NAME

       palinear - Perform Genome-Wide Association Analysis using a linear model

SYNOPSIS

       palinear [ command-line options ]

DESCRIPTION

       palinear runs a linear regression on large imputed data sets in an efficient way.

Options

   Required command line options
       -p, --pheno FILE
              Read phenotype data from FILE

       -i, --info FILE
              Read SNP information from FILE (e.g. MLINFO file).

       -d, --dose FILE
              SNP predictor (e.g. MLDOSE/MLPROB) file name.

   Optional command line options
       -m, --map FILE
              Map file name, containing base pair positions for each SNP.

       -n, --nids NUMBER
              Number of people to analyse.

       -c, --chrom FILE
              Chromosome (to be passed to output).

       -o, --out FILE
              Output file name (default is regression.out.txt ).

       -s, --skipd NUMBER
              How many columns to skip in predictor (dose/prob) file (default is 2).

       -t, --ntraits NUMBER
              How many traits are analysed (default is 1).

       -g, --ngpreds NUMBER
              How many predictor columns per marker (default 1 = MLDOSE; else use 2 for MLPROB).

       -a, --separat CHAR
              Use character CHAR to separate fields in the output file(s) (default is space).

       -r, --score
              Use the score test.

       -e, --no-head
              Do not report header line in the output.

       -f --flipmaf
              Flip  reference  and  effect  allele according to Minor Allele Frequency (MAF) such that the minor
              allele is the effect/predictor allele. If this option is set, ProbABEL will check for each genetic
              variant  whether the Freq1 column in the info file (see the --info option) is > 0.5 and if so will
              flip the probabilities/dosage of that variant such that A1 and A2 are interchanged. This will  add
              a  column called AllelesFlipped to the output file(s), indicating whether the alleles were flipped
              (1) or not (0).

       -l --allcov
              Report estimates for all covariates (large outputs!).

       -b, --interaction NUMBER
              Which covariate to use for interaction with SNP analysis (default is no  interaction,  0).  NUMBER
              indicates the column number of the covariate.

       -v, --mmscore FILE
              Score  test  in  samples  of related individuals. The FILE argument is the name of a file with the
              inverse of the variance-covariance matrix.

       -u, --robust
              Report robust (a.k.a. sandwich, a.k.a. Hubert-White) standard errors.

       --help Print help.

SEE ALSO

       probabel(1), palogist(1), pacoxph(1)

BUGS

       The bugtracker is located at
       https://github.com/GenABEL-Project/ProbABEL/issues

AUTHORS

       Lennart C. Karssen