If name option is used when mothur generate the representative sequence for each otu, will the representative sequence be different comparing to name option not used?
In other words, will this change the way mothur calculating the mininum distance to the other sequences in this OTU ?
If you are following the Costello example, your distance file will contain only the unique sequences, but the list file will contain all sequences. When mothur looks for a representative sequence it processes each sequence in the otu. For each sequence mothur sums the distances to all the other sequences in the otu. If a distance is not available, for instance because you used a cutoff when calculating the distance file or because the sequence is not unique and therefore not in the distance file, then mothur adds the cutoff, by default the cutoff is 10.05. Adding a namefile does not change how mothur finds the representative, unless you set the weighted parameter to true. If you provide a namefile and set weighted=true, then when mothur finds a redundant sequence it will use the unique sequence’s distances from the distance file. In this case more abundant sequences will then have smaller totals because their distance to the sequences they represent will be 0 instead of the cutoff. I hope this helps.
-Sarah
Thanks Sarah, the weighed option exactly address my question. Really helpful.