Implement TM-score fill-in for missing data relative to inputs #50
Labels
enhancement
New feature or request
good first issue
Good for newcomers
invalid
This doesn't seem right
Because of Foldseek's evalue-based filtering of TM-scores, we often get masking of actually high TM-scores as 0. We should, for input proteins, calculate the TM-score explicitly for all of the missing data, perhaps using something like
tmscoring
ortmtools
.The text was updated successfully, but these errors were encountered: