DefineClones default distance appears incorrect
Default distance is 0.01 in bygroup subcommand. Is this correct? Also, default distance should probably be model specific.
Comments (3)
-
-
reporter Courtesy of Moriah:
"I have been running DefineClones with different --dist parameters, however, the number of clones didn't change at each run. From looking in the R code (ClonesByDistS5F_Full.R) a bit I have noticed the following two code lines:
BinaryDist[BinaryDist<Thresh & BinaryDist>0]<-1
BinaryDist[BinaryDist>=Thresh]<-0
In the order these two lines are placed, with thresholds that are lower than 1 all the cells with a distance value lower than the threshold are replaced by 1, but in the second line they are changed to 0...
From what I understand just by changing the order of these lines this can be solved."
-
- changed status to resolved
Default distance changed to 0 (requiring identical CDR3 sequence for any bygroup model) Additionally, BinaryDist lines switched in order to fix bug
- Log in to comment
The distance is pactically set to zero, it doesn't change even when using different --dist parameters. I think there is a bug in the ClonesByDistS5F_Full.R code.