--> Currently, build-icm produces output model only in binary form. If the output model is in 
text form, glimmer2 application won't be able to read it correctly and crash.


We compared results obtained by running run-glimmer2.bat with the gene annotation at NCBI.
While the UNIX program predicts no genes in this region, the windows executable gives the 
correct prediction with more than 20 genes.


The whole NCBI genbank file is:
U00096 (http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?db=nucleotide&val=48994873)