Question: Output Bias Proportional to Sequence Length?

[dvanderwall]

Thanks for your earlier replies. Really enjoying this software. One thing I have noticed is that outputs seem biased towards returning top results for proteins that have longer sequence lengths. It also appears that there is a filter for sequence length available. What solutions/workflows have you found to be most useful for eliminating this bias, i.e. what cutoffs to use, which proteins to eliminate from index, etc. Thanks in advance.

[khb7840]

Can you share an example of this behavior please? I would like to reproduce the issue locally. Currently, if same hits were identified, the shorter ones gets higher score because Folddisco applies length-based penalty in scoring.