1
Vote

Fatal Error : SnpPairs only implemented in runGWAS_woodbury() which requires eigen->n[356] > eigen->rankU[356].

description

Hi,
when I am trying to run FaSTLMM v2.07 on Unbuntu system with command:
fastlmmc -snpPairs -file cleanF2deChr0 -fileSim cleanF2deChr0 -extractSim foodfrequency.snps.txt -pheno test_phe.txt -pheno-name foodfrequency -out test.pairs.txt -tasks 250 0

I am getting the following output:
FastLmmC v2.07.20131102 - Factored Spectrally Transformed Linear Mixed Models [Release]
Copyright Microsoft Corporation -- Licensed Only for Non-Commercial use.
Compiled Nov 2 2013 at 20:09:57 by erg00lx for Linux v3.x kernel
using MKL v11.00.04 - Build: 20130517

++ Start Processing CommandLine:
-- Process -extract: [foodfrequency.snps.txt]
-- SnpIds found: 2048
-- End Processing CommandLine:

++ Start Loading FastLmm Data:
++ Start Loading Covariance Data:
++ Processing PLINK fileset: [cleanF2deChr0]
Warning : Found 'missing' phenotype marker. Excluding [4 8196] from the study
Warning : Found 'missing' phenotype marker. Excluding [4 8202] from the study
Warning : Found 'missing' phenotype marker. Excluding [4 8292] from the study
Warning : Found 'missing' phenotype marker. Excluding [4 9190] from the study
Warning : SNP[DRGA0000658] has no variation. Filtering SNP from dataset.
Warning : SNP[DRGA0000787] has no variation. Filtering SNP from dataset.
Warning : SNP[ALGA0003880] has no variation. Filtering SNP from dataset.
 Number of Individuals Selected:     356
           Number of Phenotypes:       1
            Number of SNPs Read:   38067
            Number of SNPs Used:    1810
-- End Processing PLINK fileset: [cleanF2deChr0]
-- End Loading Test Data:

-- End Loading FastLmm Data:
          Compute/Load EigenSym:
Warning : The kernel has a few Eigenvalues that are a tiny bit smaller
       than zero and are considered to be numerically zero.
Warning : Make the kernel positive semi definite.
Warning : Setting negative eigenvalues to zero.
         Compute GWAS using LMM:
Fatal Error : SnpPairs only implemented in runGWAS_woodbury() which requires eigen->n[356] > eigen->rankU[356].

Am I missing something or doing something wrong?
Best regards,
feng zhang

comments

heckerma wrote Nov 21, 2014 at 6:35 PM

Hi Feng Zhang,

This is a known limitation in the C++ version. It is fixed in the python version at https://github.com/MicrosoftGenomics/FaST-LMM.

Best,

David