Hello. I not very skill in extract specific features from the genome. But now I have to extract all kozak sequencences from cDNA. I only need a fasta archive for every gene with NNNNATGNN (kozak context). I have the cdna.fa and cds.fa. Maybe subtract cdna and cds sequences with the restriction of keep ATGNN and eliminate the remaining sequences, even 3'UTR. If you have a better resource could be very useful. Can you help me please. Regards.