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
(Preface: I might be under caffeinated and seeing things, BUT)
I think the sharded FastQs never actually make it to mapping?
It looks like a channel ch_input_for_mapping is correctly created with them, but then all mappers in the if/else section of the MAP SWF start with: ch_input_for_mapping = reads, which pulls the original unsharded reads again... 🤔
I also get the feeling that the mapping output files should actually contain the shard number, to ensure they are kept when merging lanes. 🤔 🤔
The text was updated successfully, but these errors were encountered:
Yes you are right we are using the original unsharded reads. We should be using ch_input_for_mapping.reads instead. Under caffeinated brain works just as well as the caffeinated one ;)
(Preface: I might be under caffeinated and seeing things, BUT)
I think the sharded FastQs never actually make it to mapping?
It looks like a channel
ch_input_for_mapping
is correctly created with them, but then all mappers in the if/else section of the MAP SWF start with:ch_input_for_mapping = reads
, which pulls the original unsharded reads again... 🤔I also get the feeling that the mapping output files should actually contain the shard number, to ensure they are kept when merging lanes. 🤔 🤔
The text was updated successfully, but these errors were encountered: