Error when loading file into PLINK

Question

[mosedaled]

Hi, I wonder if anyone can help.

I am new to SNP analysis, and am in the process of loading my data into PLINK for analysis, when I received the following message:

ERROR Problem with MAP file line: 8 rs71468368 120639217

So, I am left with not being able to load that data into PLINK.The MAP file line that it tells me is a problem doesn't exist in my MAP file. The SNP it says is the error doesn't exist anywhere in the file, but the location it says that SNP has does exist in my MAP file, but for a different SNP rs6469836.

I wondered if it was to do with the build 37 data that I was using, so I converted everything to build 38 co-ordinates and got the following error:

ERROR Problem with MAP file line: 8 rs71468368 119626977

The same error, but with the co-ordinates updated. Again, this SNP (still!) doesn't exist in my MAP file, but this location refers to the same place as rs6469836 now in build 38 co-ordinates.

If I look up rs6469836 it is where I say it is, but if you look up rs71468368 (that only appears in the error message) then I get the information that it is not mapped to a particular location (and the three options are on different chromosomes to where I am looking).

Any ideas other than just excluding this SNP for now?

Thanks,
David

[ngsgene]

I am struggling with a similar MAP error - and a million SNPs so I doubt the manual part is going to save me here but liked your idea and can use unix cmds for file formatting to help solve this - can you please link out where the demo file is?

[Maxime Lamontagne]

When you download PLINK, they are in the folder. Their names are test.map and test.ped.

[ngsgene]

Thanks Maxime, got that figured out.

[BW]

Hi Dave,

I have exactly the same problem but with 500 SNPs, so I don't know how to fix it. I also tried adding --map3 but it didn't work as well....

Answer 1

[mosedaled]

I'll respond to my own post!

OK- so thinking that there might be a problem with a particular SNP I delete the data for it from both the PED and MAP files and reload the data, and get another ERROR Problem with MAP file message.

This time it gives a different SNP, which also does not exist in the MAP file. But the location is different - it is now the location of the very next SNP in the MAP file.

So on both occasions, it has read the first 10 rows of the MAP file and failed on the 11th. This seems to me to be more likely now to be an issue with the file in some fashion rather than the data in it?

But what? How does it read the first part of the file but not the rest of it?

Hope someone can help,

Dave

[Maxime Lamontagne]

Can you post the first 15 rows of your map file.

[mosedaled]

Hi Maxime,

Thanks for responding, but I have now resolved the issue. I still have no idea what the problem was, but it was some sort of file format thing. Fortunately I only have 43 SNPs, so I took the demo .map file that is provided with PLINK and manually updated each of the rows and saved it and it works. Looks identical to the file I generated automatically, but at least I am off and running.

Thanks again,
Dave

[harturchin]

Hi Dave,

Could you describe specifically how you edited that PED file? I have exactly the same problem: a MAP and PED file for a small number of SNPs, the formatting looks fine, but PLINK gives me the same error message

ERROR: Problem with MAP file line:
1    rs11309233      0    207295888

That rs# is on the 11th line in the MAP file (same as your error), but the location 207295888 is actually on the previous line of the MAP file (same as the error you described). I guess it's a file formatting problem (like a wrong end of line character) but I can't see it. Thanks a lot for suggestions how to edit.

Mike

Answer 2

[lyping1986]

Just add --map3 if your map file only have 3 columns