Because of the importance of computational genomics, I am writing this article with utmost urgency in hopes of unifying geneticists and biochemists once and for all. 

As a geneticist, I feel that I’m more equipped to write this plea for unification because I see what’s truly and fundamentally on the inside of this dispute, as opposed to biochemists, who are more used to focusing on the external and therefore superficial aspects of things. 

I'd like to break this dispute down to a science by comparing the two fields. 

By traditional definition, genetics is the study of heredity while biochemistry is the study of the chemistry of life.  Often, geneticists and biochemists are seeking answers to the same highly complex questions.  However, approaches are very different and therein lay the dispute; what approach is truly best?  Starting from the nucleus of a cell and working your way out?  Or starting from the outside of the nucleus and working your way in?  Perhaps a better way of phrasing it would be, is it best to truly get to know an organism by looking deep into its genome and then evaluating its character?  Or is it best to draw conclusions from judging an organism based on its phenotype? 

To illustrate the difference in approaches, let’s discuss a simple hypothetical experiment in detail.

We must first select the proper organism to research.  I'm used to experimenting on creatures with faces, but since I have an option and I am in a kind mood I’ll choose Saccharomyces cerevisiae or, rather, "yeast."   The aim of our hypothetical experiment is to determine if S. cerevisiae gene X and gene Y interact in the same pathway.

In biochemistry, if gene X and gene Y interact then protein X and protein Y will interact in the same biochemical complex.  In order to determine protein interaction, protein complex immunoprecipitation (Co-IP) followed by a SDS-PAGE gel analysis can be performed.  For Co-IP, an antibody is used to probe cell extract that contains numerous proteins for one specific protein.  With the antibody now attached to the protein of interest, the sample can be spun down and the antibody-protein product can be isolated for further analysis.  If protein X and Y are interacting, then it can be visualized in a SDS-PAGE gel.  

Often, several unknown proteins are discovered that are involved in the complex as well.  However, if protein X truly interacts with protein Y, then no matter what specific antibody you initially probe the cell extract with the protein interaction will always be observed in the SDS gel. 

Typically, to verify gene interactions in biochemistry it takes several exhausting rounds of Co-IP followed by SDS-Page gel analysis.

In genetics, in order to deduce gene X and Y interaction, different yeast mutants that contain nonfunctional gene copies can be constructed via transformation.  These mutants along with wildtype strains are then crossed and offspring phenotypes are scored.  In the case of yeast, we can observe prodigy typically by the next day.  Thankfully this type of experiment yields a definitive answer, so no repeat rounds are necessary and the geneticist will make it to Yoga by 4:30 no problem.

I found a paper from 2005 (Wong, et al.) that statistically compares the accuracy of genetics and biochemistry approaches.  The paper also answers our original question; which approach is superior?  The answer may surprise you. 

The genetics approach was shown to be the best way to infer relationships between interacting genes (there were other comparison papers too but I found that this one illustrated my point best).

I know it's a crushing blow biochemists.  And that's hurtful because the facts don't lie.  I realize that you feel inadequate and you're very angry.  But biochemists, please, I beg of you, put your pain aside and cooperate because, in truth, our fields are highly intertwined; you need my genes to create proteins so you have something to study in the lab.  If that doesn't persuade you then this will:  Wong et al., 2005 also suggests that when we pool our two approaches together we obtain the greatest accuracy in results. 

Since we're all being honest with each other, I will tell you I’m actually not thrilled at the idea of unification either because, like all scientists, I prefer to publish papers with a single authorship name.  But just like Wong et al., 2005 figured out, computational genomicists have also discovered that our two fields combined yield more accurate results.  Therefore, I can’t help but feel the overwhelming need to unite since in this economy, surely we can't afford to lose our jobs to droids.


Wong, SL., Zhang, LV. and Roth, FP. 2005.  Discovering functional relationships: biochemistry versus genetics. Trends in Genetics. 21(8): 424-425.