Ztrim

NAME

Ztrim.pl - Trim a Zmapqtl output file for terminal reading


SYNOPSIS

   Ztrim.pl  [-H Hypothesis]  [-l lines]  [-w width] [-t]  [-r]  
           [-s]  [-h] < input > output


DESCRIPTION

Ztrim.pl reads from the standard input and writes to the standard output. It reads the output of Zmapqtl and eliminates columns based on the value of Hypothesis. It is a quick way to make the output file fit in a terminal window without having the text wrap.


OPTIONS

-H
This option requires an integer value of 1, 10, 20, 30,31, 32 or 33. The behavior of the program for each value is explained below.

-l
This option requires an integer value greater than 1. It specifies how often the column headers should be printed. The default value of 0 means that headers will be printed at the top of each block of output. If your terminal shows 25 lines, then it is useful to reprint the header every 21 lines so as to have a header on each page of the output. (The header is 3 lines long...if you use more, then you will need one more line for it.)

-w
This option requires an integer value greater than 1. It specifies character width of the output terminal. The default is 80, which is a common value. It only effects the horizontal bars defining the headers.

-h
requires no operand. If used, it prints a usage message and exits.

-r
Prints R squared values

-t
Prints total R squared values

-s
Prints residual test statistic values


OUTPUT

Different values used with the -H option allow the user to output specific columns from the qtlcart.z file. All options print out columns for the chromosome, marker and postions. All print at least one likelihood ratio test statistic and at least one parameter estimate.

For crosses with two marker classes (backcrosses and recombinant inbreds), we can estimate one parameter value, namely the additive effect. The test is H0 (no QTL) versus H1 (QTL with an additive effect a).

For crosses with three marker classes, we can estimate additive and dominance effects. Let a be the additive effect and d be the dominance effect. We can set up four hypotheses:

H0
No QTL: a = 0 and d = 0.

H1
QTL with an additive effect: a not 0 and d = 0. We refer to the estimate of the additive effect as a1 under this hypothesis.

H2
QTL with a dominance effect: a = 0 and d not 0. We refer to the estimate of the dominance effect as d2 under this hypothesis.

H3
QTL with both an additive and a dominance effect: a not 0 and d not 0. We refer to the estimate of the additive effect as a3 and that of the dominance effect as d3 under this hypothesis.

Using the following values with the -H option yields

  1. This is the default and is useful for crosses with two genotypic classes. The output simply deletes the columns with zeros. You are left with the likelihood ratio of H1 to H0 and the estimate of the additive effect.

  2. This essentially prints the same information as 1 above except it does it for crosses with three genotypic classes.

  3. This prints the likelihood ratio for a test of dominance where the additive effect is zero. You get the H2 to H0 likelihood ratio, the estimate of dominance, the R squared estimates and the test statistic for normality of the residuals.

  4. This prints the H3 to H0 likelihood ratio. You get the hypothesis test likelihood ratio, the estimates of additivity and dominance, the R squared estimates and the test statistic for normality of the residuals.

  5. This option prints everything that option 30 prints, but adds the test of H3 to H1 and the a1 estimate.

  6. This option prints everything that option 30 prints, but adds the test of H3 to H2 and the d2 estimate.


EXAMPLE

  ./Ztrim.pl -l 20 -w 78 -H 30 -s < qtlcart.z | more

will print the test statistic H3:H0 and the estimates of the additive and dominance effects for the results in qtlcart.z. It will also reprint the column headers every 20 lines. This works well for a terminal set to 24 lines by 80 columns. The headers take 3 lines each and the more command chews up the last line of each screen.


SEE ALSO

Zmapqtl(1)


AUTHORS

In general, it is best to contact us via email (basten@statgen.ncsu.edu)

        Christopher J. Basten, B. S. Weir and Z.-B. Zeng
        Department of Statistics, North Carolina State University
        Raleigh, NC 27695-7566, USA
        Phone: (919)515-1934



Home    NCSU Home    E-mail Webmaster