Saturday, November 28, 2009

The gene for ... ?

Just how do they pack a human into 23,000 genes?

An interesting aspect of biology in this "post-genomic" age is how it has frustrated many researchers. Here we had expected genes "for" alcoholism, schizophrenia, autism, and all the other ills that ail us, but the road to find them has been rocky, tortuous, and has, in many cases, petered out to nothing. It turns out that genomes are complicated and don't come with user manuals. They didn't develop over evolutionary time in tidy ways that generate bodies and behaviors linearly from the DNA. Fruit flies have 15,000 genes, yeast cells have 6,000, and bacteria have 3,000. Most human genes are shared with bananas (i.e. the encoded proteins would function interchangeably). 99% of our DNA is shared with chimpanzees. So what makes us different, and oh-so-much better?

For every condition like eye color that can be attributed to simple mendelian variation in one or a couple of genes, there are a multitude of other conditions whose origin is not simple, but arises from the network of interaction of many genes. Coat types in dogs is another example of simple genetics, outlined in a very nice paper recently, where three genes suffice to explain most of the variation in dog hair types, from short/smooth (the wild-type) to curly, long, wavy, and wiry:

Combinations of novel alleles at three genes (FGF5, RSPO2, and KRT71) create seven different coat phenotypes: (A) short hair; (B) wire hair; (C) "curly-wire" hair; (D) long hair; (E) long, soft hair with furnishings; (F) long, curly hair; and (G) long, curly hair with furnishings.


But most aspects of biology are not so simple. Autism is the example I will focus on most, though similar observations apply to cancer, schizophrenia, personality/temperament, and on down the list of interesting and important conditions/traits.

Genes function in complex networks of regulation, both from the upstream direction of receiving regulatory signals, and downstream in the gene product's (a protein, usually) modification by other systems like phosphorylation and degradation, to eventual roles in combination with other gene products. It's all a big mess of interacting effects. What we see as the ultimate phenotype is the end result of complex mechanisms and lengthy development. Just as a cello has only four strings but an infinitude of musical expression, so genes can be played in different tissues, at different times, and in different volumes to accomplish many tasks.

For example, one set of genes, BMP1 to 20, (bone morphogenic protein), is an evolutionarily related (i.e. duplicated and diverged multiple times) family encoding small proteins that induce developmental events, like formation of cartilage and bone, when secreted by nearby cells. BMP4 is used repeatedly through development to induce notochord formation, eye formation, bone, and tooth formation, and pituitary formation, among others. To quote from one resource:
"Defects in BMP4 are the cause of microphthalmia syndromic type 6 (MCOPS6) also known as microphthalmia and pituitary anomalies or microphthalmia with brain and digit developmental anomalies. Microphthalmia is a clinically heterogeneous disorder of eye formation, ranging from small size of a single eye to complete bilateral absence of ocular tissues (anophthalmia). In many cases, microphthalmia/anophthalmia occurs in association with syndromes that include non-ocular abnormalities. MCOPS6 is characterized by microphthalmia/anophthalmia associated with facial, genital, skeletal, neurologic and endocrine anomalies."
Not a simple story, is it? BMP4 is used, reused, and reused again for similar purposes all over the body. A drug that inhibits its action would have devastating effects, though if that drug could closely control the timing and place of its effects, it might be very useful. That is one of the many challenges of drug development today.

Conversely, a single trait can be composed of the work of many genes. Down syndrome results from the duplication of an entire chromosome- many genes with slight increments in amount of product produced seems to cause a wide spectrum of altered traits. Autism seems similarly be be the consequence of the action of many genes, defects in any one of which can have similar effects. Autism spectrum disorder (ASD) has strong heritability (70% to 90% estimated), yet searches for the responsible genes have come up with not one, but scores of genes. I'll focus on one study that contributes to this story: "Efforts to map disease genes using linkage analysis have found evidence for autism loci on 20 different chromosomes." That is quite a statement, considering that we only have 24 chromosomes [a reader helpfully points out we have only 23!].

This paper used high-tech genomics to look for tiny deletions and other genetic variations throughout the genomes of families afflicted with ASD. Out of 195 autism patients and 196 controls, they found variants in 14 patients versus 2 in the control set. All the variants were heterozygous, indicating that, like in Downs, small increments or decrements in gene function may be responsible, with one normal gene copy remaining in each case. Some of the genes were expressed in the brain, while others were known to participate in retardation disorders, and others have little known about them.

Two other observations stand out. First is that most of these small genomic duplications or deletions are novel- they happened recently, and being deleterious, will die out rapidly as well. Autism "runs" in some families, but most cases result from spontaneous defects in a wide variety of genes. Second is the large number of genes estimated to be in this pool of possible ASD causers- at least 29 found so far between this and other studies, with more to come as more families are analyzed with ever more comprehensive methods. This is relevant to the rate of occurrence of this disorder, which is very high for genetic disorders, and possibly rising. It is known that autism rates go up dramatically for children of older parents.

So ASD seems to be the result of rare defects in any of numerous genes, many known to be involved in synapse formation and activity. It might be that an over-arching pathway of early brain development channels many genetic problems into the same syndrome, much as many problems with cars result in the common syndrome of "it won't start". Thus to think of genes "for" such diseases is problematic, given the complicated relations. There appears to be a developmental process that generates the syndrome, driven by many genes and susceptible to many distinct defects. And the case of ASD doesn't even touch on the separate issue of genetic variants that individually have tiny phenotypic effects on a trait, (such as height, for instance), but combine with many others to determine the overall trait quantitatively.

More deeply, for the genome at large, it isn't size that matters. Soybeans have almost three times the number of genes we have (though one-third the overall amount of DNA). It is how you use what you've got, in complex networks of regulation, combination, and reuse that makes a brain out of a bunch of cells. As the old Sun slogan had it, the network is the computer.

  • A relevant review of genome-wide association studies of "disease genes."
  • Climate disasters, a tad overdrawn.
  • The casually callous and obtuse David Brooks does in health care.
  • Does the Economist know economics?

Saturday, November 21, 2009

Treating patent addiction

Is the pharmaceutical patent system working for us, or against us?

An interesting web-book by Dean Baker makes the case that our government is socialist- but for the benefit of the rich rather than the poor. The general case has quite a bit going for it, with several trenchant points, though others are pretty weak, such as Baker's discussion of the role of the Federal Reserve.

Baker hints that there are better ways of fighting inflation than raising interest rates. And he maintains that raised rates have their main effect in cutting employment, putting the poor out of work, insinuating that this is some nafarious soak-the-poor policy. I'm not so sure. Raised interest rates raise the price of money and restrict new loans, which are the primary motor of money creation (and of economic activity, in our fallen age) and seem to efficiently combat the core issues of inflation. That this disproportionately hurts those on the bottom rung of the ladder is an automatic consequence of the overall capitalist structure of the economy, not any special property of Fed action. Could the Fed collect dollar bills and burn them? I'm not so sure, for who would give them up without compensation? Baker doesn't actually lay out his alternative plan, (other than direct wage and price controls, which are a singularly wretched tool), probably with good reason.

At any rate, Baker does present one fascinating case study- of the drug industry- which I think is on much firmer ground. To summarize, the current major pharmaceutical industry makes its money from the 20 year patent. Once a compound is discovered and filed for a patent, the clock starts ticking on this government-granted license to sell the resulting drug for whatever the company wishes to charge. (This is typically a function purely of maximizing the product of price times sales into what they imagine the market will bear, rather than recouping actual costs of research and production, or competing against other producers). This system gives drug companies huge potential incentives to create important drugs that they hope become a standard of care, with high resulting market demand.

Drug companies say that they spend an average of $800 million to bring a new drug to market, so their standard of sovency has become the "blockbuster" drug with annual revenues over $1 billion. This system has served the US well in many respects, with leading-edge pharmaceutical research, a continuous flow of new drugs in the pipeline, and a vibrant generic industry to offer rapid price reductions once drugs go off patent.

On the other hand, the system is also riddled with inefficiencies and incentives contrary to the common good. The $800 million figure is, I believe, quite inflated, since much of the work of the drug industry is devoted to making drugs with modest beneficial effects- drugs that may slightly ameliorate chronic conditions like depression, diabetes, or Alzheimer's disease. These drugs are can be extremely difficult to test because their small effect size requires large populations for trials, followed by ornate, if not ambiguous, statistics. Consider penicillin- its ability to cure infections was immediately apparent, and did not require large trials with thousands of patients.

In comparison, consider a drug like donepezil for Alzheimer's disease. It improves cognition slightly for a few months, costs ~$220 per month (still under patent in the US). Numerous studies have been done, covering thousands of patients. (The Cochrane meta-analysis says "23 trials are included, involving 5272 participants"). They typically show just-detectable benefits, which, calculated in cost-of-care terms offers no discernable benefit (cost of drug vs cost of increased care of untreated Alzheimer's sufferers).

Not only that, but each other drug company that wants a part of the action makes a copy-cat drug with slightly different chemistry and similarly marginal benefits, requiring similarly exhaustive and expensive trials to gain FDA approval. And then the drug companies complain that that the FDA is too restrictive! But the fault, dear Brutus, lies not in our stars, or in our government bureaucracies, but in the quality of the drugs being created.

The drug industry has a big problem right now- its pipelines are drying up, despite the huge advances in molecular biology and genomics over the last few decades. As the technology of drug discovery is taking quantum leaps in efficiency and comprehensiveness, there seem to be fewer drugs coming out the other end. This might be due to a true shortage of targets- we only have 23,000 genes, after all, and many of their products have multiple functions, making drug-based inhibition a perilous business, frought with side-effects. There may just be a limited number of ways to use small molecules to interfere with human biology.

New drugs approved (blue- formulations and combinations,
red- underlying novel molecular entities)

On the other hand, the problem might also be institutional in the pharamaceutical industry. To make money, drug companies have to sell lots of drugs. That means large markets and common diseases, hopefully chronic. But not all diseases are chronic or common, or afflict people with ability to pay. Diseases of the third world are notoriously under-researched and so-called "orphan diseases" and infectious (i.e. acute) diseases are likewise underfunded. Rather than do innovative research in to basic disease mechanisms, drug companies spend much of their money making copy-cat drugs in popular categories (think cholesterol reduction, acid reflux, depression, erectile dysfunction). And, of course, drug companies also spend roughly twice as much money marketing drugs than in researching them.

I think both problems are serious and getting worse. Viagra offers an interesting example. This drug was discovered serendipitously when Pfizer researchers looking for drugs for angina and hypertension, (classic categories for profitable drug hunting), all based on publically funded research on NO2 signalling, discovered a completely unexpected effect of one candidate. Now several companies have created similar drugs and great happiness has been rendered to millions. But note that this drug was found by accident, and the condition was not even on the company's radar screen. A double-accident plus a lot of publically funded research brought us this new drug category, which has made drug companies three billion dollars annually for a decade.

Does society really have to pay these exorbitant prices for such modest effort and accidental benefits? Baker claims that the excess cost of the drug patenting system over making all drugs generic from the start is roughly $400 billion per year. Is that a fair price for the research and market motivation that the current system gives us? Remember that the entire NIH, upon whose basic research virtually all pharmaceutical innovations depend in whole or in part, only costs $30 billion annually, less than one-tenth the industry figure.

Baker makes a proposal that with a doubling of the NIH buget to replace the research and development functions of the pharmaceutical industry, including modest prize incentives for development successes, we could gain all the benefits of the current pharmaceutical industry and more (i.e. targeting diseases more equitably, and focusing on significant rather than on marginal effectiveness), for a small fraction of the societal cost. I can't but agree that this is a very reasonable idea.

Indeed, the NIH is already dipping its toes into this pool of drug development, setting up programs to develop drugs for neglected diseases, one sector where our current system is AWOL. If corporate neglect where purely a function of the prevalence and harm of each respective malady, this market structure might be defensible. But often it is a function of the prospective length of treatment, (hopefully forever), the depth of the prospective patient's pockets, and prospective prevalence as juiced by energetic "informational" campaigns for what may have been unknown or minor maladies.

The clamor for buying drugs in Canada has been a sorry commentary on the dysfunction of our patent and drug system. The domestic drug industry uses its corrupt political influence to induce the government to pay whatever the patent holders ask, even for such enormous government programs as Medicare. Canada, not beholden to these companies, and given to more rational social policy generally, does negotiate prices down, attracting the interest of US consumers saddled with uncontrolled costs. Then US politicians, frustrated with their own corruption, get on the bandwagon, either cynically to gain a few votes, or possibly as a way to indirectly pressure the domestic drug industry to moderate prices. It's an insane way to get to where we should be going, which is to rethink the whole rationale of the patented drug sector.

This sector is just one example of the inefficiencies in our sclerotic political-economic system that is saddled with enormous legacies of infrastructure and vested interests as we enter this new century and try to battle our way to a better future. Current Senate rules, for instance, allow Senators representing barely 12% of the population to block any action. Deliberation is one thing, gridlock is another.

  1. Kaufman M. Decline in new drugs raise concerns: FDA approvals are lowest in a decade. Wash Post. 2002; Nov 18:A1
  2. Pollack A. Despite billions for discoveries, pipeline of drugs is far from full. N Y Times. 2002; Apr 19:C1.
  3. Dyer G. Anaemic patient needs to take its medicine: investors have fallen out of love with an industry with fewer products in the pipeline. Financ Times. 2003; Apr 16:2.

  • Others have problems with the drug industry too.
  • Meanwhile, military health care remains abysmal.
  • Religion is natural, after all.
  • Gosh- another atheist dilemma.
  • Just how does Goldman mint its gold? (But also here.)
  • For the love of Islam!
  • Another interesting story on synaesthesia, a fascinating window into consciousness and yet another argument against souls, versus a rather fallible circuitry.

Saturday, November 14, 2009

The warrior religion

Brief review of Glubb's history of early Islam

I am glad to report finding yet another gem in the hoary stacks of the local Catholic library, this time a history of early Islam, The Great Arab Conquests, by Lietenant-General Sir John Bagot Glubb (string of British orders & honors omitted here), 1963. If Amazon is to be believed, this book is out of print, and not only that, but its title was swiped by another author in 2007. This is most unfortunate, since Glubb's work is fabulous- exceedingly well written, frank, pre-politically correct, yet full of sympathy for his subject. Glubb spent his career in the Middle East, serving in both world wars and running the Arab legion, later part of the Jordanian military. He takes particular pride in clearing up a few scholarly confusions using his intimate knowledge of the ground in the Middle East, and of its military uses.

The book focuses on the first fifty years of Islam, retelling the story of Muhammad's life, background, and call, then going on to detail the careers of the first five khalifs ("successors"). Glubb is a military man and focuses on the military aspects of the story, with excellent maps throughout. But as a long-time associate of Arabs, Bedouins, and people throughout the Middle East, he also evinces sensitivity and admiration for their cultures, some of which have persisted with little change from the seventh century. Military affairs were central to the early history of Islam, and to the mindset of Muslims of that time, so this focus is incisive as well as stimulating.

One thing to note is that the quality of the khalifs was highly variable, from the high of Abu Bekr who directly followed Muhammad, to the pathetic impotence of Ali ibn abi Talib, who, along with three others of the first five, was assassinated. Glubb's portrait gives precious little evidence supporting the many partisans of Ali (Shia, or Shiatu Ali), vociferous as they are, since despite having high religious credentials and the closest personal connections with Muhammed, Ali was evidently passed over for the khalifate several times for what ultimately proved to be quite good reasons.

I won't try to retell the whole story, but just say that if you are interested in this history, you could hardly do better that this presentation, be it ever so hard to find! Glubb also wrote a sequel and several other books on the Arab cultures and his experiences.

Let me cite a few of the more striking passages, indicating Glubb's view of Islam and our relations to it.

Speaking of the first two khalifs,
We have already seen that almost the last act of Abu Bekr was to receive Muthanna ibn Haritha, who had ridden in hot haste from the Euphrates to beg for help on the neglected Persian front. The first act of Umar ibn al Khattab on assuming the Khalifate had been to dismiss Khalid ibn al Waleed from the supreme command in Syria. The second has been, as the dying Abu Bekr had ordered, ro raise a new levy for Iraq. Volunteers were at first slow in coming forward, for the Persians seem to have enjoyed the reputation of being more formidable than the Byzantines in war. As a result, recruiting proceeded by slowly, even though Muthanna himself made a speech in the mosque calling for assistance, and describing the immense plunder obtainable by those who followed the path of God and fought against the fire-worshippers. p. 160

In the book's conclusion...
The momentum of the great conquests had been so tremendous that they swept irresistibly forward without organization, without pay, without plans, and without orders. They constitute a perpetual warning to technically advanced nations who rely for their defense on scientific progress rather than the human spirit.
...
A cosmopolitan empire, with subjects professing different religions, could not constitute a devoted and homogeneous people of high morale, such as the Central Arabians had been twenty-five years earlier. p.359


Since the seventh century, many Muslim state have, at various times, established efficient legal systems and police forces, rendering private retaliation unnecessary, but the idea of revenge dies hard. In a wider sense, the right, and even the duty, of revenge has survived all modern reforms, for as a result of these early origins, it has become an accepted moral principle. This, it seems to be, is one of the directions in which Christianity differs most from Islam. Christians are never entitled to return evil for evil. In Islam, retaliation is a right, in some cases being even regarded as a moral duty. p. 367

Particularly is it noticeable that the idea of government by groups of men- cabinets, parliaments or committees- has no precedent at all in Arab history. Their idea of government is always one man. In theory he is chosen by the people. He must be humble, accessible, benevolent, pious and hospitable. Arrogant despots cannot be tolerated but nevertheless executive power must be vested in one man alone. All these traditions can be traced from the seventh century.

At various times since 1918, the Western Powers have painstakingly built up democratic, elective institutions in the countries of the Middle East. In every case, within a few years, these constitutions have collapsed and military dictators have assumed power. Perhaps this is not to be wondered at, for the military dictator is nearer the time-honoured Arab tradition than is Western Democracy. p.369

And this last parting shot:
This long-standing rivalry between Christians and Muslims has been due to political and geographical accident rather than to basic religious differences. Now that materialist atheism is challenging all spiritual values, the two religions might well make common cause against those who deny the existence of God altogether. There is, I believe, an immense field in which the two could co-operate. p. 371

One observation that struck me was the relation of Islam to power. Humans worship power- that is an unfortunate, but consistent, part of our nature (with obvious Darwinian origins). Power is an aphrodisiac to women, the source of male status, and the goal of youthful striving and careerist competition. Religion is little more than an expression of this emotion in over-wrought terms, since God is all-powerful, Jesus is Lord, and prayers and beseeching are our mode of intercession/intercourse with the imaginary beings.

Islam as refined this simple fact of human nature to the highest possible pitch, instituting and naming itself by universal submission to Allah who is great, while at the same time borrowing a bit of that greatness and demanding submission from all non-Muslims as a matter of right, whether the Dhimmi, (Christians and Jews), who are made second class citizens, or the outright infidels, who are offered conversion or death.

That is why terrorism works. The early Muslims used terror repeatedly, in quelling dissent in Medina under Muhammad, and in displaying power to quell resistance in the early conquests, when they didn't have the manpower to fully occupy the countryside (Glubb gives vivid examples in the conquest of Egypt). Terror cheaply communicates raw power and extreme dedication to one's cause, regardless of legitimacy or aims. And people respect, if not worship, power, combining a natural ability to tell which way the wind blows with a true respect for such dedication and will power. As Reagan said, "Nothing succeeds like success."

Saturday, November 7, 2009

Gray matter talks

New work on how the brain computes speech production.

A recent issue of Science had a few interesting articles on neuroscience, including one on how Broca's area in the brain processes language. But another article on functional brain imaging brought up a factoid that makes for an interesting introduction:
"Questions about functional segregation are constrained by the resolution of fMRI. For example, a voxel (volume element of several mm cubed) contains on average 5.5 million neurons, 10e10 synapses, 22 km of dendrites, and 220 km of axons."
Wow- I didn't know the brain was quite so dense. So why aren't we smarter? Honestly, one gets the impression that the brain is not very efficiently designed. No wonder vast regions of the brain can be destroyed in dementia before much of a deficit is noticed.

The current work (accompanied by a review) has its origin in rare people who have epilepsy and who get the unnerving procedure of having electrodes stuck into their brains. I have no idea why such an invasive test is done, but for Sahin et al., it was a godsend, allowing them to do the kind of electrophysiology normally restricted to other animals such as monkeys, rats, and cats. Such work has told us huge amounts about the visual system and other brain circuits, but cats can't talk, so such work can't tell us much about that human ability.

Image of Broca's area and fMRI activation by a speech task, in patient A.

Broca's area and a few others are well-known to be involved in speech production, by way of strokes and other lesions that specifically affect those abilities. But how does it work? Is processing sequential, like the visual system, where a hierarchy of processing takes signals from raw retinal input to various color, edge, motion, and shape detection, up to object recognition? Or is language different, capable of being processed in a more parallel fashion, with all elements (word choice, grammar, phoneme production) coming together at once?

Patient A, with the surface of Broca's area exposed and electrode paths indicated.

This paper supports the former model, finding sub-locations in Broca's area that activate during specific stages of speech production, indicating that, while the full region involved is far smaller than the huge areas devoted to visual scene interpretation, it is also hierarchically arranged, with computations happening sequentially.

Electrophysiology like this is still a pretty quaint and brutal way to look at the brain. Electrodes, resembling dipsticks, are stuck right into the grey matter, with the hope that not too many cells are killed, no blood vessels are blown, and that whatever neurons are near the active surfaces of the electrode give off enough electrical buzz for detection (called a local field potential). Each electrode has several channels (i.e. openings) along its length, so you can listen in to several discrete depths once it is inserted.

Illustration of probe location, with channels indicated in yellow.

The experiment was to cue the patient with a fill-in-the-blank sentence, such as "Yesterday, they ____" plus a generic word to fill in, such as "to walk", or "to think". The patient had the task of computing the right form of the verb (walked, thought) and speaking it silently. (How speaking silently actually works in these experiments is a little hard to understand.) The observation was that reading the cue sentence correlated with a small activation near the Broca area electrodes, while producing the requisite word correlated with much larger activations.

Mix of electrode traces from a few channels. The sentence with blank is presented at the cue time, and the word to be filled in is presented at the target time zero milliseconds (ms). The colored arrows point to the segments discussed below at 200, 320, and 450 ms.

The second observation was that the activations related to speech production were complex, taking place over roughly 600 milliseconds (ms), with distinct peaks and troughs at 200, 320, and 450 ms after presentation of the fill-in word, depending on the channel and electrode location. The point of the experiment was to vary the fill-in words such that more or less complex processing demands could be correlated with more or less complex electrophysiology during these periods of proccessing in Broca's area.

The strategy is much like trying to figure out how a computer works by holding a few electrodes to a computer chip while it is working- a ludicrously difficult and primitive approach to reverse engineering. But it is all we've got for the moment, until functional imaging and non-invasive EEG technologies reach higher resolution. The observations are thus correspondingly crude- that the processing of incipient speech can be broken down into sequential phases (three in this paper) of word identification, inflection processing, and phonology processing. Here is the variation they observe in word identification:


Here, the brown curve around 200 ms is higher for rare words (no examples given) than for common words (but not for short vs long, or multi-syllable vs monosyllable), indicating that this activity is related to word identification.



These curves at 320 ms indicate variation in response to verbal inflection processing- past tenses, irregular forms, etc. "Read" is the control, with no fill-in work. The patient just reads the sentence. "Null-inflect" is when a fill-in is asked for, but the proper form happens to be the same as what is cued, so there is no phonological processing, only implicit grammatical/inflection processing (Every day they ____ [walk]). And "overt" is when the fill-in demands both grammatical processing and changes to the word (suffix or change in form). (Yesterday they ____ [walk]). The experiments were run repeatedly, with flashed cards and randomized orders, with the curves reflecting averages, and curve differences given P-values of 0.01 or less.


Curves at 450 ms (note the different channel (depth) used) correlate with processing for sound construction- related to number of syllables, changes in word form, etc.

The author's conclusions are that they have dissected some aspects of speech production based on where their electrodes penetrated the patient's brains and the timing of observed electrical events, coupled with experimental variation of tasks given to the patients. While most of this was surely known indirectly, based on the many patients with known defects of Broca's and other areas from strokes and other ailments, seeing this activity and its variations in real time is certainly unprecedented, and will contribute to ongoing refinement of the functional mapping of the brain, which, as noted above, has so very far yet to go.

As a stutterer, this is fascinating, and I hope that far more is learned about the nature of speech construction, to the point that the miswiring involved in stuttering might be diagnosed (if not fixed). And of course it also indicates yet again that our behaviors are not magical products of souls, but are computational products of brains.