Any piece of dsDNA
can be
“read”
in six possible ways. There are two strands, either of
which can
be read in the 5' 
3' direction, and on
either strand there are three possible start points, beginning
at the 1st,
2nd, or 3rd nucleotides. Each of these
six possibilities is called a reading
frame.
Because 3 of
the 64 possible
DNA triplets correspond
to mRNA stop codons, a DNA
sequence read at random will
have 3-letter stop signals approximately
once in every 20 triplets. The occurrence of
multiple stops in a particular reading frame indicates that it does not
code for a
polypeptide: this is a "closed"
reading frame. In contrast, an Open
Reading
Frame
(ORF) can be read through several
hundred triplets without encountering a
stop
sequence. ORFs are therefore candidates for
protein-coding exon regions: the
inferred amino-acid sequence can be compared with GenBank
to identify possible analogous proteins.
In the example
above, the longest open runs in reading
frames ##1, 2, 3, 4,
& 6 are very short
runs [gold boxes,
typically < 100
bases] of amino acids without stops. Reading
frame #5 includes an ORF [blue box]
of more than 2,000 nucleotides, corresponding to a protein of more than
600
amino acids.
Figure © 2000 by Griffiths et al. ; text © 2008 by Steven M. Carr