Francis Collins and genetic medicine

August 18, 2009 • 10:11 am

by Greg Mayer

Concerns have been raised by several people (PZ Myers, Sam HarrisRussel Blackford, Steve Pinker, and Jerry) about the appointment of Francis Collins as NIH director, mostly to do with whether he might let his strong religious convictions interfere with scientific judgments. A different point is raised by Ken Weiss at The Mermaid’s Tale (and expanded on here).

…on the surface it [Collins’ appointment] suggests a complete and total victory for the genetic view of life. That might have been fine for the Genome Institute, but seems much less so for NIH overall, because many if not most problems in both medicine and public health are not about genes or genetic variation (though they involve them at least indirectly) but are about environments, many kinds of therapies, prevention, and so on. One doesn’t have to ignore the fact that genetics is certainly fundamental to life, and that molecular biology will become increasingly important, to know that (for example) most common diseases have little to do with genetic variation in any sensible way. [emphasis added]

(Hat tip: John Hawks)

A similar thought occurred to me at the time of the announcement of the completion of the draft human genome in 2000, which I expressed in a letter to the editor of the Racine Journal Times:

Few diseases are caused by a “gene.” Most diseases, in fact, are caused by the invasion of the body by another organism (bacteria, viruses, protozoa). Our susceptibility and resistance to disease may often have a genetic basis, but these too are usually the result of multiple genes in interaction with the environment. Even when a disease does have a singular genetic cause, finding the gene does not necessarily lead easily to treatment or prevention (e.g. cystic fibrosis).

21 thoughts on “Francis Collins and genetic medicine

  1. Yes, but he has an MD, and his Ph.D is in physical chemistry. So it’s not as if his training and focus has been exclusively on genes.

    His management capabilities in the human genome project have been praised by his supporters, while detractors have a different version. Whatever the truth about that is, one should note that management often (not always, of course) translates well, and it is possible that physical chemistry may have been a significant aspect even of his management.

    You’re probably going to have a specialist of some sort as the head of the NIH most of the time, and in most cases you could probably say that the NIH is about so much more than that specialty.

    Genetics may not be especially useful for understanding diseases now, and probably won’t be in the future, but it is likely that a disproportionate amount of cutting-edge research will be in genetics (in oncology, for instance), for which a gene guy at the head of the NIH may be good.

    I thought Collins was fair game on the religion that he wears on his sleeve, and which appears to affect his understanding of neuroscience. This “concern,” though, has no legs that I can see.

    Glen Davidson

    1. I agree. Collins was quick to bail out of Biologos when presented with the opportunity to do real science again. To have failed to bail out quickly would have revealed him to be anything but politically astute.

      Comments here reveal the highly political nature of the job which Collins has taken on. Whatever one’s feelings about his current appointment nobody can deny that he has a proven ability to manage a large-scale scientific enterprise.

      In relation to concerns about Collins’ religious views affecting his capacity to act in the best scientific interest of public health I think Newsweek’s ,Lisa Miller is more to the point. Ebryonic stem cell research could prove to be a litmus test.

      Collins’ solution of allowing the use of embryos surplus to the requirements of fertility treatment looks more to me like a political fix than a religiously motivated partial block – though it could conceivably be both.

      It may be that Collins’ personal ambition is a stronger motivating factor for him than his faith.

  2. Actually, while genomewide mapping is not yielding the bonanza of simple causes for complex disease that was promised, many of results suggest infectious etiology, even of traits like schizophrenia, various forms of cancer, some forms of bowel disease, macular degeneration, and even heart disease.

    The data I refer to is from genomewide searches for susceptibility differences but of course that’s just a flag to tell us that the disease is ultimately infectious (and most of the genetic variation risks are themselves complex or modest in extent)

    1. I agree too. There was a news story this morning where Collins said “There is a need to reinvigorate and empower the scientific community”.

      One of his most immediate missions, he said, is entering the ongoing debate of whether and how to reform the nation’s health care system.

      “This is a national conversation that the NIH should be right in the middle of,” he said. “Not to provide political advice, but provide evidence that is useful in assisting decision makers’ direction (on what) we need to get better health care for (the) nation.”


      “I don’t want anyone in NIH or outside of NIH to think I have a religious agenda in coming to this role. I do not,” he said. “I am the scientific director and I’m going to keep that focus.”

      I will take him at his word.

  3. “Few diseases are caused by a “gene.””

    The one gene, one phenotype oversimplification may be a good thing to remind the media, but I’m fairly positive most geneticists are privy to it.

  4. This is a bigger deal the suggested by some. Note his recent comments about some important questions like “What are all the transcripts in a cell?” This is not science, this is list making. Any fucking primate can generate reams and reams of data, how about doing the more difficult work of figuring out what these lists mean and what important knowledge can be gleaned from genomic sequences.

    One fear I had was that Collins would push more “big science” based on his background. Where “big” means a bazillion data points followed up with a bazillion speculations/conclusions and little rigorous follow up. It looks like this fear may bear fruit.

    1. So your concern is that in the future the NIH won’t train/hire enough bioinformatists and statisticians to write the programs and design the algorithms needed to make sense of all the data?

      Btw., the Broad Institute’s sequencer to scientist ratio is far far greater than that at the NIH.

      A greater concern in my opinion is the National Center for Complimentary and Alternative Medicine. I’m just not sure what Collins’ views are toward it.

      He’s presenting at the NIH this week and taking questions from the crowd. Perhaps I should ask him.

      1. So your concern is that in the future the NIH won’t train/hire enough bioinformatists and statisticians to write the programs and design the algorithms needed to make sense of all the data?

        I disagree with your premise and as such can not answer it. The results of a bioinformatist is not a real world data point, but a computer generated interpretation. Ths could then be tested in the laboratory to become a real world data point, but usually isnt.

        Btw., the Broad Institute’s sequencer to scientist ratio is far far greater than that at the NIH.

        The point being? My concern isn’t what NIH does intramurally per se, but that there will be a push of funds towards more high-throughput data point generating science. Top that off with the push for even more translational research and the basic sciences are being pushed ever towards the periphery.

        re: NCCAM. Agreed. Please do ask, though since congress mandated NCCAM there isnt much changing there, no is Collins likely to cause any ruffles this early. Plus, if the ACGT sequence of the human genome = goddidit, then NCCAM may be the benefactor of some big research dollar increases.

  5. While I share some concern for banking too much on Big Science (i.e. high-throughput, marginally hypothesis-driven data generation), the limitations of this approach are certainly not lost on Collins. He gave the keynote address at the 2009 Endocrine Society meeting in D.C., and was careful to be honest and realistic about the limited successes and usefulness of whole genome (or transcriptome, or proteome) approaches as a predictive tool. He was careful to admit that many disease processes are proximately epigenetic, which may or may not be associated with genetic lesions. That said, his attitude appears to be that given the rapid increase in the rate of data throughput, and the ever-reducing cost of DNA sequencing, we’re bound to come up with something.

    1. Your last sentence hits home at least with me. I have seen too many graduate student seminars where my interpretation of their research focus was “We couldnt think of anything to do or a viable hypothesis, so we did a microarray figuring that we will then somehow be able to come up with a hypothesis”

      1. I see it all the time, too. Surprisingly, formulating hypotheses seems to be one of the harder lessons for graduate students. It’s rewarding to see how much better their science becomes once they properly articulate the questions they’re asking.

  6. Even though genetics may not cause many diseases, I think people overlook that there may be a large genetic component to effectiveness of particular drug therapies, susceptibility to adverse drug reactions, etc. Various types of infection may not be the result of genetics, but if you can tell ahead of time which antibiotics are most likely to help, that’s a huge gain.

    1. I agree – this is becoming increasingly clear. I think advances in pharmacogenetics like you describe is seen as the primary potential benefit of genomic approaches, more so than finding causative disease alleles per se.

  7. I laugh at his religious statements as much as the next person here but I don’t have a problem with his scientific output or stated goals. This whole debate about whether high throughput methods or small scale research is better is a false dichotomy. Both offer advantages and indeed both are needed in order to make the sort of rapid progress we require to solve important medical issues like cancer or HIV.
    The suggestion you occasionally hear that genome sequencing has failed to immediately produce multiple cures is a ridiculous strawman. I can’t think of any serious genomicist who suggested anything like that. It provided a base of knowledge from which to begin to tackle these diseases, not an immediate solution.
    As for overall costs to put things in perspective, a single stealth bomber costs about about a third the cost of the US Genome project.

    1. Speaking of false dichotomies or at least associations. I don’t think anyone here concerned with costs/funding is comparing the cost of sequencing with the military budget. Every dollar used for one sequencing center isnt being printed de novo or coming from the military budget, its being shifted from other NIH funded research areas. Now if you think its important to know every transcript within a cell because that would be really cool and whatnot, go ahead and support that. I, however, think this is trivial information without any goals and funding with how to follow it up.

      BTW I am of the mind, based on my experience, that 1 genome sequence/transcriptional profile generates much more data and leads to many more interesting hypotheses than can be addressed by a single lab.

      Because NIH has been hit hard by budget cuts for almost a decade (Im not counting the stimulus) many labs have closed up shop or are having a hard time staying afloat. A shift of more dollars to sequencing centers that generate tons of data looks good in the headlines, but will lead to even fewer labs to pursue any of these potentially interesting/important findings.

      Hey, but its cheaper than a bomber.

      1. I’m afraid the argument of whether a limited amount of money should either go to the high throughput centers or should be used to support large numbers of smaller groups is not one of science, but of politics. The fact that there will be fewer labs left to pursue findings is not a scientific problem either – it will be the better labs that survive. The social darwinism of the academic scientific career is not a pleasant reality to face for the average scientist (or even the above average scientist). Grant review boards are not asked to consider the career paths of the scientists who make applications, merely to pick the application that guarantees the most results for the money – and in this high throughput science has the advantage.
        One of the worst aspects of this scenario is the realization that research, traditionally a topic associated with mentally creative individuals, is perfectly amenable to being done by robots. Paradoxically more investment in research may lead to greater mechanization and thus result in lower numbers of scientists. At times like this I wish I could meet again my university supervisor who convinced me that a career in research would be good for me – so that I could shake his hand – and then promptly knee him in the testicles.

  8. What the new NIH director thinks of NCCAM matters almost not-at-all. He could fervently hope that it dried up and blew away (as one NIH director likely did) and almost no one would know it.

    The question is what does Congress think about NCCAM.

Leave a Reply to Lorax Cancel reply

Your email address will not be published. Required fields are marked *