One enhancement we've discussed for the next PICRUSt release is using NSTI values to inform strain selection in sequencing. This would be a new script inside PICRUSt that greedily selects genomes for sequencing in order to lower NSTI scores for a given environment, represented by an OTU table.
Required Inputs:
-- Tree (assume Greengenes by default?)
-- Count Table (needed only to know which strains have information).
Optional Inputs:
-- Target OTU table. If provided, look to minimize NSTI vs. the organisms actually present in the OTU table.
-- File providing a list of available strain ids [e.g. if someone is looking at a culture collection]. If not provided, assume all tips can be cultured (will be much slower).
Outputs:
-- Greedily selected set of strains for sequencing with estimated cumulative reduction in NSTI per strain.
One enhancement we've discussed for the next PICRUSt release is using NSTI values to inform strain selection in sequencing. This would be a new script inside PICRUSt that greedily selects genomes for sequencing in order to lower NSTI scores for a given environment, represented by an OTU table.
Required Inputs:
-- Tree (assume Greengenes by default?)
-- Count Table (needed only to know which strains have information).
Optional Inputs:
-- Target OTU table. If provided, look to minimize NSTI vs. the organisms actually present in the OTU table.
-- File providing a list of available strain ids [e.g. if someone is looking at a culture collection]. If not provided, assume all tips can be cultured (will be much slower).
Outputs:
-- Greedily selected set of strains for sequencing with estimated cumulative reduction in NSTI per strain.