Researchers have come a long way from initially cracking the DNA code since the time of Watson and Crick, to now unveiling the complex layers of molecular codes that make up the cell’s molecular fingerprint.

These codes are no longer restricted to the 4 nucleotide codes of the DNA sequence, but rather a complex web of coding systems that regulate every stage of gene expression, including the epigenetic codes (transcriptional), microRNA codes (translational), as well as codes derived from alternative splicing of RNA transcripts (post-translational). While the existence of these codes are now dogma to most cell biologists, precisely how these codes dictate the identity of cells in a multicellular organism still remains elusive.

A recent Nature News and Views article describes how researchers in the University of Toronto have developed an ingenious computer model to tackle the cell’s alternative splicing code, which differs between cell types from different tissues (Gene regulation: Breaking the second genetic code.

Nature. 465, 45-6; 2010). Using information about regulatory splicing sequences in pre-mRNA transcripts as parameters, the computer model successfully predicted the differential splicing of pre-mRNA transcripts between tissue types in mice. While the success suggests that scientists have finally cracked the splicing code, the news article also reminds us that alternative splicing is also influenced by epigenetic codes in the cell. This is not surprising as recent published studies provided strong evidence for the robust interactions between splicing codes, epigenetic codes (Luco R.F. Science 327, 996-1000; 2010) and even microRNA codes (Kolastra A. Genes Dev. 24, 653-8; 2010).

It is therefore apparent that the computer model still needs to factor in the influence of the cell’s epigenetic and microRNA codes on alternative splicing. Overall, while computer models are a great way to tackle the complex splicing code, they should still be regarded as simplified models of the complex biological cell. Caution should be taken when interpreting computer predictions of the splicing code.