I noticed on the new version that phase information was now included in the output. I was wondering how to interpret this information. Below is an example:
The last columns in the output of OrderMarkers are the phases. There is a single phase for each family in the order they occur in the data (I will try to add column names in the next update of LM2).
The phases are 0- or 1- for paternally-only informative markers. They define how to map genotypes into haplotypes. If phase is 0-, then genotype with the first paternal allele is the paternal haplotype 0 and the other is 1. If phase is 1-, then the haplotypes are the opposite.
Similarly for maternally-only informative markers the phases are -0 and -1.
For markers informative in both parents, phases are 00, 01, 10 and 11. If phase is 00 then the genotype with first parental allele corresponds to haplotype 0 and the second allele to 1, whereas the phase 11 is the opposite. In phase 01 first paternal and second maternal alleles correspond to haplotype 0, whereas 10 is the opposite.
Phase -- is for markers not informative at all.
For example, parents are "1 2" and "1 2":
offspring genotype "1 1", phase 00 => haplotypes "0 0"
offspring genotype "1 2" phase 01, haplotypes "0 0" or "1 1"
offspring genotype "2 2" phase 10, haplotypes "0 1".
Lep-MAP2 only need and outputs "parental" phase. If grandparental phase is needed (QTL mapping?), then these phases can be obtained (say for a subset of markers) from the grandparental genotypes and the phases of each family and chromosome need to reversed if the parental and grandparental phases are opposite.
Hope this helps, the phasing is quite complicated...
Cheers,
Pasi
Last edit: Pasi Rastas 2016-03-17
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I added option outputPhasedData to OrderMarkers module. This will output phased data for each (non-duplicate) marker and provides easy way to get phased data.
Cheers,
Pasi
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi Pasi,
Thanks for Lep-MAP2. It's really fantastic. It worked beautifully to put together a linkage map for a new organism we're working on. Next I'd like to use the map for QTL applications, and I think the phase output you've added should make this easier. But I'm not clear on how to interpret the output.
Below is a little bit of output (from a much larger dataset). Suppose that the parents of these individuals were haplotypes AB x CD. I take it that, for example, the first two columns of phase output, here all 11, represent the haplotypes inherited by the first individual, so individual one inherited haplotypes B and D. The next individual is all 00, and so inherited haplotypes A and C. Is that right?
Thanks for any guidance!
Matt
Your interpretation is correct. Maternal and paternal haplotypes are interleaved, so phased data for the first individual is in the two first digits, second individual in digits 3 and 4, and so on. The data is imputed, so there are no missing values at all.
Cheers,
Pasi
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear Pasi,
I noticed on the new version that phase information was now included in the output. I was wondering how to interpret this information. Below is an example:
mar male female err fam1_phase ....
220 110.433 0.829 ( 0.2 ) -0 -0 -- -0 -0 -1 -1 -0 -0
140 111.506 0.829 ( 0.2 ) 0- 0- 0- 0- -- -- -- 1- 1-
249 111.580 0.829 ( 0.2 ) 1- 00 -- 00 00 -1 -1 10 --
What do the various number and dashes mean? Thank you for your time and your excellent software.
Regards,
Kris Christensen
Thanks Kris for a good question,
The last columns in the output of OrderMarkers are the phases. There is a single phase for each family in the order they occur in the data (I will try to add column names in the next update of LM2).
The phases are 0- or 1- for paternally-only informative markers. They define how to map genotypes into haplotypes. If phase is 0-, then genotype with the first paternal allele is the paternal haplotype 0 and the other is 1. If phase is 1-, then the haplotypes are the opposite.
Similarly for maternally-only informative markers the phases are -0 and -1.
For markers informative in both parents, phases are 00, 01, 10 and 11. If phase is 00 then the genotype with first parental allele corresponds to haplotype 0 and the second allele to 1, whereas the phase 11 is the opposite. In phase 01 first paternal and second maternal alleles correspond to haplotype 0, whereas 10 is the opposite.
Phase -- is for markers not informative at all.
For example, parents are "1 2" and "1 2":
offspring genotype "1 1", phase 00 => haplotypes "0 0"
offspring genotype "1 2" phase 01, haplotypes "0 0" or "1 1"
offspring genotype "2 2" phase 10, haplotypes "0 1".
Lep-MAP2 only need and outputs "parental" phase. If grandparental phase is needed (QTL mapping?), then these phases can be obtained (say for a subset of markers) from the grandparental genotypes and the phases of each family and chromosome need to reversed if the parental and grandparental phases are opposite.
Hope this helps, the phasing is quite complicated...
Cheers,
Pasi
Last edit: Pasi Rastas 2016-03-17
Dear all,
I added option outputPhasedData to OrderMarkers module. This will output phased data for each (non-duplicate) marker and provides easy way to get phased data.
Cheers,
Pasi
Hi Pasi,
Thanks for Lep-MAP2. It's really fantastic. It worked beautifully to put together a linkage map for a new organism we're working on. Next I'd like to use the map for QTL applications, and I think the phase output you've added should make this easier. But I'm not clear on how to interpret the output.
Below is a little bit of output (from a much larger dataset). Suppose that the parents of these individuals were haplotypes AB x CD. I take it that, for example, the first two columns of phase output, here all 11, represent the haplotypes inherited by the first individual, so individual one inherited haplotypes B and D. The next individual is all 00, and so inherited haplotypes A and C. Is that right?
Thanks for any guidance!
Matt
Hello Matt,
Thanks for good question.
Your interpretation is correct. Maternal and paternal haplotypes are interleaved, so phased data for the first individual is in the two first digits, second individual in digits 3 and 4, and so on. The data is imputed, so there are no missing values at all.
Cheers,
Pasi
Terrific. Thanks!