with data entered by someone who did not know the Map Manager convention. I will be adding a function to correct this type of coding error.
  • As of version b4, the Mapmaker import function has been corrected so that it should produce the right phenotype coding.
• Why doesn't the "to chromosome" menu work in the Interval Mapping dialog?
  • This menu is not enabled in the current version of Map Manager QT. It will always show the same chromosome as the "from chromosome" menu.
• Why are most of the loci in the "controlling" chromosome menu dimmed?
  • In the current version of Map Manager QT, the maximum number of loci for controlling other QTLs is seven. This "control" chromosome should not be a normal chromosome. It should be a small collection of loci (perhaps from several chromosomes) which are strongly associated with the trait.
• Why do the map distances in the interval mapping figure disagree with the map distances in the Map window?
  • History. In the current version, distances in the Map window are recombination fractions, because that is what they have always been. Distances in the interval mapping figure are map distances derived from the Haldane map function (no interference), because that is what Haley & Knott used. Eventually, we will reconcile these, offering optional map functions for both.

QT Find and Links Report

• When will we be able to do a Find or QT Links Report with control for other QTLs?
  • QTb5 provides this feature for the Links Report (not yet for the Find command).

Return to Map Manager QT home page
Map Manager Classic/QT On-line Documentation