bug in code

[jzohren]

Hi Christopher,

Thanks for sharing your script to parse RM output. I just used it as part of the FastTE pipeline - very useful!

I went through the R script more or less line by line today as I wanted to change the output format slightly and then noticed a bug. I traced this back to the way you are reading in the RM output in line 26. Basically, if the first column (sw_score) does not start with a white space, all of the columns of that line are shifted to the left and the resulting data frame is incorrect. I found a fix to get around this issue, see below.

Lines in raw RM output:

27245    0.0  0.0  1.7  Chr1      20619926 20623101  (9804570) + TE_00000209     __ClassII_DNA_CACTA_nMITE        2683   5805  (1708)  7595
  625    2.4  0.0  2.4  Chr1      20623765 20623848  (9803823) + TE_00000209     __ClassII_DNA_CACTA_nMITE        5806   5887  (1626)  7596
 1409    0.6  0.0  0.0  Chr1      20624203 20624367  (9803304) + TE_00000209     __ClassII_DNA_CACTA_nMITE        5878   6042  (1471)  7597

Original representation in the R data frame (first row should start with 27245, but all columns are shifted to the left):

sw_score	perc_div	perc_del	perc_insert	qry_id	qry_start	qry_end	qry_left	matching_repeat	repeat_id	matching_class	no_bp_in_complement	in_repeat_start	in_repeat_end
0	0	1.7	Chr1	20619926	20623101	-9804570	+	TE_00000209	__ClassII_DNA_CACTA_nMITE	2683	5805	-1708	7595
625	2.4	0	2.4	Chr1	20623765	20623848	-9803823	+	TE_00000209	__ClassII_DNA_CACTA_nMITE	5806	5887	-1626
1409	0.6	0	0	Chr1	20624203	20624367	-9803304	+	TE_00000209	__ClassII_DNA_CACTA_nMITE	5878	6042	-1471

R data frame after fixing:

sw_score	perc_div	perc_del	perc_insert	qry_id	qry_start	qry_end	qry_left	matching_repeat	repeat_id	matching_class	no_bp_in_complement	in_repeat_start	in_repeat_end
27245	0	0	1.7	Chr1	20619926	20623101	-9804570	+	TE_00000209	__ClassII_DNA_CACTA_nMITE	2683	5805	-1708
625	2.4	0	2.4	Chr1	20623765	20623848	-9803823	+	TE_00000209	__ClassII_DNA_CACTA_nMITE	5806	5887	-1626
1409	0.6	0	0	Chr1	20624203	20624367	-9803304	+	TE_00000209	__ClassII_DNA_CACTA_nMITE	5878	6042	-1471

This is how I fixed it (replace line 26 with these two lines of code):

str.res <- unlist(stringr::str_trim(x))
str.res <- unlist(stringr::str_split(str.res, "\\s+"))

Hope this is useful to others as well.

Best wishes,
Jasmin