- attached Rplot.pdf
When no junction found, MakeDb imgt doesn't report j_germ_start correctly
Issue #178
new
This line in MakeDb.py imgt
works only if there is a junction:
result['j_germ_start'] = int(junction['5\'J-REGION trimmed-nt nb'] or 0) + 1
When the junction can’t be identified, there is no J-REGION info in the junction file, and j_germ_start
gets value of 1, which is not always correct. Attached an example where j_germ_start should be 4. 5'J-REGION trimmed-nt nb
is empty, J-REGION partial 3prime missing nt nb
is 0 and junction length is 35. MakeDb reports
j_germline_start j_germline_end
1 35
The AIRR formatted output reports:
j_call j_germline_start j_germline_end j_germline_alignment
Homsap IGKJ1*01 F 4 38 gacgttcggccaagggaccaaggtggaaatcaaac
I think that later, CreateGermlines
doesn’t create the germline correctly because of the _start and _end. See the J alignment in the ugly figure attached, showing the alignment. The deleted J nucleotides should be the first 3, not the last 3 (shown with alpha), and end in ‘aaac’.
Comments (2)
-
reporter -
reporter It think the same for D.
- Log in to comment