corresponding position are: identical (i.e., fully conserved), conser-
vation between different amino acid groups of highly similar prop-
erties, conservation between different amino acid groups of low
similarity and no similarity, respectively.
Various sequence feature options are given (active site, domain,
helix, chain, etc.) at the left pane and checking the corresponding
boxes enable displaying these features on the aligned sequences
with distinct colors. It is possible to easily observe the site specific
information on the aligned sequences and to infer the properties of
PQby observing if the same amino acid presents in the feature
carrying column on the output MSA. Clustal Omega tool generates
a guide tree from the output of the MSA, which is useful to observe
the inferred evolutionary relationship between the aligned
sequences.
As mentioned above the MSA parameters can be adjusted when
the procedure is run using the Clustal Omega interface (http://
http://www.ebi.ac.uk/Tools/msa/clustalo/). FASTA formatted
sequences of to be aligned proteins can be copy-pasted to the
sequence window. The results page is similar to the UniProt Align
tool interface. It is possible to download the phylogenetic tree,
which is constructed using the resulting MSA.
As a result of MSA, the functional positions can be inferred on
PQ. In the case of no (or limited) functional region/site informa-
tion due to either low conservation or the lack of region/site
specific annotation for the aligned sequences, it may still possible
to identify potential active sites on the sequences using MSA out-
putsviaevolutionary tracing methods or by trying alternative
approaches, explained starting from Subheading2.4.2.4 Evolutionary
Trace (TraceSuite II)
Search functionally important sites via TraceSuite II web server
(http://mordred.bioc.cam.ac.uk/~jiye/evoltrace/evoltrace.html)
either by uploading or by copy-pasting the MSA output. It is
optional to enter PDB id(s) belong to the protein structure ofPQ,
if this information is available. If so, placePQat the top of the
alignment and change its FASTA header to “>PDBid”, where
PDBid should be replaced with the actual 4 digit PDB structure
entry identifier of the protein. If there is no PDB annotation forPQ
but there is a BLAST resulting protein that satisfy the conditions
explained in Subheading2.1,item a, then its PDB id can also be
used here. When provided, the structural information will help
determining the functional residues according to their location
(i.e., as either at the core or on the surface of the proteins).
After hitting the start trace button, the algorithm displays the
results page following the run. In this page, there are two types of
output: (1) a partitioned phylogenetic tree, and (2) the evolution-
ary trace demonstrating the inferred active sites on the partitions of
aligned sequences, where the buried residues are highlighted. It is
also possible to display the evolutionary trace mapped to eachPhylogenetics-Based Prediction of Functional Sites 59