Hi,
I'm just reporting here what could be considered an incompatability with the narrowPeak description from ENCODE (https://genome.ucsc.edu/FAQ/FAQformat.html#format12) which led me to some confusion before figuring it out. According to their description of the 10th field, "-1" is used in the final field if no summit is called.
However, if using finemo extract-regions-modisco-fmt with -p using a peak file with -1 in this field, it will set the "start" coordinate as start-(width/2)-1 (as far as I can tell), when I think the expected behaviour would be that it should set start simply as the start position. This is easily solved by making the 10th field the width of the peak. However, I thought I'd report it in case someone else is having the same problem.
Otherwise, thanks for a great tool!
Elias
Hi,
I'm just reporting here what could be considered an incompatability with the narrowPeak description from ENCODE (https://genome.ucsc.edu/FAQ/FAQformat.html#format12) which led me to some confusion before figuring it out. According to their description of the 10th field, "-1" is used in the final field if no summit is called.
However, if using finemo extract-regions-modisco-fmt with -p using a peak file with -1 in this field, it will set the "start" coordinate as start-(width/2)-1 (as far as I can tell), when I think the expected behaviour would be that it should set start simply as the start position. This is easily solved by making the 10th field the width of the peak. However, I thought I'd report it in case someone else is having the same problem.
Otherwise, thanks for a great tool!
Elias