**What it does**

This tool reads an annotation file in AGS format and uses it together
with an genome object (GIO) to generate labeled genomic positions for
the selected content. Positive examples are generated as they are
annotated. We use annotated segments of different types as negative
examples. The output is a file in content prediction format (CPF)
providing the chromosome/contig, start and stop position, strand and the label of the
example.

--------

**Example**


------

**About formats**

**GIO format** Genome Information Object is an internal format used to
store relevant information about the considered genome (as extracted
from a FASTA file). Includes for example number and length of
chromosomes/contigs. File is generated by applying the GenomeTool.

**AGS format** Annotation Gene Structure Object is an internal structure that
efficiently stores the information from a GFF3 file. Can be generated
with the GFF2Anno tool. The Anno2GFF converts it back to a GFF3 file.

**CPF format** Content Prediction format is similar to the SPF format,
except that it requires exactly an additional field, as it contains
information on segments not on individual positions::

    1. chrom - The name of the chromosome (e.g. chr1, chrY_random).
    2. contentName - possible choices include tss, tis, acc, don, cdsStop, cleave.
    3. scoreName - possible choices include label, output, Conf, Conf_Cum
    4. chromPos1 - The start position of a segment.
    5. chromPos2 - The end position of a segment.
    6. strand - Defines the strand - either '+' or '-'.
    7. score - The score between -infinity and infinity. If scoreName is 'label', then the score should be either -1 or 1.
