.\" DO NOT MODIFY THIS FILE!  It was generated by help2man 1.47.4.
.TH EVAL_PREDICTIONS "1" "May 2016" "eval_predictions 1.4" "User Commands"
.SH NAME
eval_predictions \- Compares predicted genes with "real" (or annotated) genes.
.SH SYNOPSIS
.B eval_predictions
\fB\-r\fR <real_fname_list> \fB\-p\fR <pred_fname_list>
\fB\-l\fR <seq_len_list> [OPTIONS]
.SH DESCRIPTION:
Compares predicted genes with "real" (or annotated) genes.
Reports standard measures of prediction quality.  The following
measures are reported:
.IP
\- nucleotide sensitivity (Sn)
.IP
\- nucleotide specificity (Sp)
.IP
\- approximate correlation (AC)
.IP
\- correlation coefficient (CC)
.IP
\- exon sensitivity (ESn)
.IP
\- exon specificity (ESp)
.IP
\- proportion of real exons correctly predicted (CRa)
.IP
\- proportion of real exons partially predicted (PCa)
.IP
\- proportion of real exons with overlapping predictions (OLa)
.IP
\- missed exons (ME)
.IP
\- proportion of predicted exons that are correct (CRp)
.IP
\- proportion of predicted exons that are partially correct (PCp)
.IP
\- proportion of predicted exons that overlap real ones (OLp)
.IP
\- wrong exons (WE)
.PP
All quantities are computed as described in "Evaluation of Gene\-Finding
Programs on Mammalian Sequences," by Rogic et al. (Genome Research
11:817\-832).  Note that CRa + PCa + OLa + ME = 1 and CRp + PCp + OLp +
WE = 1.  Note also that each set (predicted and real) should consist of
non\-overlapping groups of features (see 'refeature').
.SH OPTIONS
.HP
\fB\-r\fR <real_fname_list>
.IP
(required) List of names of files defining real genes (GFF).
.HP
\fB\-p\fR <pred_fname_list>
.IP
(required) List of names of files defining predicted genes
(GFF).  Must correspond in order to <real_fname_list>.
.HP
\fB\-l\fR <seq_len_list>
(required) List of lengths of sequences.
Needed to compute
.IP
certain nucleotide\-level statistics.
.HP
\fB\-f\fR <feat_list>
.TP
List of names of all features denoting exon regions.
By
.IP
default, equal to the single name "CDS".
.HP
\fB\-d\fR <fname_prefix>
Dump full coords of correct, partially correct, wrong, missed,
and overlapping exons to a set of files having the specified
file name prefix.
.HP
\fB\-n\fR <nbases>
Also report stats on "nearly correct" exons, that is, incorrect
exons whose boundaries are within <nbases> of being correct.
Columns will be labeled "NCa" and "NCp".
.HP
\fB\-h\fR
Print this help message.
.PP
NOTE: be sure stop codons are included in CDSs in both the predicted
and real sets, or in neither set.