geneid main program actions:
- Initialize accounting structure
- Read setup options from command line
- Set ratios (for sites/exons memory allocation) from file size estimation
- Request memory for main geneid data structures
- Read parameters file (statistical model to predict genomic elements)
- [optional]. Read annotations and homology SRs from input file
- Read first DNA sequence from input file (more than one allowed)
- Processing sequence: upper case, reverse and complement
- Divide sequence into overlapped fragments (length = LENGTHSi)
- For every fragment do:
- Measure G+C content to select the proper isochore
- Forward strand predictions: sites, exons (both scored and filtered)
- Reverse strand predictions: sites, exons (both scored and filtered)
- Merge forward and reverse predicted exons (and annotations)
- Integrate external information: evidence and HSPs
- Sort by left (minor) position the whole set of predicted exons
- Print genomic elements: sites and exons (selected format)
- Run genamic algorithm to assemble the best gene possible
- Save necessary information to go on computing the next fragment
- Delete the rest of information (to save memory)
- Print best prediction: multiple gene output (selected format)
There is a partial prediction mode: reading exons directly from input file
and running assembling algorithm (genamic) to display the best gene predicted
|