Hi Pat,
I totally agree with you that the sequences should be made to overlap in the same alignment space using the filter.seqs function, but it seems a bit harsh to remove columns across 100,000’s of sequences when there could just be 1 “.” in the column. Is it possible to specify a soft limit eg remove columns when “.” occurs in 10% of the sequences or even 1% of the sequences?
I may be clutching at straws here but we have a couple of samples that have low number of reads, and instead of reducing the number of sequence reads further by filtering out the shorter sequences, we would like to include these without compromising the length of the overall alignment space!
Thanks