search

bioinformatics-ptp / Zanardi

3 stars 3 forks source link

FImpute pedigree check #2

Open nalnahhas opened 7 years ago

nalnahhas commented 7 years ago

When option --fimpute was used, Zanardi indicated a problem in the pedigree file: "Male parent in pedigree file (2nd column) is not present as individual (1st column). Check male parent: Sire." A simple check in R with the command all(ped$sire[ped$sire >0] %in% ped$ID) indicate that all male (and female) parents are present in the ID column of the pedigree.

I run Zanardi with the option --pedigchk and surprisingly, the pedigree file passed the checking process. Could it be just a bug with --fimpute option?

P.S: the pedigree file format is: ID;Sire;Dam;Date_of_birth;Sex as indicated in Zanardi user manual. Thanks.

nicolazzie commented 7 years ago

Hi, It's difficult for me debug this without seeing the debug log. I'll give it a try: What the error is saying is not that there is a blank sire. It's saying that the pedigree contains a sire (in the second column) that is not present as individual (in the first column). This is a FIMPUTE thing. zanardi is only helping you debugging this. FIMPUTE requires ALL individuals, including sires and dams to be present as individuals in the pedigree. If you don't have pedigree for sires and dams, you can use "null" ids (UUUUUU... or 000000.. i don't remember) for them. Make sure that birthdate preceeds the birth date of their sons/daughters. You can see fimpute docs to see how the pedigree is supposed to be.

Hope it helps. Best, Ezequiel

Sent from my iPhone

On Oct 12, 2016, at 10:50, nalnahhas notifications@github.com wrote:

When option --fimpute was used, Zanardi indicated a problem in the pedigree file: "Male parent in pedigree file (2nd column) is not present as individual (1st column). Check male parent: Sire." A simple check in R with the command all(ped$sire[ped$sire >0] %in% ped$ID) indicate that all male (and female) parents are present in the ID column of the pedigree.

I run Zanardi with the option --pedigchk and surprisingly, the pedigree file passed the checking process. Could it be just a bug with --fimpute option?

P.S: the pedigree file format is: ID;Sire;Dam;Date_of_birth;Sex as indicated in Zanardi user manual. Thanks.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

nalnahhas commented 7 years ago

Hi Ezequiel,

I tried one more thing, removing the header line form the pedigree file, and the pedigree passed all ckecks without any problem.

Best regards,

Nabeel.