|
The problem
You might be a research scientist wishing to make a peptide and
antibody for probing a protein's structure and function, or one
developing a vaccine or an immunodiagnostic test,
or you might be an antibody producing
company wishing to expand its product portfolio. At the outset of any of
these pursuits, possibly the most frequently expressed concern is
which peptide sequences to choose for generating antibodies.
 Alternatively,
you may be experiencing difficulty producing antibodies to a purified
protein. Further, you may even be considering artificially tagging
your target protein with an unrelated epitope.
Or, you may be frustrated by the lack of commercially
available antibodies to regions other than protein N- and C-termini,
and, thus, be considering producing your own antibodies.
You may even wish to produce a library of peptides
and antibodies to target specific regions of a protein of interest.
In any of these situations you might elect to choose
between attempting epitope prediction and analysis yourself, which is
difficult, or enlisting assistance from others, such as ePitope
Informatics, who specialise in epitope prediction and target protein analysis.
In addition, we are research-active in these areas.
|
Our solution
Is to provide epitope prediction and analysis services that
address the following issues:
|
 Epitope prediction
accuracy, and the
evidence
that it is better to
use a variety of algorithms
and structure
prediction
approaches. |
|
The evidence that some
epitope prediction
algorithms are better
than others. |
|
Ranking predicted
epitopes in order of likely
antigenicity. |
|
Weighting given to a
region predicted to be
surface located
but not
predicted to be antigenic,
or vice versa. |
|
Optimal "window"
settings of certain
algorithms, and the
possibility of adjusting
these (which changes the
"smoothing
effect" of
the algorithm) to obtain
extra information. |
|
Degree of importance
attached to findings of
secondary structure
prediction. |
|
Analysing predicted
epitopes located on solved
3-D structures. |
|
The presence of amino
acid sequence motifs and
protease cleavage sites
providing further evidence
that a predicted epitope is
surface-located on a target
protein. |
|
Possible antibody
cross-reactivities
identified by searching
primary protein
databases for sequences
showing identity to
predicted epitopes. |
|
The effect on primary
protein database searching
of
low complexity regions
in epitope query
sequences. |
|
Primary protein
database
search parameters that are
recommended for use with
shorter query
strings
(e.g. predicted epitopes). |
|
Setting primary
protein
database search
parameters to allow
gaps
(insertions and
deletions) in
returned
sequence alignments. |
|
Frequently occurring
amino acid signature
sequences
in primary
protein databases, which,
if also present in the
predicted epitope query
sequence, may lead to
important alignments
being
missed. |
|
