• JoinMap
• MapQTL
• General information
• Evaluate
• Release notes
• Maintenance updates
• User manual
• FAQ List
• MQConvert
• Citations
• Shop

MapQTL reads its license file 'MapQTL.lic' in its program file directory (typically: 'C:\Program Files\MapQTL7' for version 7 or 'C:\Program Files (x86)\MapQTL6' for version 6). This is a protected directory and to place a file in it you need to have 'Administrator' privileges as Windows user. Copy your individual license file to this directory and rename it to 'MapQTL.lic'. With version 6 you need first remove the current license file 'MapQTL.lic', as it will be the evaluation license file. The easiest way to install a license is to start MapQTL from within the installation procedure (by placing the appropriate checkmark in the final screen of the installation), and next use the 'Install license' option of the MapQTL Help-menu.

When you get this message, it means that MapQTL does not have an individual license file installed. This situation only allows evaluation of the software, with limited functionality. If you obtained an individual license file (on your installation CDROM or by e-mail), then you must install this file according to the answer to question 1. The individual license file usually has a file name with the extension '.MQ7win' (or '.MQ6win').

MapQTL 7 runs on 64-bit Windows 10. MapQTL 6 runs as 32-bit software on 32-bit and 64-bit Windows 10. Both programs are expected to run properly on Windows 11, although this was not tested extensively.

MapQTL stores various view settings of its main window (position and size on the screen, partitioning of panels) in the file 'Documents\MapQTL7\MapQTL.Settings.sqlite' or 'in the 'Documents\MapQTL6\MapQTL.ProgSetup'. You should close the program, remove this file and start the program again; it will reset all view settings of the program. For version 7 this will also remove your default Analysis and Environment options.

Under certain circumstances you can get irregular lod profiles. They can be characterised by:

1. very irregular lod profiles, i.e. a (very) high lod at one marker next to a low lod at a neighbouring marker at short distance,
2. high lod scores often inbetween neighbouring markers that have a large distance to each other, while low lods at these markers,
3. high numbers of iterations at the high lod scores (>20),
4. non-normally distributed trait data (e.g. ordinal, binomial or uniform data),
5. unexpected QTL genotype contrasts (e.g. at an <nnxnp> type marker the major contrast is between mu_ac and mu_ad on the one hand (i.e. approximately mu_ac=mu_ad) and mu_bc and mu_bd on the other,
6. lower amounts of information in the markers, e.g. dominant genotypes or <nnxnp> markers and alike.

Basically, the results are a problem of the mapping algorithm, which is a maximum likelihood method for the separation of a mixture of distributions. In this algorithm the separation is based upon both genetic information, i.e. the marker genotypes, and phenotypic information, i.e. the quantitative trait data. Now, when there is not much genetic information and the trait is rather non-normal, then it may happen that the phenotypic information will outweigh the genetic information. For instance, when at an <nnxnp> type marker more or less two underlying distributions could be distinguished (especially at ordinal data on a scale of 1 to 5 or so, and also at uniformly distributed data), then these distributions become fitted to the ac and bc qtl-genotypes as well as to the ad and bd qtl-genotypes (i.e. a genetic effect within the first parent in the cross P1 x P2), while there is no (direct) genetic information from the marker about segregation within the first parent. As such the fitting happens by chance to the genotypes. Therefore, a major symptom of phenotypic information outweighing the genotypic information is the large number of iterations.

Serious errors in the linkage map may also be a cause of the problem.

Two other possible causes of irregular lod profiles are outliers in the quantitative trait data and erroneous marker data. Sometimes it can be better to replace suspect (marker or trait) data by unknowns.

To a certain extent there are some possibilities to inspect such odd results. An interval mapping analysis without map (enter 'none' as map file) or a Kruskal-Wallis analysis will show no significance on these markers. Another option (version 6 only) is to set the 'maximum number of neighbouring markers used' parameter to a higher number; this may increase the amount of genetic information; on the other hand this may also increase the odd results or introduce them in other areas.

QTL mapping with the regression approach is less affected than the maximum likelihood interval mapping approach.

This means that the expected trait values for the QTL genotypes cannot be estimated. It occurs when the probability for one (or more) of the QTL genotypes, given the marker genotypes and the position on the map for which the LOD is being calculated, is zero. Such a probability usually is never zero inbetween two markers (because there is always a non-zero probability of recombination), but it can be zero on top of a marker. This can only occur if all genotypes are known for the marker. In practice this means that in such a case the GIC value is 1.0; it can be lower but then the individuals with some missing marker information also have missing trait observations.

This can only occur if the number of distinct trait observations is equal to or smaller than the number of possible QTL genotypes. Usually this happens when the analysed trait was observed on an ordinal (or even binomial) scale. In such instances it may happen that the interval mapping maximum likelihood (ML) algorithm iterates towards a situation that the different QTL genotypes coincide exactly with the different trait classes. Usually this doesn't occur on top of markers, in such a case even the marker (!) genotypes would coincide exactly with the trait observations. But it occurs most often inside intervals (i.e. in between markers) where the ML algorithm will iterate towards this situation, which is signalled as a 'perfect fit'. [NB: One of the main assumptions for interval mapping is that the trait residuals behave according to a normal distribution; this assumption is violated when interval mapping is applied on ordinal traits.]