I was wondering why some of the maternally or paternally informative markers are not incorporated into the linkage groups. I have around 8000 maternally or paternally informative markers, but only ~4500 are being incorporated into linkage groups.
Thanks for your help.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I assume that we are talking about the output of SeparateChromosomes.
Does changing lodLimit change the number of used markers? Maybe the markers not used cannot reach your lodLimit. Have you run JoinSingles after SeparateChromosomes?
Cheers,
Pasi
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Sorry, yes we are talking about the output of SeparateChromosomes, and definitely the lodLimit influences the number of markers.
Is the LOD limit the only factor which will determine if markers are incorporated into linkage groups? Are there any other filters in the SeperateChromosome function? How are the number of maternally and paternally informative markers determined? Thank you so much for your time.
Regards,
Kris
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
If you are running Filtering module first, that will remove some markers from the data and those cannot be incorporated to the maps anymore.
From SeparateChromosomes, InformativeMask, families, sizeLimit and (fe)malePrior/fe(male)Theta all have an effect. A marker is informative maternally (paternally) if the mother (father) has heterozygous genotype at that marker.
Cheers,
Pasi
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Very nice explanation, thank you. I was wondering if there was documentation on InformativeMask, and the other criteria for filtering, or if you could elaborate a little on each of them. You have been very helpful, and I really appreciate your time and help.
Best Regards,
Kris
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The informativeMask simply takes markers that are either only paternally (1) only maternally (2) or paternally and maternally (3) informative. Default 0123 will take all markers.
This option has been most useful with datasets with relatively small number of offspring and low (or no) recombination in one sex. Using only markers with less recombination will separate chromosomes more robustly.
Also in some cases the marker ordering might work more robust (and is faster) by taking only single parent informative markers (1 or 2).
For other filtering, I guess the dataTolerance is the most important parameter in the Filtering module. It will remove markers that are not segregating very evenly (segregation distortion). Some of this distortion can be biological, but more often it is indicating problems with data. Especially sequencing based datasets have some very strange markers (that are clearly wrong), that are easy to remove by segregation filtering. Having these distorted markers when for example separating chromosomes, you might join different chromosomes into one linkage group. You can try adding some of the filtered markers to linkage groups using JoinSingles with the same data with less filtering.
Usually I do much filtering before generating the genotype data. It is up to the user whether to do it that way or with Lep-MAP.
Cheers,
Pasi
Last edit: Pasi Rastas 2015-12-02
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
I was wondering why some of the maternally or paternally informative markers are not incorporated into the linkage groups. I have around 8000 maternally or paternally informative markers, but only ~4500 are being incorporated into linkage groups.
Thanks for your help.
Hello Kris,
Thank you for question.
I assume that we are talking about the output of SeparateChromosomes.
Does changing lodLimit change the number of used markers? Maybe the markers not used cannot reach your lodLimit. Have you run JoinSingles after SeparateChromosomes?
Cheers,
Pasi
Hello Pasi,
Sorry, yes we are talking about the output of SeparateChromosomes, and definitely the lodLimit influences the number of markers.
Is the LOD limit the only factor which will determine if markers are incorporated into linkage groups? Are there any other filters in the SeperateChromosome function? How are the number of maternally and paternally informative markers determined? Thank you so much for your time.
Regards,
Kris
Dear Kris,
If you are running Filtering module first, that will remove some markers from the data and those cannot be incorporated to the maps anymore.
From SeparateChromosomes, InformativeMask, families, sizeLimit and (fe)malePrior/fe(male)Theta all have an effect. A marker is informative maternally (paternally) if the mother (father) has heterozygous genotype at that marker.
Cheers,
Pasi
Hello Pasi,
Very nice explanation, thank you. I was wondering if there was documentation on InformativeMask, and the other criteria for filtering, or if you could elaborate a little on each of them. You have been very helpful, and I really appreciate your time and help.
Best Regards,
Kris
Hello Kris,
The informativeMask simply takes markers that are either only paternally (1) only maternally (2) or paternally and maternally (3) informative. Default 0123 will take all markers.
This option has been most useful with datasets with relatively small number of offspring and low (or no) recombination in one sex. Using only markers with less recombination will separate chromosomes more robustly.
Also in some cases the marker ordering might work more robust (and is faster) by taking only single parent informative markers (1 or 2).
For other filtering, I guess the dataTolerance is the most important parameter in the Filtering module. It will remove markers that are not segregating very evenly (segregation distortion). Some of this distortion can be biological, but more often it is indicating problems with data. Especially sequencing based datasets have some very strange markers (that are clearly wrong), that are easy to remove by segregation filtering. Having these distorted markers when for example separating chromosomes, you might join different chromosomes into one linkage group. You can try adding some of the filtered markers to linkage groups using JoinSingles with the same data with less filtering.
Usually I do much filtering before generating the genotype data. It is up to the user whether to do it that way or with Lep-MAP.
Cheers,
Pasi
Last edit: Pasi Rastas 2015-12-02