I know that when RNA is transcribed from the original strand of DNA it contains introns and exons, and that the introns are spliced out of the strand to provide genetic diversity. However, what I don't understand is, how does whatever is doing this splicing know whether the section it is reading is an intron or an exon? Are there start and stop codes like there are for polypeptides, or is it determined by epigenetic factors like methyl markers? Or is it neither?
Quick answer: we don't really know.
As WYSIWYG said, splice sites do have a sequence signature. The image below (taken from ) shows the consensus for human acceptor and donor sites:
In the images above, the size of a nucleotide represents its frequency at that location. As you can see, there is a clear signal around the splice sites and this signal is used by various programs that do splice site prediction. What is not quite clear yet is how the cell recognizes these signals. Sometimes a "perfect" (identical to the consensus) splice site is ignored by the cell in preference to one that we would consider "worse". This is further complicated by the presence of various downstream and upstream signals such as splicing enhancers, silencers and structural elements (loops, hairpins etc) in the mRNA molecule.
So, to answer your question yes there are start/end markers for introns/exons but they are far more complex than the simple START and STOP codons of transcription. We know know a lot about it but we still don't fully understand the details of splicing.
There are some signature sequence which mark intron-exon boundaries. Usually introns start with a
Also, there are protein regulators of splicing such as SR proteins; other proteins such as Fox-1/2 can control alternate splicing events.