sequence ancestor reconstruction outputs protein ancs sequences with residues instead of gaps

[famotsuka]

Dear phyml developers or skilled users

I have been using phyml in my linux for the main reason to predict ancestrals sequences from a set of aligned protein sequences (msa.phylip) plus a tree (tree.nwk) made in other software. The command I usually run is:
/home/default/phyml-master/src/phyml -i msa.phylip -d aa -m LG -f e -v 0 -a e -c 12 -o tlr -b 0 -u tree.nwk --freerates --no_memory_check --print_site_lnl --ancestral

Then, I get the standards outputs from the run: *_tree.txt, *_lk.txt, *_stats.txt, *_ancestral_seq.txt, *_ancestral_tree.txt.
Intuitively, I now that in the file *_ancestral_seq.txt have the protein sequences for each node of the tree.

But, when I extract the ancestral sequences of each node, places where should be a gap is filled with a residue. Then I get protein sequences too big. For instance, the query protein is 280 residues long, and its ancestor is 380 residues long.

My question is, are there ways to get ancestor sequences whitout gaps been filed with residues?
I think that someone would have a script or other program to get rid of these residues interpreted as gaps. I would appreciate to know any of them.

Thank you