GWAS results from 

Tobacco and Genetics Consortium. Meta-analyses of genome-wide
association studies implicate multiple loci for smoking behavior. Nature
Genetics. 2010;42:441-7.

See the paper for methods. 

There are four files. 

FILE						CONTAINS
tag.cpd.tbl.gz				results for cigs per day, continuous
tag.logonset.tbl.gz			log age-onset smoking, continuous
tag.evrsmk.tbl.gz			Ever vs never smoked, discrete
tag.former.tbl.gz			Current vs former smoker, discrete

All are tab delimited text with header. Header is:

CHR     SNP     BP      A1      A2      FRQ_A   FRQ_U   INFO    OR      SE      P

chr=hg18

For all, FRQ_A=FRQ_U (only had "freq1")

INFO=1 for all (unavailable)

OR is not an OR !!! Is the linear regression beta for the continuous variables and 
logistic regression beta for the discrete variables. 


md5sum tag.*
ee43737cec4f1b184074ce07ea600b85  tag.cpd.tbl.gz
9216b2742390819b4d5d0e4673926885  tag.evrsmk.tbl.gz
a7c3500db3fbcca9d6590bf6d2deba32  tag.former.tbl.gz
44cf8df18e1ada22f67ab914ff54871c  tag.logonset.tbl.gz


pfs
30DEC2011