How to correctly interpret the output of calcDivergenceFromAlign.pl?

[ala98412]

Hi,

I am using calcDivergenceFromAlign.pl and createRepeatLandscape.pl to plot the repeat landscape, but I've encountered some difficulties.

This is the repeat landscape plotted by the script createRepeatLandscape.pl. The repeat landscape shows that LINE/L2 with a k-distance of 0 has a proportion of 0.463% (basepair 3,023,067) in the genome (genome size 652,930,317).

However, according to the divsum file (output of calcDivergenceFromAlign.pl, sorted in ascending order.):

Class	Repeat	absLen	wellCharLen	Kimura%
LINE/L2	Zebrafish_L2-54_DR	292	148	-50.85
LINE/L2	DR0171017	180	97	-26.53
LINE/L2	rnd-1_family-43	843825	842394	1.88
LINE/L2	DR0172227	6	6	1.95
LINE/L2	rnd-1_family-45	3084378	3077852	2.12
LINE/L2	rnd-1_family-23	3396	3353	2.32
LINE/L2	rnd-1_family-41	397235	393867	2.58
LINE/L2	rnd-1_family-69	1123890	1121156	2.6
LINE/L2	rnd-1_family-20	503263	499225	2.68
LINE/L2	rnd-1_family-18	1105936	1101045	3.13

First, there are no elements with a k-distance of 0.

Second, if kimura% < 0 represents the category of k-distance 0, they cannot occupy 0.463% of the genome, because both Zebrafish_L2-54_DR and DR0171017 only have a small proportion according to the .out file of RepeatMasker:

Repeat	Kimura%	count	basepair
Zebrafish_L2-54_DR	-50.85	6	806
DR0171017	-26.53	12	586

I think I might be misunderstanding these outputs. How should I interpret them correctly?

Best, Jui-Hung

[rmhubley]

Something appears to be wrong with your input to calcDivergenceFromAlign.pl. Kimura divergences should not be negative. Do you mind sharing your alignment data so that I can try to reproduce this?

[rmhubley]

Closing this for now. Please let me know if you continue to have problems.