Thank you for your great suggestions. I ran the leave-one-chromosome-out LMM test in GCTA and went throug ldsc regression again. Here is my response and the log file.
Looking forward to your further suggestions. Thank you.
*********************************************************************
* LD Score Regression (LDSC)
* Version 1.0.0
* (C) 2014-2015 Brendan Bulik-Sullivan and Hilary Finucane
* Broad Institute of MIT and Harvard / MIT Department of Mathematics
* GNU General Public License v3
*********************************************************************
Call:
./munge_sumstats.py \
--out dbgap.ea.nd \
--merge-alleles w_hm3.snplist \
--N 6970.0 \
--sumstats dbgap.ea.nd.mlma.assoc
Interpreting column names as follows:
A1: Allele 1, interpreted as ref allele for signed sumstat.
p: p-Value
A2: Allele 2, interpreted as non-ref allele for signed sumstat.
b: [linear/logistic] regression coefficient (0 --> no effect; above 0 --> A1 is trait/risk increasing)
rs: Variant ID (e.g., rs number)
Reading list of SNPs for allele merge from w_hm3.snplist
Read 1217311 SNPs for allele merge.
Reading sumstats from dbgap.ea.nd.mlma.assoc into memory 5000000.0 SNPs at a time.
Read 3916453 SNPs from --sumstats file.
Removed 3282355 SNPs not in --merge-alleles.
Removed 0 SNPs with missing values.
Removed 0 SNPs with INFO <= 0.9.
Removed 0 SNPs with MAF <= 0.01.
Removed 0 SNPs with out-of-bounds p-values.
Removed 52 variants that were not SNPs or were strand-ambiguous.
634046 SNPs remain.
Removed 0 SNPs with duplicated rs numbers (634046 SNPs remain).
Using N = 6970.0
Median value of b was 0.000417715, which seems sensible.
Removed 23 SNPs whose alleles did not match --merge-alleles (634023 SNPs remain).
Writing summary statistics for 1217311 SNPs (634023 with nonmissing beta) to /scratch/xyyin/impute/depLDSC/dbgap.ea.nd.sumstats.gz.
Metadata:
Mean chi^2 = 1.078
Lambda GC = 1.062
Max chi^2 = 91.437
43 Genome-wide significant SNPs (some may have been removed by filtering).
Conversion finished at Sat Mar 26 14:21:46 2016
Total time elapsed: 19.46s
*********************************************************************
* LD Score Regression (LDSC)
* Version 1.0.0
* (C) 2014-2015 Brendan Bulik-Sullivan and Hilary Finucane
* Broad Institute of MIT and Harvard / MIT Department of Mathematics
* GNU General Public License v3
*********************************************************************
Call:
./ldsc.py \
--h2 dbgap.ea.nd.sumstats.gz \
--ref-ld-chr eur_w_ld_chr/ \
--out dbgap.ea.nd.h2 \
--w-ld-chr eur_w_ld_chr/
Beginning analysis at Sat Mar 26 14:22:05 2016
Reading summary statistics from dbgap.ea.nd.sumstats.gz ...
Read summary statistics for 634023 SNPs.