The last word on junk DNA

December 5, 2012 Leave a comment

Sean Eddy has written an excellent article that covers all the issues stirred up by the recent headline grabbing paper from the ENCODE consortium.   He also wrote a similar blog post in response to that publication.

Sean Eddy (2012). The C-value paradox, junk DNA and ENCODE. Current Biology 22, R898-899. [journal pdf] [preprint]

Categories: Bad Science, genomes

The physicist delusion

November 21, 2012 Leave a comment

I was reminded of this appropriate SMB cartoon by dmgregory while browsing the comments on PZ Myers’ post this morning.

Categories: Media, Nonscience

Worms don’t get cancer

November 20, 2012 5 comments

My post yesterday, criticising the idea that cancer cells provide a window into what single-celled life looked like about a billion years ago, got me reflecting on the incidence of cancer in different organisms. All organisms have to control their cell numbers and coordinate cell proliferation rates, so we should expect cancer to be a feature of all multicelled life. However, the one organism that I am most well acquainted with, apart from myself, doesn’t seem to particularly prone to cancer. This tiny (adults are 1-2 mm long) nematode worm, called Caenorhabditis elegans, or just C. elegans to it’s friends, rarely acquires the kinds of tumours that can be readily observed in other animals. When I mention this as a throw away line in non-specialist seminars, this is often seized upon in the question and answer sessions that the follow the seminar. So, why is this, and should we rush out and start making elixirs from ground up nematode worms as anti-cancer agents?  Sadly the answer is no. The reason why cancer is rare in C. elegans is to do with basic differences between how cell numbers are maintained in worms compared to mammals.

At the molecular level there are lots of similarities between worms and us: most of the same molecular pathways that go wrong in human cancer are present in C. elegans (and we can use it as a powerful experimental system to study their properties), and many of the same defects that underlie cancer in humans do cause excessive cell proliferation in C. elegans, it’s just that the effects are rather modest. C. elegans somatic cells seem to be uptight and inhibited compared to their exuberant mammalian counterparts. The extra cells generated are generally well behaved and simply toe the line joining the other cells that form the tissue in which they are generated.

C. elegans can develop tumours, but only in the germ line cells (the ones that produce the egg and sperm cells). The reason for this is that germ cells are stem cells: a population of cells that are maintained throughout the life of the animal, which continue to divide and produce new cells. This creates a continual conveyer belt of cells that gradually turn into sperm or eggs (most C. elegans are females that for a short time can make sperm, which they store to fertilise the eggs that they subsequently make: they have no need for males and thus are the ultimate feminist model organism). In order to become egg or sperm cells though, the cells have to stop dividing. When things go wrong with the mechanism that switches off cell division, as happens in worms carrying certain mutations, the cells never stop dividing, resembling the behaviour of cells in human cancers. In other words the only cells that form genuine tumours in C. elegans are stem cells. This is certainly significant, since our current theory for the origin of human cancers is that they arise either from stem cells, or from cells with stem cell-like properties.

So, the likely reason why C. elegans doesn’t really get cancer is that, unlike in mammals (and many other animals) there are no adult somatic stem cells that can serve as the origin of a tumour. Every adult C. elegans has precisely 959 somatic cells; a strikingly small number of cells compared to our tens of trillions of cells. Importantly, if any of these cells die they cannot be replaced by new cell division since there are no somatic stem cells to provide replacement cells. In contrast, we are profligate with our cells, and can easily afford to replace lost cells by the generation of new cells from our many stem cell populations (such as the mesechymal cells in our bone marrow that generate a variety of cell types, including the red and white blood cells that are constantly lost through cell death). Being able to replace damaged and decrepit cells is clearly a significant boon, but one that comes at a cost. The price we pay for this advantage is cancer.

Kirienko, N. V., Mani, K., & Fay, D. S. (2010). Cancer models in Caenorhabditis elegans. Developmental dynamics 239, 1413–1448. doi:10.1002/dvdy.22247

Cancer does not give us a view of a bygone biological age

November 19, 2012 9 comments

Scientific enquiry often benefits from outsiders bringing a fresh insight. The awesome success of molecular biology in the second half of the 20th century, for instance, was driven by a significant number of physicists, attracted by the challenge of understanding the answer to the question, posed by Erwin Schrödinger, “What is life?“.

However, outsiders can also arrive ignorant of important background knowledge and thus make contributions that do not advance understanding because they misunderstand, or omit important aspects of a research topic. Often the proponents of these “new hypotheses” are so enamoured with their idea that they ignore all reasoned objections (and evidence that refutes them). This morning’s Comment is Free section of The Guardian contained a good (or rather bad?) example of this: “Cancer can teach us about our own evolution” by the physicist Paul Davies. He wrote a similar article for The Guardian last year, and I remember a rather confused student citing this after one of my lectures on cell division at the time. The latest article offers no new insight, and crucially no actual evidence.

Davies writes, cancer “is embedded in the basic machinery of life”, a statement that few biologists would take issue with. The problems arise when he explains why he thinks this is the case. His hypothesis, co-proposed with Charles Lineweaver (another physicist), is that cancer cells have somehow reverted back to an ancient state resembling aspects of the single-celled life that existed “on Earth before a billion years ago”. He appears to be arguing that cancer cells are activating a set of genetic pathways that are normally silenced in healthy cells, and that these pathways date back from the time when our ancestors were single-celled. There is a grain of truth in this view, all cells are generated by cell division, and so must contain the machinery and pathways that regulate and activate this process. Since these processes are ancient, actively dividing cells, such as cancer cells, maintain active cell division machinery. However, the same is also true of the stem cells in your bone marrow, which must continually divide throughout your life to maintain the cells that make up your blood, but that doesn’t mean these cells are somehow recapitulating the conditions that existed a billions years past.

Also, although it is true that single-celled organisms, by definition, cannot get cancer, they do not “have one imperative – to go on replicating”. Even single-celled organisms must still coordinate cell division with other cellular processes and signals from outside the cell. For instance, beginning cell division when there are insufficient nutrients, or when the cell is too small leads to catastrophe. Thus, in both single and multi-celled organisms, cell division must be tightly controlled.

The article takes several other liberties with biology, but I don’t have time to take issue with all of them. The estimable PZ Myers has previously upbraided Davies for his use of the long discredited view that human embryos recapitulate key events in vertebrate evolution as they develop: Aaargh! Physicists! This erroneous conception appears again here: “Every human, for example, possesses tails and gills for a time in the womb.” No! Both fish embryos and human embryos have the same general structures that become differentially specialised as they develop. At no point does a human embryo develop gills!

Davies uses this erroneous statement to make the leap that because genes active early in embryonic development are also those “reawakened in cancer” (this is true, for the most part), then this must be because the cancer cells are reverting not only to a state that existed in the early embryo, but to an ancestral, single-celled state. This is a curious logical jump made less for the strength of the evidence and more because it fits the author’s hypothesis.

There is more at stake here than simply getting the biology wrong, or proposing hypotheses that don’t advance our understanding of cancer. Davies is the principal investigator of one of twelve Physical Sciences – Oncology Centres launched by funding from the US National Cancer Institute, at approximately $2,000,000 apiece. To be sure, there may be benefits from this multidisciplinary approach, but only if those involved get the biology right.

Update PZ Myers has just posted an excellent take down commentary of the Physical Biology paper that yesterday’s Guardian article was based on. Apparently there was also a Torygraph article earlier in the year that I missed, framing this dog’s breakfast of a “hypothesis” as “The final frontier in the war on cancer”!

Categories: cell biology, evolution, Media

Susan Greenfield says that the internet will eat your brains.

September 6, 2012 3 comments

“I didn’t say, and I’ve been misquoted universally, that [technology] rots the brain and it’s bad, I’ve never given value judgements, ever,”
Professor Susan Greenfield, 2011.

I’m writing this, my first ever blog entry, after having attended the talk that Professor Susan Greenfield gave as part of the 2012 British Science Festival in Aberdeen, entitled The 21st Century Mind. The topic that she was speaking about, the influence that our increasing use of computer technology might be having on the development and physiology of our brains, is one that she has presented in numerous television, radio and newspaper interviews and articles.

Her opinion (and it is just that) is that our use of screen-based media, such as social networking and video games, is having a detrimental effect on our behaviour. This is obviously a controversial area and also one that the popular press (especially the Daily Mail) is fond of championing. The problem is that there is scant scientific evidence to support such a claim, and Professor Greenfield’s critics have suggested (for some years now) that perhaps, as a neuroscientist, she should try and address this shortfall in evidence. For a nice summary of Professor Greenfield’s view of this topic and her critics response see Happy Science‘s recent post.

Professor Greenfield began her talk by stating that she was merely bringing her concerns to the attention of society to encourage studies that would address the issue; she was definitely NOT scaremongering (lol!). The first third of the lecture was essentially setting out the evidence, for a lay audience, that the brain is plastic: it is remodelled by our experiences (though apparently this is not the case for goldfish brains). So far, so uncontroversial.

She then set out her hypothesis that computer-based media were having a negative effect by remodelling our brains, making us less empathic and less able to handle social interactions. This latter term she defined as being physical social interactions; where we can see and touch one another. Thus, her concern is that our reliance on social networking and mobile devices means that although we are having social interactions, these lack the richness that comes from actually being in the same room as someone.

She then went on to document the types of interaction that people might have online, illustrating them with Powerpoint slides. It was at this point that the lecture became a Daily Mail article. She essentially presented a caricature of social networking, describing it as people posting banal statements about their cats. However, I was particularly struck by her description of a Google search. She described what happens when you search for the meaning of the word “honour” using Google, and presented a collage of the images that such a search returned. Images. Not the text results that one would usually obtain. This for me was key; she used images because these support her claim about the shallowness of the online life. This selectiveness (cherry picking) characterised most of the “evidence” to support her claim.

Throughout her lecture she was essentially making value judgments – books = good, video games/social networking/mobile devices = bad. At one point she cited some research which showed that more children in the study were able to recognise a dalek than they were a magpie; making it clear that, in her view, being able to recognise the latter was more laudable.

She presented a picture of three children sitting around smiling and laughing, each of them talking on their phones, snidely pointing out that these children, although together, were not interacting with each other. But they were at least interacting. What if she’d shown a picture of three children reading (remember, value judgement: “books = good”)? Zero social interaction, so surely books should be tightly regulated too?!

Then she moved on to video games, but I’m not even going to bother describing her line of reasoning here. You already know how that must have played out (though she had one gem of a slide comparing a Japanese RPG (I think; it was a PS2 game) to War and Peace).

She then covered a study showing a generational decline in empathy (not that we know how robust these findings are) and explicitly linked this to increases in the use of digital technology. Really. There was no discussion of being cautious about correlation and causation.

Oh, and she compared her championing of the link between behavioural problems and computer technology use to that of Sir Richard Doll‘s efforts to show the link between smoking and lung cancer!

A summary slide calling for more studies (yes, that would be a good idea) and then it was all over. There were questions (I asked one about why she had used images as an example of the Google search output), but they were not really dealt with with anything other than defensiveness. I was left with the impression that Professor Greenfield was not really interested with obtaining scientific answers that would allow us to determine the actual risks that digital technology pose, or even having a rational debate, but rather her culturally embedded, value-based, dogmatic views about digital technology would not actually be swayed by any kind of evidence. I  wonder if that’s why she doesn’t really seem keen on carrying out a proper scientific investigation into this issue.

Categories: Media, Nonscience