I'm trying to use 4928 high quality SNPs (that were used to estimate linkage with Plink) to make a linkage map for my species. To give you a little bit more background, my species has 25 chromosomes and undergoes achiasmatic meiosis (no recombination in female) and I have genotype information from a mom and dad and their sixteen offsprings.
So this is what I did.
I took the ped file from plink, converted it to linkage format and then used Filtering, Separate Chromosomes, JoinSingles and OrderMarkers modules to get the linkage map.
However, after running these steps in the final output, all markers came out as duplicates. For your reference, please find the first few lines of the input linkage file (only a few columns for six individuals) and output file (order1.txt), and the standard output for the modules below.
It would really appreciate it if you could give me some pointers about what I'm doing wrong. Let me know if you need more details from my side.
1) Input file
family 1 0 0 1 0 2 1 2 1
family 2 0 0 2 0 1 1 1 1
family 3 1 2 0 0 1 1 2 2
family 4 1 2 0 0 2 2 1 1
family 5 1 2 0 0 2 2 1 1
family 6 1 2 0 0 1 1 2 2
java -cp /data/programs/lepmap2/bin SeparateChromosomes data=FaHetMoHom.linkage sizeLimit=56 lodLimit=4.3 informativeMask=1 >map1.txt
***
LOD(16, 0) = 4.350864189212862
LOD(15, 0) = 4.0594197333830895
LOD(15, 1) = 2.699498039411941
LOD(14, 2) = 1.20801156790841
LOD(16, 0) = 4.350864189212862
LOD(15, 0) = 4.0594197333830895
LOD(15, 1) = 2.699498039411941
LOD(14, 2) = 1.20801156790841
Family family
Number of different markers = 352
4.3
number of LGs = 150 singles = 4628
Removing LG 26 (as too small) with size 54
Removing LG 27 (as too small) with size 54
Removing LG 28 (as too small) with size 53
Removing LG 29 (as too small) with size 52
Removing LG 30 (as too small) with size 51
.
.
.
Number of LGs = 25, markers in LGs = 2371, singles = 2557
My guess is that all marker are identical, that is why you get all markers collapsing into one. Does this happen on all linkage groups or only on (chromosome=) 2? By adding parameter outputPhasedData=1 LM will ouput the most likely segregation patterns. These can provide more information.
With only 16 individuals, the linkage mapping does not necessary work that simply from the paternal markers. Based on the output of SeparateChromosomes and JoinSingles, you are throwing away too many linkage groups and markers with sizeLimit=56. (sizeLimit could be omitted or be set to about 3-4, this will yield much more groups).
I would try to get the chromosomes from the maternal markers. As there is no recombination, you should find 16 groups easily. These can be used, e.g. to verify your paternal linkage groups. If you have a mapping to a genome, it is quite easy to link paternal and maternal linkage groups (by eyeballing or maybe even using ScaffoldHMM)
Anyway, you probably need some manual work to finish your maps.
Cheers,
Pasi
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi Pasi and Lep-map users,
I'm trying to use 4928 high quality SNPs (that were used to estimate linkage with Plink) to make a linkage map for my species. To give you a little bit more background, my species has 25 chromosomes and undergoes achiasmatic meiosis (no recombination in female) and I have genotype information from a mom and dad and their sixteen offsprings.
So this is what I did.
I took the ped file from plink, converted it to linkage format and then used Filtering, Separate Chromosomes, JoinSingles and OrderMarkers modules to get the linkage map.
1) java -cp /data/programs/lepmap2/bin/ Filtering data=FaHetMoHom.linkage dataTolerance=0.001 >FaHetMoHom.f.linkage
2) java -cp /data/programs/lepmap2/bin SeparateChromosomes data=FaHetMoHom.linkage sizeLimit=56 lodLimit=4.3 informativeMask=1 >map1.txt
3) java -cp /data/programs/lepmap2/bin JoinSingles map1.txt data=FaHetMoHom.linkage lodLimit=3 lodDifference=1 >map1_js.txt
4) java -cp /data/programs/lepmap2/bin/ OrderMarkers map=map1_js.txt data=FaHetMoHom.f.linkage chromosome=2 initRecombination=0.05 0 >order1.txt
However, after running these steps in the final output, all markers came out as duplicates. For your reference, please find the first few lines of the input linkage file (only a few columns for six individuals) and output file (order1.txt), and the standard output for the modules below.
It would really appreciate it if you could give me some pointers about what I'm doing wrong. Let me know if you need more details from my side.
1) Input file
2) Output file
3) Filtering
4) SeparateChromosomes
5) JoinSingles
6) OrderMarkers
Regards.
R
Dear R Sheldon,
Thank you for your detailed question.
My guess is that all marker are identical, that is why you get all markers collapsing into one. Does this happen on all linkage groups or only on (chromosome=) 2? By adding parameter outputPhasedData=1 LM will ouput the most likely segregation patterns. These can provide more information.
With only 16 individuals, the linkage mapping does not necessary work that simply from the paternal markers. Based on the output of SeparateChromosomes and JoinSingles, you are throwing away too many linkage groups and markers with sizeLimit=56. (sizeLimit could be omitted or be set to about 3-4, this will yield much more groups).
I would try to get the chromosomes from the maternal markers. As there is no recombination, you should find 16 groups easily. These can be used, e.g. to verify your paternal linkage groups. If you have a mapping to a genome, it is quite easy to link paternal and maternal linkage groups (by eyeballing or maybe even using ScaffoldHMM)
Anyway, you probably need some manual work to finish your maps.
Cheers,
Pasi