On 3/18/24 8:13 AM, hmad3--- via modeller_usage wrote: > I am currently modelling a heteromeric protein using a homomeric > protein as a template, as follows: > - heteromer subunits (target): 2 x aaaaaaaa + bbbbbbbbb > - homomer subunits (template): 3 x cccccccccc > > When aligning the monomer sequences to be used in the .ali file, > should I align aaaaaaaa and cccccccccc separately, then bbbbbbbbb and > cccccccccc and use the sequences generated in the FASTA file of each > alignment in the .ali file -- or should I align all three sequences at > once and use the sequences in the FASTA file in the .ali file? (In the > first case, there would be two FASTA files and "cccccccccc" would be > aligned differently, and both alignments of cccccccccc would be used in > the .ali file. In the second case, there would be one FASTA file.)

Usually for modeling you would want a single alignment containing all the sequences. See the FAQ for an example: https://salilab.org/modeller/FAQ.html#1

Your alignment would end up looking something like

temp1: aaaaaaaa/aaaaaaaa/---------* temp2: --------/--------/bbbbbbbbb* target: cccccccc/cccccccc/ccccccccc*

How you produce the alignment (e.g. manually combining two or more initial alignments in a text editor) is up to you.

Ben Webb, Modeller Caretaker