Differences in annotate_SJ.R diposited in Baltica and scripts folder

[Antgomoplx]

Hi,

I saw differences in Baltica's actual release regarding this script ( the one in Baltica's folder) compared to the one located in scripts folder. In the first one I see that the function filter_hits_by_diff is not used, so as far this is the main point of the results integration among all 4 methods (https://dieterich-lab.github.io/Baltica/integration.html#annotating-the-results) I want to know if this has changed by the use of the simple merge by coordinates in the script ( line 354)

Many thanks in advance!

[tbrittoborges]

Hi, Antgomoplx, you are correct. Currently, filter_hits_by_diff is not used as we recently preferred to keep the integration simple. I will correct the error in the documentation. Thank you for the feedback!

Just to detail:

• filter_hits_by_diff method allows any SJ with a maximum n nt (defaults to 2) difference to "cluster" - we assume these are the same.
• the current approach uses integrate_coordinates to find the most common difference in nt between the query and the reference annotation and uses this number to offset the coordinates.

(sorry about the delay, I am out of the office)

[tbrittoborges]

fixed in bad2483