I am trying to do some testing using the LOD output matrix from the orderMarkers module, but I am struggling to understand what the positions in the columns represent.
For a single linkage group I mapped 793 markers across 3 families. The resulting maps have:
- female map: 101 unique positions
- male map: 61 unique positions
- sex-averaged map*: 157 unique positions.
The LOD file has the correct number of rows (793), one row per mapped marker, but it only has 148 columns, thus 147 unique positions when removing the 'position' column.
Could you please explain what the columns in the LOD output file represent?
*Note that I obtained the sex-averaged position myself by doing (male+female)/2.
Kind regards,
Lorenzo
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I did not find any obvious problem with the code for calculating unique map positions (this routine is only used in the LOD calculation). I even tried in on real data, I got 147 unique positions from the LOD and from the map.
Are you using the latest version of LM3? Some older versions used identicalLimit parameter in finding these unique positions.
The LOD output contains the marker names, did you check the positions for these markers from the map?
Cheers,
Pasi
Last edit: Pasi Rastas 2024-05-17
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
We are using the version from December 2023. Should we try a newer version?
The names in the LOD output match the names and position of markers in the map file.
We first ran orderMarkers for each LG 10 times and selected the order with the best likelihood, and then re-ran orderMarkers with the best order to produce the LOD file with the below command.
We are using the version from December 2023. Should we try a newer version?
The names in the LOD output match the names and position of markers in the map file.
We first ran orderMarkers for each LG 10 times and selected the order with the best likelihood, and then re-ran orderMarkers with the best order to produce the LOD file with the below command.
The problem you are having is probably due to not using the same phasing for LOD computation and map evaluation. Safer option is to calculate LOD scores for each 10 runs and pick the LOD result based on the likelihoods (instead of evaluateOrder). You could also use the map from evaluateOrder output. The parameter phasingIterations=3 might improve the evaluated order's phasing...
Cheers,
Pasi
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello Pasi,
I am trying to do some testing using the LOD output matrix from the orderMarkers module, but I am struggling to understand what the positions in the columns represent.
For a single linkage group I mapped 793 markers across 3 families. The resulting maps have:
- female map: 101 unique positions
- male map: 61 unique positions
- sex-averaged map*: 157 unique positions.
The LOD file has the correct number of rows (793), one row per mapped marker, but it only has 148 columns, thus 147 unique positions when removing the 'position' column.
Could you please explain what the columns in the LOD output file represent?
*Note that I obtained the sex-averaged position myself by doing (male+female)/2.
Kind regards,
Lorenzo
Dear Lorenzo,
Thank you for your question.
I did not find any obvious problem with the code for calculating unique map positions (this routine is only used in the LOD calculation). I even tried in on real data, I got 147 unique positions from the LOD and from the map.
Are you using the latest version of LM3? Some older versions used identicalLimit parameter in finding these unique positions.
The LOD output contains the marker names, did you check the positions for these markers from the map?
Cheers,
Pasi
Last edit: Pasi Rastas 2024-05-17
Hello Pasi!
Thank you for the follow up.
We are using the version from December 2023. Should we try a newer version?
The names in the LOD output match the names and position of markers in the map file.
We first ran orderMarkers for each LG 10 times and selected the order with the best likelihood, and then re-ran orderMarkers with the best order to produce the LOD file with the below command.
Hello Pasi!
Thank you for the follow up.
We are using the version from December 2023. Should we try a newer version?
The names in the LOD output match the names and position of markers in the map file.
We first ran orderMarkers for each LG 10 times and selected the order with the best likelihood, and then re-ran orderMarkers with the best order to produce the LOD file with the below command.
Dear Lorenzo,
I think your version of LM3 is recent enough.
The problem you are having is probably due to not using the same phasing for LOD computation and map evaluation. Safer option is to calculate LOD scores for each 10 runs and pick the LOD result based on the likelihoods (instead of evaluateOrder). You could also use the map from evaluateOrder output. The parameter phasingIterations=3 might improve the evaluated order's phasing...
Cheers,
Pasi