Description:

Read an external file of exons. These annotations will be added in the final gene prediction either forcing to include them or mixing with the ab initio predictions (depending on the existence of value in the group field (gff) of annotations to preserve or not the complete annotated gene). GFF format line (tab as field separator): Name: locusname of the input sequence. Source: program used to produce the output. Type: biological feature described here. Begin: left position in the sequence. End right position in the sequence. Score: probability to be a real prediction. Strand: forward or reverse reading sense. Frame: number of nucleotides of the left non complete codon. [group]: The way to group a set of exons forming a gene. geneid wildcards: represented with a dot '.' Score: maximum score will be assigned to the annotation. Frame: three exons will be created from this one (3 reading frames). Group: If is present, the set of annotations (exons) with the same group will be preserved as a complete gene as long as they match the gene assembling rules. If not, annotations will be mixed with ab initio predictions as long as they might contribute to be part of the best genes. -- Annotations MUST be ordered by left position (beginning) --

Briefing:

long ReadExonsGFF (char *FileName, packEvidence* pv, dict* d)

Input exons from an external file. They must be sorted by the left position (increasing order). It returns the number of created annotations. Exons with unknown (not in gene model) features are skipped, displaying one warning for every one.

packEvidence* SelectEvidence(packExternalInformation* external, char* Locus)

Select the group of annotations to be integrated into the predictions according to current Locus name (using a hash table - dictionary of locus).

Enrique Blanco Garcia © 2003