A well-known weakness of many topolgy prediction algorithms is that they often
confuse transmembrane regions with signal peptides, as both these types of regions
in general consist of a stretch of mainly hydrophobic residues.
When applied to whole genome data, it is important for a topology predictor not to falsely
predict actual signal peptides as transmembrane regions as this may lead for example to
inaccurate functional assignments.
SPOCTOPUS extends the original topology prediction algorithm of OCTOPUS by also predicting signal peptides.
The SPOCTOPUS algorithm is a pre-processor to OCTOPUS and uses a neural network to predict a signal peptide
preference score for each of the 70 most N-terminal residues. If these scores are high enough a signal peptide
Its exact location is determined by a hidden Markov model before the remainder of the topology
prediction is performed by OCTOPUS.
The final output contains information if a residues is predicted to be
N-terminal of a signal peptide (n), in a signal peptide (S), in a transmembrane (M), interface (I), close loop (L) or
globular loop (G) environment and the preference for each residue to
be on the inside (i) or outside (o) if the membrane.