9.3 Running lm_auto example and sample output

The syntax for running a MORGAN program is:

./<program> <parfile> [> <output file name>]

The lm_auto example can be run under the subdirectory `IBD/'

./lm_auto jv_rep_auto.par > auto.out

Below are sections of the output file `auto.out', generated by running lm_auto using the parameter file `jv_rep_auto.par'. Note, as for the program ibddrop, the exact values of the probability estimates will depend on the value of the random seed. The first table of calculations (about halfway through the output) gives probabilities of gene ibd patterns for each marker and the trait locus (in the map order).

Probabilities of IBD patterns Proband gamete set 1: 531 1 531 0 331 0 333 1 pattern marker-1 marker-2 trt-geno marker-3 marker-4 marker-5 label 1 1 1 1 .1985 .2960 .3480 .2920 .2430 .1840 0 1 1 1 2 .1400 .2280 .2690 .2205 .1720 .1145 1 1 1 2 1 .1265 .1495 .2050 .1225 .0975 .0985 3 1 1 2 2 .0145 .0165 .0200 .0145 .0125 .0070 4 1 1 2 3 .0340 .0350 .0515 .0255 .0220 .0130 5 1 2 1 1 .0235 .0150 .0030 .0095 .0185 .0275 6 1 2 1 2 .0755 .0425 .0110 .0640 .0705 .0775 7 1 2 1 3 .0595 .0330 .0085 .0310 .0515 .0570 8 1 2 2 1 .0380 .0260 .0030 .0275 .0300 .0435 9 1 2 2 2 .1035 .0605 .0255 .0860 .1175 .1530 10 1 2 2 3 .0540 .0370 .0150 .0385 .0490 .0580 11 1 2 3 1 .0200 .0080 .0035 .0060 .0140 .0210 12 1 2 3 2 .0495 .0255 .0155 .0385 .0645 .0700 13 1 2 3 3 .0100 .0045 .0060 .0030 .0095 .0170 14 1 2 3 4 .0530 .0230 .0155 .0210 .0280 .0585 15 Probabilities of IBD for pattern set for windows of 3 loci Proband gamete set 1 Pattern set: 0 4 IBD wndw 1 wndw 2 wndw 3 wndw 4 0 0 0 .5325 .4900 .4740 .5385 0 0 1 .0975 .0870 .0625 .0685 0 1 0 .0295 .0440 .0400 .0550 0 1 1 .1275 .0665 .0555 .0315 1 0 0 .0445 .0465 .1330 .1125 1 0 1 .0130 .0085 .0240 .0250 1 1 0 .0255 .1130 .0975 .1030 1 1 1 .1300 .1445 .1135 .0660

Interpretation of these results is similar to that of ibddrop See Running ibddrop example and sample output. Briefly, the probabilities are summarized by ibd pattern. A pattern is a series of integers, one representing each gamete listed in the `set proband gametes' statement. The order of gametes in the output file patterns is the same as the order in which the gametes were listed in `set proband gametes'. Numbers that are the same indicate gametes that are ibd. For instance, in the first row of the table above, the pattern is `1 1 1 1', which means that the values in the first row represent probabilities that all four gametes are ibd at each marker locus and at the trait locus. Likewise, `1 2 1 1' means gametes 1, 3, and 4 are ibd while gamete 2 is not ibd with the others; `1 2 3 4' means all four gametes are not ibd.

The second table in the above output is a result of the window size and ibd pattern statements in the parameter file. Its interpretation is similar to the output of ibddrop when statement `set locus window' was used, See Running ibddrop example and sample output. Recall that in ibddrop, the values in the `IBD' column of the output indicate whether the two gametes specified in the `set proband gametes' statement are ibd (indicated by a `1') or not (indicated by a `0'). With lm_auto, the user can specify additional ibd patterns of interest over two or more gametes. In this example, the parameter file `jv_rep_auto.par' includes the statement `set window patterns 0 4', which indicates that we are interested in ibd patterns `0' and `4', corresponding to `1 1 1 1' and `1 1 2 2', respectively, as discussed in the previous section. That is, we would like to know the probability that either all four gametes are ibd or that the first two are ibd and the second two are ibd, but the pairs are not ibd with one another for each window of three loci. Consequently, interpretation of the `IBD' column of the lm_auto output is as follows. The row headed by `0 0 0' gives probabilities that the gametes do not follow either of the two ibd patterns of interest at all three loci for each window. The row headed by `0 0 1' gives probabilities that the gametes do not follow either of the two ibd patterns of interest at the first two loci in the window, but at the third loci either all four gametes are ibd or the first two are ibd and the last two are ibd, but the pairs are not ibd with one another. In this section of the lm_auto output, the order of the marker and trait loci is the same as above. In this example, the trait locus was between markers 2 and 3. Therefore, the windows are as below:

window

loci

wndw 1

marker 1, marker 2, trait

wndw 2

marker 2, trait, marker 3

wndw 3

trait, marker 3, marker 4

wndw 4

marker 3, marker 4, marker 5

For more information regarding the MCMC parameters and diagnostic output, See MCMC parameters and options.