NAME

computeRT.pl - Compute retention time

SYNOPSIS

./computeRT.pl [--method=METHOD [--saveparam=PARAM] [--current=CURRENT]|[--learnfrom=LEARN] [--settings=SET]] | [--readparam=PARAM] [--calibratefrom=CAL] [--out=OUT] (filein1 [ filein2[...]] | - )

OPTIONS

--method

The algorithm to be taken (irrelevant if parameters are to be read from a file). Currently available algorithms are: Hodges(averaged sum of amino acid coefficients), Krokhin (modified Hodges) and Petritis (neural network).

--current

For Hodges algorithm, precomputed coefficients can be chosen among several references in the literature: Guo86 (default), Guo86_2, Meek80, Su81, Krokhin04, Sagasawa82, Meek81, Browne82, Casal96, Meek80_2, Browne82_2, Petritis03.

--out

Output text file (or STDOUT if no such argument is given)

--learnfrom

Learn from experimental data in LEARN the coefficients in Hodges algorithm or neural network weights in Petritis. File format of blank-separated columns: "amino acid sequences" "retention time" "additional cols" ...

--settings

XML file with additional settings for learning, filtering, etc. See 'rtset.dtd'.

--calibratefrom

Calibrate the algorithm by fitting some experimental data. Same format than for learning.

--saveparam

Save current params in PARAM, including current coefficients and calibration.

--readparam

Retrieve current params in PARAM, including current coefficients and calibration.

--help

Print a brief help message and exits.

--man

Print the manual page and exits.

All remaining arguments are names of input files; or '-' for the standard input.

DESCRIPTION

This program will estimate the HPLC retention time for a given list of peptides. Several prediction algorithms can be chosen. It is also possible to train the coefficients used in the algorithm or to calibrate the estimates to a set of experimental values.