Halftyped Issue --Test runPM on CHP markers

[changebio]

Pedigree 1007 Person 1007_40 halftyped (1/0)

The halftyped came from phasing haplotypes. In some case, only one Haplotype is phased.
paramlink2 can handle the halftyped issue. However, pseudomarker will report an error. One solution is to set all halftypeds to missing genotypes (0/0). @gaow

Reading input files...
Reading map file PM.map
(Map file is in PLINK format)
 LOC100379224 ...
(Number of markers : 1)
...done.
(Pedigree file in pre-makeped format)
(Pedigree has disease phenotype)
Reading pedigree file: PM.ped
...failed
Pedigree 1007 Person 1007_40 halftyped (1/0) at marker_index: 0, line 6
 *** Problem with pedigree file format? ***

...aborting program due to errors!

progress is written, eg. /home/joe/
Enter folder containing programs, eg. /home/joe/bin/
Enter "1" (the number 1) if detailed information should be saved for each variant
Running pseudomarker...
**Marker 1
Error running pseudomarker. Do next marker.

[gaow]

paramlink2 can handle the halftyped issue

How does it handle this?

One solution is to set all halftypeds to missing genotypes (0/0). @gaow

Looks like the easiest thing to do, but could you double check with Jurg?

[changebio]

Okay. I will double check with Jurg.

I don't know how exactly paramlink2 does. Based on my understanding, we can calculate the likelihood of a variant only if there are no mendelian errors.
I think pseudomarker must set rules to filter markers with halftyped in some reasons.

[gaow]

I don't know how exactly paramlink2 does.

Can you investigate?