fhormoz / caviar

50 stars 24 forks source link

GSL error #6

Closed kathush closed 6 years ago

kathush commented 6 years ago

Hello,

I'm trying to run CAVIAR, not very successfully..

I tried first with my own data, but it kept failing. I tried with the sample data that you provide and I get the same error. This is my code: ./CAVIAR –l /cvar/jhlab/Kathy/Plink/caviar/CAVIAR-C++/sample_data/50_LD.txt –z /cvar/jhlab/Kathy/Plink/caviar/CAVIAR-C++/sample_data/50_Z.txt –r 0.95 –o outputresults

And this is the error that I keep getting: gsl: init_source.c:29: ERROR: matrix dimension n1 must be positive integer Default GSL error handler invoked. Aborted (core dumped)

Do you have any suggestions what should I change in order to fix this? Can you also comment which file formats can CAVIAR tolerate? only .txt and .ld? or also .csv?

Thank you!

fhormozdiari commented 6 years ago

Hi,

This is very strange. Can you please run "make clean" and run "make" command. Then, run the sample code. Usually, I get this error when I run CAVAIR without specifying the LD or Z-score file or using the Z-score file instead of LD file.

CAVIAR inputs should be tab/space delimiters. The Z-score file contains the SNPID and Z-scores. The LD files is a (n*n) matrix where n is the number of SNPs in a locus.

Thanks,

kathush commented 6 years ago

Hi Farhad,

Sorry about the delay - I reinstalled the program and ran make clear followed by make. By running the sample code you mean just to execute ./CAVIAR? When I do that I get the same error as previously. I have GCC, Python, GSL and openblas loaded. I also noticed that each time I'm trying to run something, an -rw file "core.xxxxx" is created in the directory.

Any clues what might be going on? Thanks! Kathy

fhormozdiari commented 6 years ago

Hi Kathy,

After you make the program.

1) Run the program as follow: ./CAVIAR -h I expect that it will show a short manual of CAVIAR.

2) Run the sample data as follow: ./CAVAIR -z sample_data/50_Z.txt -l sample_data/50_LD.txt -c 2 -o test -f 1 -r 0.95

Please let me know if any of these commands failed.

In addition, can you email me the output of these result at hormozdiari@hsph.harvard.edu

Thanks