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
It would be a good idea to implement some sort of scoring function for each tool individually as each tool uses a slightly different approach for fusion detection.
STAR-Fusion
Ericscript (own score)
Fusioncatcher
Pizzly
Squid
Could clustering be an option here?
This issue is used for discussions and suggestions.
The text was updated successfully, but these errors were encountered:
Some weighted voting system maybe? I guess all tools have pretty different benchmarking results, so could be weighted accordingly based on real-world tests...
e.g. benchmark on some datasets and then justify using the ones with highest SE and SP giving them most weight.
Closing and moving the issue to a new repository where the discussion can continue: Clinical-Genomics/fusion-report#2.
I've implemented a basic scoring function, we will see what will be the feedback.
It would be a good idea to implement some sort of scoring function for each tool individually as each tool uses a slightly different approach for fusion detection.
Could clustering be an option here?
This issue is used for discussions and suggestions.
The text was updated successfully, but these errors were encountered: