How to set the parameter for augur LBI?

Hi, community members

Is there any general idea about how to set the parameters for the interpretation of LBI?
How to evaluate these parameters?

usage: augur lbi [-h] --tree TREE --branch-lengths BRANCH_LENGTHS --output
                 OUTPUT --attribute-names ATTRIBUTE_NAMES
                 [ATTRIBUTE_NAMES ...] --tau TAU [TAU ...] --window WINDOW
                 [WINDOW ...] [--no-normalization]

Calculate LBI for a given tree and one or more sets of parameters.

optional arguments:
  -h, --help            show this help message and exit
  --tree TREE           Newick tree (default: None)
  --branch-lengths BRANCH_LENGTHS
                        JSON with branch lengths and internal node dates
                        estimated by TreeTime (default: None)
  --output OUTPUT       JSON file with calculated distances stored by node
                        name and attribute name (default: None)
  --attribute-names ATTRIBUTE_NAMES [ATTRIBUTE_NAMES ...]
                        names to store distances associated with the
                        corresponding masks (default: None)
  --tau TAU [TAU ...]   tau value(s) defining the neighborhood of each clade
                        (default: None)
  --window WINDOW [WINDOW ...]
                        time window(s) to calculate LBI across (default: None)
  --no-normalization    disable normalization of LBI by the maximum value
                        (default: False)

Best wishes,
Yang

The parameters of the LBI for viruses like seasonal influenza viruses are generally expected to be a fraction (roughly 1/10) of the coalescence time scale. So this would translate to a few months for flu. See here for the theoretical background:

Hi, @rneher

Thanks for your reply.
One more thing I wonder to know about predicting influenza evolution between fitness and LBI in here.

  1. What are the differences between the predictions between fitness and LBI since the LBI can preictally imply fitness variation?
  2. Are there any available methods that can predict the fitness of the influenza virus using nextstrain?

Best,
Yang