Query PredGPI web server.

PredGPI web server is a predictor of GPI modification sites.

get_pred_gpi(data, ...)

# S3 method for character
get_pred_gpi(data, spec = 0.99, progress = FALSE, ...)

# S3 method for data.frame
get_pred_gpi(data, sequence, id, ...)

# S3 method for list
get_pred_gpi(data, ...)

# S3 method for default
get_pred_gpi(data = NULL, sequence, id, ...)

# S3 method for AAStringSet
get_pred_gpi(data, ...)

Arguments

data	A data frame with protein amino acid sequences as strings in one column and corresponding id's in another. Alternatively a path to a .fasta file with protein sequences.Alternatively a list with elements of class SeqFastaAA resulting from read.fasta call. Alternatively an AAStringSet object. Should be left blank if vectors are provided to sequence and id arguments.
...	currently no additional arguments are accepted apart the ones documented bellow.
spec	Numeric in the 0-1 range, indicating the threshold specificity.
progress	Boolean, whether to show the progress bar, at default set to FALSE.
sequence	A vector of strings representing protein amino acid sequences, or the appropriate column name if a data.frame is supplied to data argument. If .fasta file path, or list with elements of class "SeqFastaAA" provided to data, this should be left blank.
id	A vector of strings representing protein identifiers, or the appropriate column name if a data.frame is supplied to data argument. If .fasta file path, or list with elements of class "SeqFastaAA" provided to data, this should be left blank.

Source

http://gpcr.biocomp.unibo.it/predgpi/pred.htm

Value

A data frame with columns:

id

Character, name of the submitted sequence truncated to 30 characters.

omega_site

Integer, indicating the sequence position of the omega-site.

specificity

Numeric, indicating the specificity of the prediction.

is.gpi

Logical, is the specificity of the prediction higher than the set threshold specificity

Note

This function creates temporary files in the working directory. If the server is unable to make a prediction due to non-amino acid letters or length of the sequence, the returned prediction is NA (is.gpi column). Protein ids should be shorter then 30 characters.

References

Pierleoni A. Martelli P. L. Casadio R. (2008) PredGPI: a GPI-anchor predictor. BMC Bioinformatics 9: 392

See also

get_big_pi

Examples

if (FALSE) {
library(ragp)
data(at_nsp)

gpi_pred <- get_pred_gpi(at_nsp[1:20,],
                         sequence,
                         Transcript.id)
gpi_pred
}