Wednesday, November 28, 2007

Scientific faith is an oxymoron

This past Sunday (11/24/2007) the New York Times ran an Op-Ed column titled “Taking Science on Faith”, written by Paul Davies, a physicist at Arizona State University. I cannot decide which is more irresponsible; that a member of the scientific profession wrote such a column or that the Times deemed it fit for publication. Davies’ article contains numerous incorrect statements and non-sequitur arguments, and several people have already written detailed rebuttals of it in the scientific blogosphere. I am writing this simply to address Davies’ central premise because I think that it has significant implications in the current cultural and educational landscape---and is therefore worth examining. The problem I have is not with Davies’ opinion---he is entitled to having it and to expressing it. The problem is that Davies makes broad, sweeping and incorrect postulates about the nature of scientific investigation itself, and appears to present these as fact.

Early on in his column, Davies outlines his central premise as follows:

The problem …….. is that science has its own faith-based belief system. All science proceeds on the assumption that nature is ordered in a rational and intelligible way. You couldn’t be a scientist if you thought the universe was a meaningless jumble of odds and ends haphazardly juxtaposed.”

Those three sentences can be evaluated as follows: NOT TRUE, NOT TRUE, and, NOT TRUE.

The scientific method is defined (by the American Heritage Dictionary) as “The principles and empirical processes of discovery ….. generally involving the observation of phenomena, the formulation of a hypothesis concerning the phenomena, experimentation to demonstrate the truth or falseness of the hypothesis, and a conclusion that validates or modifies the hypothesis”. Scientists don’t study phenomena with the “faith” that they will find order, rationality or logical programming underlying in every phenomenon. Indeed, one of the first things scientists do when they encounter a new phenomenon is to determine (based on observation, for instance) whether there is any reason to hypothesize that there may be any order or program underlying that phenomenon. I don’t know about physicists but I can certainly add this from a biologist’s perspective---randomness and haphazardness are not rare in biology, and there is no reason to approach any new phenomenon with a faith that one will find an underlying rational order or program. Anyway, the significant point here is that all science does not proceed on the assumption that nature is ordered in a rational and intelligible way; as a matter of fact, by its defining principles, none of science proceeds on that assumption or faith. Science is based on objective, evidence-based, reasoning; there is no room in science for biased thinking, preconceived notions and prejudice.

It is evident from Davies’ column that he uses the term ‘faith’ to mean “belief in the absence of proof”, not “confidence or optimism based on existing data”. If a person expresses confidence/optimism based on a body of evidence and data then that is, by definition, not faith anymore. If a person expresses confidence/optimism in the absence of supporting evidence or data then that is, by definition, not science anymore. While this may sound like nitpicking on semantics, it is not. The problem is that the logic behind Davies’ premise is critically flawed in a scientific context, i.e. his claim on faith in science is a contradiction in itself. The failure to understand this may explain how popular misconceptions about science arise and gain traction in public thinking. It may explain how, in the year 2007, we find ourselves living in a society where people demand that faith-based (not evidence-based) ideas such as “intelligent design” be taught as science in classrooms.

Davies also goes on to draw fundamentally incorrect parallels between religious belief and science, and here is a good representative example it. He says, “Clearly, then, both religion and science are founded on faith — namely, on belief in the existence of something outside the universe, like an unexplained God or an unexplained set of physical laws, maybe even a huge ensemble of unseen universes, too. For that reason, both monotheistic religion and orthodox science fail to provide a complete account of physical existence”. Now, I do not know what laws he accuses scientists of invoking, but I think he is mistaking hypotheses for facts---not surprising, as faith-based thought frequently represents unsubstantiated hypotheses as fact. But, while scientists may invoke hypotheses as starting points for their investigation into phenomena, they do not represent these as facts until they have sufficient evidence to that effect.

Finally, there is that cryptic line about “failing to provide a complete account of physical existence”. He goes back to this in his concluding paragraph, and I quote, “In other words, the laws should have an explanation from within the universe and not involve appealing to an external agency. The specifics of that explanation are a matter for future research. But until science comes up with a testable theory of the laws of the universe, its claim to be free of faith is manifestly bogus”. So basically his argument, as I understand it, is, “Science cannot explain everything in the universe. Yet scientists continue to explore nature. Therefore they have faith that everything in the universe is rational and ordered and that they will eventually find/understand this underlying order”. But that is just more flawed logic. As I mentioned before, scientists continue to study nature, not on the faith that order and logic will eventually be found universally, but despite having already encountered numerous phenomena that do not follow order, logic or precise programming. The mandate of science is to try and understand nature as it is, to the extent possible---not to hunt selectively for evidence that will support someone’s notion of what it should be. Scientists simply take the results of their studies for what they are and make objective conclusions based on the data---without pre-conceived bias, i.e. without faith.

People with anti-scientific agendas will doubtless try and use the views of Davies and his ilk as support for their cause of equating faith-based ideas (such as Intelligent Design) to science. To these people I have this to say: If you happen to find some policemen that are corrupt, would you conclude that corruption is a defining principle of law enforcement? I’m guessing that you wouldn’t. If you happen to find that some priests are pedophiles, would you conclude that pedophilia is a defining principle of priesthood? I’m guessing that you wouldn’t. So, just because some scientific professionals (by injecting faith into their ‘scientific’ thought process) may be guilty of flawed scientific method, don’t rush to conclude that science as a whole operates on faith. Making such an inference would be, to borrow a phrase, manifestly bogus.

Sunday, November 25, 2007

A PSA regarding the PSA

I just saw this headline “Obesity Affects Test of Prostate” and my mediocre mind immediately came up with, “Well, the gloved finger cannot get a good feel for the status of the gland when it has to navigate ponderous valleys of flesh, so no surprise there….” But my thinking was not only crass (as usual) but incorrect as well. The report was actually about a blood test for the Prostate Specific Antigen (PSA). It turns out that since obese individuals have a greater volume of blood they could show lower concentrations of PSA. This, in turn, could lead to misleading diagnoses that such individuals are disease-free.

Thursday, November 22, 2007

Getting with the (re)program

An important breakthrough in stem cell research:

Two teams of scientists reported this week (here and here) that they have successfully reprogrammed mature human skin cells to become pluripotent stem cells that appear to be equivalent to Embryonic Stem (ES) cells. The NY Times report and the LA Times report on this work cover the attendant implications pretty well, so I won’t rehash it in great detail. The popular press reaction to this has naturally been to speculate that this work could eliminate the need for ES cell research. And, of course, the ignorant and the malicious policymakers alike in Washington have jumped on it to further their fundamentalist agendas (and, dare I say, antediluvian ideas) using pseudo-scientific claims.

Anyhooo, I think that this is a tremendous breakthrough, especially for the research in understanding the development and progression of many, many diseases. One can now take actual patient cells, reprogram them (to make induced pluripotent stem cells (iPS) as one of the groups of scientists termed them) and perform a multitude of crucially important studies that were previously impossible. So I have nothing but unreserved enthusiasm regarding this breakthrough, on that front.

Now, as to the therapeutic potential of iPS cells, I would urge a lot more restraint in optimism. And here are my reasons for the caution;

1) While the early results are promising, there is a lot more research to be done to determine whether iPS cells and ES cells are indeed equivalent in a variety of circumstances over time.

2) From the genes that are used to reprogram the skin cells, to their vehicle of introduction (retroviral vector), the reprogramming involves engineering whose long-term consequences are just not known. iPS cells will not come close to FDA approval in anything remotely resembling their current technological iteration.

3) Most importantly, we are dealing with reprogrammed adult differentiated cells. I refer you back to my previous post on cancer---adult differentiated cells can acquire damage over the course of their lifespan. There is no evidence to suggest that the aforementioned reprogramming can also magically rid the cells of this accumulated damage or that it can restore their genomes to their formerly pristine or clean states. If a cell with damage acquires the capacities for pluripotency and self-renewal, it could very well end up being a cancer stem cell. I think this issue could end up being the largest hurdle that stands in the path of using iPS cells in a therapeutic context.

It should be noted that such reprogramming was done with mouse cells last year (by one of groups of scientists mentioned above) and the reprogrammed mouse cells were used to generate whole mice. So the power of the reprogramming has been well demonstrated in an animal model system. It is also noteworthy that 20% of the mice generated from reprogrammed cells developed cancer. This was attributed to the fact that one of the genes used for reprogramming was c-myc, a known oncogene. But I don’t think it was proven that c-myc was the only reason.

The best approach in science is to let the system guide you and tell you what’s important. If a certain mature cell-type is not normally self-renewed in the body, then it is also likely that there has not been sufficient evolutionary selective pressure to keep that cell-type from accumulating damage over the course of its lifespan. It is very important to bear that in mind when thinking of ‘reprogramming’ such cells and re-introducing them into the human body. Of course, to appreciate that point one would have to appreciate (or at least acknowledge) evolution-----So I expect Washington to welcome this new breakthrough as an ultimate solution and escalate the war on all other types of ES cell research in 5, 4, 3, 2……

Tuesday, November 20, 2007

Emerging insights into cancer

The adult human body contains approximately a hundred trillion cells, a majority of which are post-mitotic (i.e. incapable of self-replication). The inappropriate and uncontrolled proliferation of such post-mitotic cells results in cancer (at least that’s what we’ve known/thought for a while now). Anyway, the transformation of normal differentiated cells is not a simple event/process, because the body has its checks and balances. For a normal cell to turn cancerous, it would have to acquire the hallmarks of cancer, i.e. obtain and sustain proliferation signals, acquire limitless proliferative capacity, circumvent the apoptotic (programmed-cell-death) pathway, elude the immune system and, finally, acquire the capacity to invade tissues and metastasize. Therefore cancer has historically been thought to result from a multi-step process involving the progressive accumulation of damage to post-mitotic cells that eventually leads to their neoplastic transformation.

Over the past couple of years however, a different aspect of cancer has begun to emerge---one that is connected to the basic biological process of cell replenishment. The human body sheds (and replenishes) billions of cells each day. One simple way to replenish these cells would simply be to replicate differentiated cells of the appropriate type (so, to replace shed skin cells the body could just replicate some already-differentiated skin cells in the area) to make up the deficit. But that is not what happens. In reality, cell-replenishment involves a more laborious, energy-intensive process wherein the requisite cells are ‘made from scratch’ each time. This is achieved by drawing on populations of adult pluripotent stem cells and stimulating them to proliferate and differentiate into the required cell types.

What advantage could such a strategy confer on the system? A very important one actually, as it could reduce the chances of somatic evolution. What does that mean? Well, differentiated cells and tissues have already gone through a number of cell divisions and developmental programming----and have probably acquired damage/mutations over time. Proliferation of such differentiated cells would create the risk of propagating deleterious mutations and abnormal cells, some of which could be cancerous. So, by going back to adult stem cell pools for cell replenishment, the body essentially goes back to a ‘clean’ genetic lineage and reduces the chances of propagating aberrant cells.

Here’s where it gets interesting with respect to cancer. A significant implication of the aforementioned strategy is that the genes that regulate patterning and differentiation in our body are not just required during embryonic development. Rather, it is likely that most of these genes also function on a daily basis to direct the differentiation of adult stem cells into myriad cell-types for cell replenishment. What can happen then, if an adult stem cell acquires a mutation that disrupts any of these differentiation programs? Well, one would predict that this could result in the generation of rogue stem cells that, when stimulated to differentiate, may fail to check their proliferation at the appropriate time after differentiation. Remember, these cells are already endowed with the capacity to proliferate, spread and invade tissues (as they are required to do during normal cell replenishment); therefore a single error that disrupts the differentiation program can be sufficient to render these stem cells cancerous. This concept of cancer stem cells is now very hot in cancer research. For one thing, the presence of cancer stem cells may hold the simplest explanation for why many cancers reappear after treatment. [Surgical removal of tumors or chemical/radiological elimination of (rapidly dividing) cancer cells will likely not be able to harm the (slowly dividing) mutant stem cell population. And the next time a mutant stem cell contributes to cell replenishment, the cancer will naturally reappear].

But what is even more interesting about this recently uncovered aspect of cancer is that it invokes a paradigm shift in thinking on the origins of (and possible future therapies for) cancer. Cancer need not only be the result of multiple events in post-mitotic cells. A single mutation in an adult stem cell could cause it. Furthermore, that mutation could very well be in a gene that regulates patterning or differentiation.

A corollary of this paradigm-shift is that cancer research is no longer limited to the signaling police, cell-cycling enthusiasts and the undertakers of apoptosis research. Rather, it has now also clearly entered the realm of differentiation and the evolution of developmental systems. Man, is there nothing evo-devo cannot tackle? Evo-devo would now like a vodka martini, shaken not stirred.