You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In various places in the code, we use chromosome coordinates in the form [chr1, pos1, o1]. This has a couple drawbacks:
List usage. Despite having a fixed number of arguments, list usage means we're unable to enforce static type-checking on each argument as a result. Prefer shifting to namedtuples
We have different "edge" definitions that concatenate these coordinates across multiple intervals, which makes indexing into these edges confusing.
ex: adding a discordant edge here in the form [chr1, pos1, o1, chr2, pos2, o2, sr_count, sr_flag, sr_cn, lr_count, reads, cn]. No typing, 12 indices to remember the unique meanings of. Modularizing the coordinates would help greatly, ex:
class ChromosomeCoordinates(NamedTuple):
chrom_num: int
position: int
orientation: Literal["+", "-"]
class DiscordantEdge(NamedTuple):
start_location: ChromosomeCoordinates
end_location: ChromosomeCoordinates
....
The text was updated successfully, but these errors were encountered:
In various places in the code, we use chromosome coordinates in the form [chr1, pos1, o1]. This has a couple drawbacks:
namedtuples
[chr1, pos1, o1, chr2, pos2, o2, sr_count, sr_flag, sr_cn, lr_count, reads, cn]
. No typing, 12 indices to remember the unique meanings of. Modularizing the coordinates would help greatly, ex:The text was updated successfully, but these errors were encountered: