Several eukaryotic proteins associated to the extracellular leaflet of the plasma membrane carry a Glycosylphosphatidylinositol (GPI) anchor, which is linked to the C-terminal residue after a proteolytic cleavage occurring at the so called ω-site. Computational methods were developed to discriminate proteins that undergo this post-translational modification starting from their aminoacidic sequences.
Evaluates the degree of presence of the C-terminal sequence signal in a query proprotein sequence based on sequence properties extracted from a learning set. big-Pi predictor allows large-scale database searches for potentially Glycosylphosphatidylinositol (GPI)-anchored protein. It predicts GPI modification sites in precursor sequences. The software integrates terms evaluating amino acid type preferences at given motif positions but also terms judging the conservation of physical properties in the query sequence which represent correlation between few or many motif positions.
A prediction system for GPI-anchored proteins. PredGPI is based on a support vector machine (SVM) for the discrimination of the anchoring signal, and on a Hidden Markov Model (HMM) for the prediction of the most probable omega-site.
A Kohonen self-organizing map was developed that predicts GPI-anchored proteins with high accuracy. In combination with SignalP, GPI-SOM was used in genome-wide surveys for GPI-anchored proteins in diverse eukaryotes. Apart from specialized parasites, a general trend towards higher percentages of GPI-anchored proteins in larger proteomes was observed.
A computational system based on the tandem use of a Neural Network predictor and a Hidden Markov Model predictor. The Neural Network is used to select the potential GPI-anchored sequences and the Hidden Markov Model classifies the selected sequences according to four different levels of precision (highly probable, probable, weakly probable, potential false positive). The Hidden Markov Model proposes also up to three possible locations for the anchor/cleavage site.
A program for the prediction of compatibility of query protein C-termini with the plant GPI lipid anchor motif requirements. Validation tests show that the sensitivity for transamidase targets is approximately 94%, and the rate of false positive prediction is about 0.1%. Thus, the big-Pi predictor can be applied as unsupervised genome annotation and target selection tool. The program is also suited for the design of modified protein constructs to test their GPI lipid anchoring capacity.