Closed
Description
To provide more fine-grained read classifications, LAAVA will need the coordinates of each ITR, not just the overall vector region (status quo). The two ITRs (left and right, or WT-ITR and mITR) are not necessarily the same size, and are not necessarily labeled with the same strings in every vector design that the annotation BED is derived from.
Tasks:
- Add 2 more input fields for the ITR names as they will appear in the annotation BED file, e.g.
itr_l_labelanditr_r_label, oritr_label_1anditr_label_2. - Read those labels from the annotation.
- Use those labels in the Python scripts for classification.
TBD / later:
- The outer coordinates of the two ITRs should be identical to the annotated vector region -- therefore, having these two annotations is redundant with the single "vector" annotation that's currently required. Consider dropping the requirement for a "vector" label in the annotation BED, and use the ITR coordinates to calculate the "vector" coordinates instead.
- Keep the "vector" option as a legacy code path that omits ITR-dependent classifications, or strictly require ITR labels right away?
Metadata
Metadata
Assignees
Labels
No labels