Hi there, I am having some issues trying to get the residuals after running my models.
I am using this tutorial as an example, and I am running BSLMM. The file has steps for using GEMMA to fit a BSLMM and then get predicted values of the phenotype from BSLMM estimates. Then my plan is to subtract predicted values of phenotype from observed values of phenotype, which would get me the residuals.
I made my kinship matrix with a neutral set of SNPs:
./gemma -bfile "$GENO_DIR/caespitosa_gemma_neutr" \
-gk 1 \
-o "caespitosa_kinship"
Then ran BSLMM on each separate trait
PHENO_FILE="input_data/gemma_pheno/gemma_pheno_caespitosa.txt"
BFILE="../gen_data/caespitosa_gemma"
for TRAIT in 1 2 3 4 5 ; do
echo "Running trait $TRAIT with LMM"
nvalue=$((TRAIT + 2))
./gemma -bfile ${BFILE} \
-k output/caespitosa_kinship.cXX.txt \
-p ${PHENO_FILE} \
-n $nvalue \
-bslmm \
-o caespitosa_pheno_T${TRAIT}
done
And then am trying to get the predicted values the same way:
PHENO_FILE="input_data/gemma_pheno/gemma_pheno_caespitosa.txt"
BFILE="../gen_data/caespitosa_gemma"
for TRAIT in 1 2 3 4 5; do
echo "Running trait $TRAIT with predict"
nvalue=$((TRAIT + 2))
# Run GEMMA prediction
./gemma -bfile ${BFILE} \
-p ${PHENO_FILE}.txt \
-n ${nvalue} \
-epm output/caespitosa_pheno_T${TRAIT}.param.txt \
-emu output/caespitosa_pheno_T${TRAIT}.log.txt \
-predict \
-o caespitosa_pheno_T${TRAIT}_prdt
done
But I am getting the same error every time
Running trait 1 with predict
GEMMA 0.98.5 (2021-08-25) by Xiang Zhou, Pjotr Prins and team (C) 2012-2021
Reading Files ...
ERROR: Enforce failed for Problem reading FAM file (phenotypes do not match geno file) in src/gemma_io.cpp at line 600 in ReadFile_fam
I don't understand why it does not work, as it works fine in my bslmm. This is what my input files look like
(outlier_analysis) [mvan0045@m3-login1 gemma]$ head -n 10 input_data/gemma_pheno/gemma_pheno_caespitosa.txt
GCMa05 GCMa05 NA 18.9859945586842 1.00176543053093 0.626840007778175 0
CHMa01 CHMa01 NA 18.6030023090797 1.15752251830044 0.715276050768868 0
CPMa10 CPMa10 NA 19.2448561913047 1.05030718781501 0.682523328403877 0
GLMa02 GLMa02 NA 18.6357921316338 1.18760930990084 0.801529994231572 0
PIFa10 PIFa10 9.21044036697652 NA 0.671512768604101 0.693483505944199 1
CPMa01 CPMa01 NA 19.6791417364033 0.921480575339401 0.383187016894056 0
GLMa01 GLMa01 NA 18.2463273687124 1.15752251830044 0.844676718586457 0
FPMa06 FPMa06 NA 20.1854115425048 1.0885856600443 0.715276050768868 0
GLFa07 GLFa07 8.5173931714189 NA 0.351086219275306 0 1
FPMa02 FPMa02 NA 20.1144598066585 0.833861106417453 0.569030896263964 0
(outlier_analysis) [mvan0045@m3-login1 gemma]$ head ../gen_data/caespitosa_gemma.fam
GCMa05 GCMa05 0 0 0 -9
CHMa01 CHMa01 0 0 0 -9
CPMa10 CPMa10 0 0 0 -9
GLMa02 GLMa02 0 0 0 -9
PIFa10 PIFa10 0 0 0 -9
CPMa01 CPMa01 0 0 0 -9
GLMa01 GLMa01 0 0 0 -9
FPMa06 FPMa06 0 0 0 -9
GLFa07 GLFa07 0 0 0 -9
FPMa02 FPMa02 0 0 0 -9
(outlier_analysis) [mvan0045@m3-login1 gemma]$ head ../gen_data/caespitosa_gemma.bim
chromosome_10 chromosome_10_265148 0 265148 G C
chromosome_10 chromosome_10_299552 0 299552 A T
chromosome_10 chromosome_10_427034 0 427034 C A
chromosome_10 chromosome_10_427051 0 427051 C T
chromosome_10 chromosome_10_597889 0 597889 T A
chromosome_10 chromosome_10_598222 0 598222 T C
chromosome_10 chromosome_10_610851 0 610851 G A
chromosome_10 chromosome_10_611080 0 611080 A G
chromosome_10 chromosome_10_611081 0 611081 T C
chromosome_10 chromosome_10_611101 0 611101 A T
(outlier_analysis) [mvan0045@m3-login1 gemma]$ head output/caespitosa_pheno_T1.param.txt
chr rs ps n_miss alpha beta gamma
chromosome_10 chromosome_10_265148 265148 0 1.269917e-04 5.626064e-02 2.340000e-03
chromosome_10 chromosome_10_299552 299552 0 -2.212754e-04 -1.234974e-01 1.490000e-03
chromosome_10 chromosome_10_427034 427034 0 -5.300191e-05 -1.037424e-01 3.100000e-04
chromosome_10 chromosome_10_427051 427051 0 6.667576e-05 3.782375e-02 1.600000e-03
chromosome_10 chromosome_10_597889 597889 0 -1.533992e-04 -3.244584e-01 2.500000e-04
chromosome_10 chromosome_10_598222 598222 0 -2.044246e-04 -1.249578e-01 1.080000e-03
chromosome_10 chromosome_10_610851 610851 0 -1.236232e-04 -5.875985e-02 2.440000e-03
chromosome_10 chromosome_10_611080 611080 0 -1.276230e-04 -4.073762e-02 2.530000e-03
chromosome_10 chromosome_10_611081 611081 0 3.461184e-04 1.834051e-01 1.420000e-03
(outlier_analysis) [mvan0045@m3-login1 gemma]$ head -n 50 output/caespitosa_pheno_T1.log.txt
##
## GEMMA Version = 0.98.5 (2021-08-25)
## Build profile = /gnu/store/8rvid272yb53bgascf5c468z0jhsyflj-profile
## GCC version = 7.5.0
## GSL Version = 2.6
## OpenBlas = OpenBLAS 0.3.9 - OpenBLAS 0.3.9 DYNAMIC_ARCH NO_AFFINITY Prescott MAX_THREADS=128
## arch = Prescott
## threads = 1
## parallel type = threaded
##
## Command Line Input = ./gemma -bfile ../gen_data/caespitosa_gemma -k output/caespitosa_kinship.cXX.txt -p input_data/gemma_pheno/gemma_pheno_caespitosa.txt -n 3 -bslmm -o caespitosa_pheno_T1
##
## Date = Wed Oct 8 08:49:43 2025
##
## Summary Statistics:
## number of total individuals = 152
## number of analyzed individuals = 66
## number of covariates = 1
## number of phenotypes = 1
## number of total SNPs/var = 27037
## number of analyzed SNPs/var = 27031
## REMLE log-likelihood in the null model = -87.2542
## MLE log-likelihood in the null model = -85.7597
## pve estimate in the null model = 1.76384e-06
## se(pve) in the null model = 0.701472
## vg estimate in the null model = 0
## ve estimate in the null model = 0
## beta estimate in the null model =
## se(beta) =
## estimated mean = 10.5162
##
## MCMC related:
## initial value of h = 1.76384e-06
## initial value of rho = 1
## initial value of pi = 0.000369946
## initial value of |gamma| = 10
## random seed = -1
## acceptance ratio = 0.300477
##
## Computation Time:
## total computation time = 1.75702 min
## computation time break down:
## time on calculating relatedness matrix = 0 min
## time on eigen-decomposition = 1.38e-05 min
## time on calculating UtX = 0.00058885 min
## time on proposal = 0.0137673 min
## time on mcmc = 1.75345 min
## time on Omega = 0.88554 min
##
If someone could help me see why it is not working I would very much appreciate it. If there is a different way to get the residuals of the model I would also love to hear about it.
Thank you!
Hi there, I am having some issues trying to get the residuals after running my models.
I am using this tutorial as an example, and I am running BSLMM. The file has steps for using GEMMA to fit a BSLMM and then get predicted values of the phenotype from BSLMM estimates. Then my plan is to subtract predicted values of phenotype from observed values of phenotype, which would get me the residuals.
I made my kinship matrix with a neutral set of SNPs:
Then ran BSLMM on each separate trait
And then am trying to get the predicted values the same way:
But I am getting the same error every time
I don't understand why it does not work, as it works fine in my bslmm. This is what my input files look like
If someone could help me see why it is not working I would very much appreciate it. If there is a different way to get the residuals of the model I would also love to hear about it.
Thank you!