NAME

data2gff.pl

A script to convert a generic data file to GFF format.

SYNOPSIS

data2gff.pl [--options...] <filename>

Options:
--in <filename>
--ask
--chr <column_index>
--start <column_index>
--stop | --end <column_index>
--score <column_index>
--strand <column_index>
--name <text | column_index>
--id <column_index>
--tags <column_index,column_index,...>
--source <text | column_index>
--type <text | column_index>
--unique
--zero
--out <filename> 
--gz
--version
--help

OPTIONS

The command line flags and descriptions:

--in <filename>

Specify an input file containing either a list of database features or genomic coordinates for which to convert to GFF format. The file should be a tab-delimited text file, one row per feature, with columns representing feature identifiers, attributes, coordinates, and/or data values. Files may be gzipped compressed.

--ask

Indicate that the program should interactively ask for column indices or text strings for the GFF attributes, including coordinates, source, type, etc. It will present a list of the column names to choose from. Enter nothing for non-relevant columns or to accept default values.

--chr <column_index>

The index of the dataset in the data table to be used as the chromosome or sequence ID column in the gff data.

--start <column_index>

The index of the dataset in the data table to be used as the start position column in the gff data.

--stop <column_index> =item --end <column_index>

The index of the dataset in the data table to be used as the stop or end position column in the gff data.

--score <column_index>

The index of the dataset in the data table to be used as the score column in the gff data.

--name <text | column_index>

Enter either the text that will be shared name among all the features, or the index of the dataset in the data table to be used as the name of each gff feature. This information will be used in the 'group' column.

--id <column_index>

The index of the dataset in the data table to be used as the unique ID of each gff feature. This information will be used in the 'group' column of GFF v3 files only. The default is to automatically generate a unique identifier.

--strand <column_index>

The index of the dataset in the data table to be used for strand information. Accepted values might include any of the following "+, -, 1, -1, 0, .".

--tags <column_indices>

Provide a comma delimited list of column indices that contain values to be included as group tags in the GFF features. The key will be the column name.

--source <text | column_index>

Enter either a text string or a column index representing the GFF source that should be used for the features. The default is 'data'.

--type <text | column_index>

Enter either a text string or a column index representing the GFF 'type' or 'method' that should be used for the features. If not defined, it will use the column name for either the 'score' or 'name' column, if defined. As a last resort, it will use the most creative method of 'Experiment'.

--unique

Indicate whether the feature names should be made unique. A count number is appended to the name of subsequent features to make them unique. Using a base text string for the name will automatically generate unique names.

--zero

Input file is in interbase or 0-based coordinates. This should be automatically detected for most known file formats, e.g. BED.

--out <filename>

Optionally specify the name of of the output file. The default is to use the input file base name. The '.gff' extension is automatically added if required.

--gz

Indicate whether the output file should (not) be compressed with gzip.

--version

Print the version number.

--help

Display the POD documentation

DESCRIPTION

This program will convert a data file into a GFF version 3 formatted text file. Only simple conversions are performed, where each data line is converted to a single feature. Complex features with parent-child relationships (such as genes) should be converted with something more advanced.

AUTHOR

Timothy J. Parnell, PhD
Howard Hughes Medical Institute
Dept of Oncological Sciences
Huntsman Cancer Institute
University of Utah
Salt Lake City, UT, 84112

This package is free software; you can redistribute it and/or modify it under the terms of the Artistic License 2.0.