I use the GERPplusplus software to find constrained elements in multiple sequences alignements (msa). First I calculate the Rejected Substitution (RS) score for each positions using the command gerpcol. Then, the command gerpelem characterize constrained regions using the ouptup of the first command.
According to the documentation, the outputs of gerpelem are:
start end length RS-score p-value
But my outputs do not match:
region 1975 3125 2269.7 7.66825e-28 1150 0 0
region 3279 3684 841.284 9.91585e-17 1555 0 0
region 825 884 140.106 9.70729e-09 1614 0 0
region 494 698 398.796 7.29589e-06 1818 0 0
region 405 427 54.86 7.61356e-06 1840 0 0
The order and number of the columns do not match. And even though the sixth column seems to be the length of the region, assuming column 2 is start and column 3 end of the region, it is the case for the first line, not for the following lines.
I am very tempted to believe that column 3 is the RS/GERP score, but with a limited understanding of the results, it seems difficult to trust these numbers.
Has anyone used gerpelem before? Is it a normal behaviour for this software?
