For this assignment, we will build models of a sequence for which an
experimental crystal structure has already been determined (A chain of PDB
code 1ujc). (Obviously, no credit will be given for models built using the
1ujc structure as the template!)

1. Template selection. Use Modeller's profile.build() command to search
for suitable templates in the PDB_95 sequence database for the 1ujc
sequence. (Tip: see part 1 of the basic modeling tutorial on the
Modeller web site.). You should receive only one hit (the 1ujc sequence
itself in PDB_95).

Add related sequences to your query using profile.build() against
uniprot90. (This step may be too demanding for computers with slower CPUs
and less memory. If so, you may download a pre-built profile: input.prf.)

Repeat the search using the resulting profile, and pick one of the resulting
hits as your template. Why did this second search find templates? Justify your
choice of template.

2. Model building. Build several models of the 1ujcA sequence, and pick
the best one by a suitable assessment method (not by comparing it to the
experimental structure!). Justify your choice.

3. Comparison. Compare the final model to the experimental PDB
structure. What are the possible applications of this model? Propose at
least one experiment that could be used to validate the model. Suggest some
applications of a higher accuracy model, and ways to achieve such a model.