Wednesday, January 18, 2012

Teaching Science: Ur still doin' it wrong

Ok, I know this post is really long so the tl;dr version is: don't go to premed2 for information on genetics.

Alright, it's time for premed2's second video: More on Genes!

This video was by far the worst of the two because it had way more glaring errors. The last video pretty much only had one gigantic error in it that could be attributed to the Scientific American article itself being misleading. I can't really imagine what our self-proclaimed "premedical sciences" tutor was thinking when he produced the second video.

Now where to begin....  I know! How about with this image!

The most obvious error in the entire video was the reversal of "transcription" and "translation". I probably would have let it slide as just a typo if he hadn't:
                                  a) had it right in the first video
                                  b) read the two terms in the wrong order multiple times
                                  c) taken the time to underline the words
I mean, really? He couldn't even be bothered to put an annotation over top or re-recorded it? It's not like he spent a load of time creating the video in the first place. Transcription and translation may be obvious to people with a background in this sort of thing but this is youtube. The reason this sort of thing irritates me so much is because people are learning things that are incorrect and it's much harder to re-learn things properly the second time around.

Now, premed2 goes on to talk about how the Humane Genome Project (HGP) was supposed to result in a golden age of medicine. According to him, one of the secondary reasons the golden age of medicine has not become a reality is, in part, because of epigenetics. Now, let's keep in mind that the term "epigenetics" itself was first coined in 1942 by C. H. Waddington. It's not like the scientists participating in the HGP were taken completely by surprise when it came to epigenetics and human gene expression.

Moving right along, premed2 dazzles us with the revelation that the scientists working on the HGP learned that RNA could "act as a protein"... only they didn't. RNAs that have enzymtic properties (a ribozyme) were first theorized by Crick, Orgel and Woese in 1967. Thomas Cech was the first to actually identify ribozymes in nature during the early 1980s and eventually won a Nobel prize (along Sidney Altman) for his work 1989. Keeping in mind that the HGP was initiated in 1990 and completed in 2003, it's hard to imagine it resulted in a discovery that was made more than two decades earlier.

Premed2's next interesting "fact" involves the hypothetical number of genes that code for functional RNA. Chris and I spent quite a bit of time trying to figure out what his source for this estimate was but we were unsuccessful. Premed2, if you're reading this, mind helping us out here? We really have no idea what you're talking about.

We did find some cool papers that discussed the potential number of genes that could code for functional RNA. The best of those papers gave a few estimates combined with a lot of disclaimers about how they are highly speculative (not unlike the original estimates made for protein coding genes). This means that the estimates could be off by orders of magnitude, but here's what we know from this paper published in 2007(unfortunately not open access):
  •  De novo gene-prediction programs have identified ~30,000 regions that could code for structured RNA. Again, it is important to note that the false positive rate for the programs is expected to be quite high.         
  • ~4000 noncodingRNA (ncRNA: RNA that does not code for a protein product) genes have been detected using conservative automated methods
  • Only about 1500 ncRNA have been manually identified.     
Again, as the paper points out, most of these numbers are highly speculative. It's incoherent to declare ribozymes as rare or common at this point. The one thing I can tell you for sure is that we have been aware of their existence for quite some time. Epigenetics and ncRNA genes were not discovered during the HGP nor were they unexpected.

Now this next claim deserves a timestamp because it's really strange. At 5:30, we are told that another reason the golden age of medicine has been delayed is because an assumption made by the "vast majority of scientists" was demonstrated to be incorrect. What is this assumption? The Common Disease-Common Variant (CDCV) hypothesis, of course!

The hypothesis basically went like this: common genetic diseases in a given population are likely to be caused by common variants (alleles/copies of a gene). In human beings, autosomal chromosome genes have two copies, one from your father and one from you mother. It gets a bit complicated with sex chromosomes but that will have to be saved for another post. The "versions" of a gene you inherit are called alleles (also sometimes referred to as variants). Typically, a healthy individual will have two alleles per gene (one on each chromosome). Within a population there can be any number of alleles present for a given gene. In population genetics, we typically describe an allele as common if it is present in a frequency ≥ 1%. In other words, 1% or more of the total copies of a given gene are of a specific allele.

I'm not really sure why premed2 thinks that the CDCV hypothesis was widely held before the HGP. It was never widely held. In fact, it seems it was first proposed in 1996 (6 years after the HGP began). It was never accepted as more than hypothetical and has been discredited for some time now. For a cool paper that discusses why the CDCV hypothesis was a bit unusual this would be worth checking out.

I'm sure there's more I could pick apart but this has gone on long enough. Premed2, if you can't be bothered to get things right, don't bother at all.


  1. I died a little inside when I saw that he'd muddled Transcription, and Translation. I guess there probably are a lot of smallish points also, that he doesn't get completely right; DNA->RNA->Protein, is the sequence hypothesis, not the Central Dogma.

    I also don't really like how he defined a gene in some way that I though poor, then go on to demolish that definition and have a sneaky jibe at biologist - accusing them of being dogmatic.

    1. Yeah that was a pretty ridiculous mistake. I was going to go after him on the Central Dogma mistake but it seemed pretty minor relative to everything else that was wrong. If I addressed everything who knows how long this post would have been haha.

      Still, according to premed2 and those who think like him - I'm just being pedantic by nitpicking minor details. :P