Mystery Rays from Outer Space

Meddling with things mankind is not meant to understand. Also, pictures of my kids

January 19th, 2010

The good old days

Ladies & Gentlemen, I give you The Fever Districts of the United States, as of 1856 (click for a larger version):

Keith 1856 Fever Districts of the USA

Note the outlining of the Intermittent Fever districts, including Lansing, MI, where I live. Note the intense yellow rim of Yellow Fever.  Note the Small Pox Measles Scarlatina Consumption Endemic region along the Eastern seaboard, the large-case TYPHUS, the DYSENTERY, the casual “And many epidemics” tacked on to the main Yellow Fever, the serpentine red band tracing cholera. 1 There’s goitre in the Midwest and Mexico, elephantiasis down in South America, “Dia. & Dys. (severe)” in tiny writing down in the Bahamas, and the Bermudas are “generally healthy: Influenza, Rheumatism, Dysentery, Yellow Fever”.  And so much more.  (Compare to the map of Malaria in the USA, 1870.)

Keith 1856 USA Health & DiseaseThis amazing map is a mere afterthought, an inset of a map whose awesomeness goes up to 11.  The US map to the right2 (again, click for a larger version) is still just a small fraction of the whole, and that’s not even mentioning the jaw-dropping charts and graphs, also inset, showing “Consumption: Proportion of Deaths in the different quarters of the Globe”, “Comparative Value of Life in Different Countries”, “Proportionate Mortality of European Residents in Foreign Countries” and still more and more.

This map is “The geographical distribution of health & disease, in connection chiefly with natural phenomena. (with) Fever districts of United States & W. Indies, on an enlarged scale,” and it’s from:

The physical atlas of natural phenomena
by Alexander Keith Johnston, F.R.S.E., F.R.G.S., F.G.S.
William Blackwood and Sons, Edinburgh and London, MDCCCLVI 3

I’d run across reverent mentions of this map — especially the Fever Districts inset — here and there in old books, and I just stumbled across it in downloadable form.  You must go at once to The David Rumsey Collection and pore over it for several hours, at the highest resolution.


  1. Lansing seems to have been just barely cholera-free, at least in 1856.[]
  2. The colors refer to “zones of disease” – Torrid (brown), Sub-torrid & temperate (green), sub-temperate & arctic (blue) []
  3. That’s 1856, for those of you who, like me, need to pause a while in thought when confronted with years in Roman numerals[]
December 8th, 2009

Malaria and mosquitoes: Not 1908, not Cuba

A couple of days ago I posted this map of malaria in the USA. It got picked up by Grant Jacobs, who made some interesting and useful comments, and that in turn got picked up by someone who posted it to boingboing.net.  Unfortunately, whoever wrote it up for boingboing tried to add some value by offering a couple of points on the history of malaria, both of which were wrong. 1 In particular, he claimed that “It wasn’t until 1908 that a Cuban doctor made the connection with mosquitoes”.  To set the record straight:

RECENT researches by Surgeon Major Ronald Ross have shown that the mosquito may be the host of parasites of the type of that which causes human malaria. Ross has distinctly proved that malaria can be acquired by the bite of a mosquito, and the results of his observations have a direct bearing on the propagation of the disease in man. Dr P. Manson describes the investigations in a paper in the British Medical Journal, and sums them up as follows: –The observation tend to the conclusion that the malaria parasite is for the most part a parasite of insects; that it is only an accidental and occasional visitor to man; that not all mosquitos are capable of subserving it; that particular species of malaria parasites demand particular species of mosquitos; that in this circumstance we have at least a partial explanation of the apparent vagaries of the distribution of the varieties of malaria. When the whole story has been completed, as it surely will be at no distant date, in virtue of the new knowledge thus acquired we shall be able to indicate a prophylaxis for malaria of a practical character, and one which may enable the European to live in climates now rendered deadly by this pest.

Nature, Sept. 1898.  p. 523

The earliest probable reference I can find2 is from 1896:
The Goulstonian Lectures on the Life History of the Malaria Germ Outside the Human Body. P. Manson. The British Medical Journal, 1896

Update: I just realized what the boingboing.net poster had in mind with his comment that “It wasn’t until 1908 that a Cuban doctor made the connection with mosquitoes”: He was thinking about yellow fever, a virus rather than a parasite. Here and there about the web it’s suggested that yellow fever was shown to be mosquito-borne, in 1908, by a Cuban doctor, Carlos Finlay.  Unfortunately that’s also not correct; it probably was originally a typo somewhere that got spread around.

Finlay (who was, I believe, American, though he worked in Cuba) originally published his observations in 18813 and then in English in 18891886.4  His theory wasn’t immediately accepted, but by 1900 it was confirmed by a medical commission that included the famous Walter Reed.


  1. Also, he didn’t credit me, which is probably for the best, since my pathetic hosting would have undoubtedly crashed[]
  2. I haven’t read the text of this yet[]
  3. C. Finlay. El mosquito hipoteticamente considerado como agente de trasmislon de la flebre amarllla. An. de la Real Academia de ciencias med. … de la Habana, vol. 18, pp. 147-169 (Aug 14 1881) []
  4. C. Finlay. Yellow Fever, its transmission by means of the Culex mosquito. Am. Journ. Med. Sci. vol. 92, pp. 395-409 (1886) []
December 3rd, 2009

Gone in 60 (milli)seconds

Gone in 60 (milli)seconds

Intracellular proteins have to be degraded, more or less at the same rate as new proteins are produced (or the cell would eventually burst). On the other hand, you can’t go about degrading proteins willy-nilly.  There are vast and complex systems for identifying proteins that should be destroyed, tagging them, and then moving them into a controlled destruction chamber.

The most important of these systems is the ubiquitin-proteasome degradation pathway.  Proteins that are destined for destruction are tagged with a chain of ubiquitin molecules.1  There are multiple steps in this pathway, in which ubiquitin is prepared for tagging, target proteins are identified, and ubiquitin is transferred from the activating components to the targeted protein.

Target proteins are destroyed when a chain of ubiquitin molecules (head to tail) are attached to them. An unanswered question has been how this works. Is the ubiquitin chain formed first, and then transferred to the target en bloc? Or are single ubiquitin transferred one at a time, sequentially, first to the target protein and then to the previously-attached ubiquitins?  The problem has been that the process goes so fast that it’s been hard to distinguish between the possibilities.

Now, in a gorgeous series of experiments, Pierce et al2 were able to watch ubiquitination happening over fractions of a second:

… we performed our single-encounter reactions on a quench flow apparatus that allowed us to take measurements on a timescale ranging from 10 ms to 30 s2

And the answer looks pretty clear: Ubiquitins are transferred sequentially, not en bloc.

Even at this timescale, though, they weren’t able to catch the very first event — the transfer of the first ubiquitin to the target.  That happens, apparently, in less than 10-20 milliseconds.  They also draw the conclusion that target tagging is critically dependent on the kinetics of ubiquitin chain elongation (as you’d expect) which are governed by ubiquitin off-rates, and this mode of regulation is probably a billion years old.

Pierce et al (2009) Fig. 3d: Ubiquitin addtion

Figure 3d: Kinetics of ubiquitin addition and elongation2(Click for a larger version)


  1. Ubiquitin being a small, abundant protein[]
  2. Pierce, N., Kleiger, G., Shan, S., & Deshaies, R. (2009). Detection of sequential polyubiquitylation on a millisecond timescale Nature, 462 (7273), 615-619 DOI: 10.1038/nature08595[][][]
November 9th, 2009

Making charts with Numbers

Apple’s iWorks ’06 package was interesting, but ended up being too simplified to really compete with MS Office.  But iWorks ’09 was a big step forward, and I now use Pages for almost all my word processing, and Numbers for about 75% of my spreadsheets.  (I still use Powerpoint for most of my slideshows; I don’t find any compelling reason to use Keynote instead, and Powerpoint does have some distinct advantages.)

“Numbers” looks fairly similar to Excel — they are both spreadsheet programs, so there’s only so many ways of usefully presenting information there — but the editing and so on can be quite different from Excel, which can be frustrating if you’re coming from an Excel background.  Rosie Redfield was just complaining about the non-intuitiveness of Numbers.  I don’t think it’s non-intuitive, just different from Excel.

So I put together a couple quick screencasts of making a line graph and a scatter plot, in the hope it would give a starting point for people new to Numbers.  (Flash movies, 7.8 and 5 MB respectively.  No sound, because my kids are still asleep.)  I’ve never tried this before, but hopefully they’ll work.

October 31st, 2009
August 6th, 2009

On stupidity and virologists

I’ve quoted this before, but without attribution, because I didn’t know who originally said it:

”The stupidest virus is smarter than the smartest virologist.”

Apparently it was George Klein.

July 25th, 2009

Busy, Back soon

I’m off for a week’s vacation with the family – camping, etc., and with limited internet access.  I’ll be back in the first week of August some time.  Talk amongst yourselves.

July 1st, 2009

Happy Canada Day

(A real post tomorrow, I hope)

June 15th, 2009

Conspicuous consumption

CTuberculosis and the Grim Reaper
Tuberculosis and the Grim Reaper

A while ago I made the point that many of the biggest killers of 19th-century London were almost unknown today, because of vaccination (“hooping cough”, measles, smallpox) and sanitation (typhus, cholera) (see “Life & Death, pre-vaccination“).

I have a small confession to make: I kind of rigged that chart, because I wanted to avoid a complicated story that I didn’t know much about.  I still don’t know much about it, but I’ll share my ignorance with you, dear readers, because it’s a fascinating story and because it hooks up with something else I want to talk about, maybe later this week.

The “rigging” was pretty minor.  If you look at the table I took the mortality info from (from the Journal of the Statistical Society of Landon, Vol. XII, 1850)  you’ll see the infectious causes that I listed, neatly clustered together at the bottom.  In 1847, mortality from these diseases looked something like this:

Mortality in London, 1847

What’s missing?  The biggest killer of them all;1 it’s not included in this section because before 1882,2  they didn’t know tuberculosis (“consumption”) is an infectious disease.

Here’s what happens when we include “tubercular diseases” in the same chart — watch the scale!

Mortality in London, 1847

In developed countries most of us don’t think about tuberculosis much, but in the 19th century it was everywhere.  The poor — crowded and malnourished — were at the most risk, but consumption spared no one.  Rich and poor, merchant or noble or laborer, everyone had friends and family who died of consumption.

What happened to it?  Why did Tb go from causing 20% of all deaths, to only infecting 0.01% of the population (and killing a far smaller fraction)?  Unlike the other infectious diseases, vaccination and sanitation can only explain a part of that.  The death rate of Tb was already dropping drastically well before 1882, when Koch showed that it was infectious:3

Trends in Tb mortality

From “Pulmonary Tuberculosis
Maurice Fishberg (Lea and Febiger, Philadelphia, 1922)

So although antibiotics and, to some extent, vaccination were to help push Tb to obscurity in the 20th century, the disease was already, very slowly, fading before that.  (Tb rates had exploded in the 18th century, as urbanization crowded the poor together.  It wasn’t for many years that cities became self-sustaining and didn’t reply on immigration.)

Consumption (19th-century physician)
From “Passages from the diary of a late physician
Samuel Warren (Baudry’s European Library, 1838)

Why was Tb becoming less common? Well, this is the part I don’t actually understand very well, but according to Arthur Newsholme4 in 19085 this was indirectly because of the Poor Law of 1834 (Wikipedia on Poor Laws).  The Poor Laws were very, very primitive versions of welfare; the 1834 Act brought about a system of workhouses, where the desperately poor — and they had to be utterly desperate — were fed (barely) and housed (kind of ) and generally abused.  The point being, the poorest of the poor were kept in these workhouses; because the Tb sufferers, who couldn’t work normally, tended to be the poorest of the poor, they were housed in the workhouses and essentially quarantined.  After Koch demonstrated that Tb is contagious, in 1882, quarantine became a deliberate policy6 and rates dropped still more; and when antibiotics were introduced in the 1940s, rates of Tb dropped still more.

So although I said the biggest killers of the 19th century have been almost eliminated by sanitation and vaccination, that’s not really true of tuberculosis.  Antibiotics broke the back of the disease, but it was already being controlled to a large extent by social factors and then by medical opinion — one of the few cases where formal medicine actually had an influence on these diseases.


  1. Well, “Diseases of the lungs and other Organs of respiration” killed more people, but that’s not a single disease[]
  2. Koch, R. 1882. Die Aetiologie der Tuberculose. Berl. Klin. Wchnschr., xix: 221-230.[]
  3. And it wasn’t for years after that that an effective treatment or vaccine was available[]
  4. The Prevention of Tuberculosis, by Sir Arthur Newsholme.
    Methuen, 1908 []
  5. Supported by others since, e.g.  Wilson, L. (2004). Commentary: Medicine, population, and tuberculosis International Journal of Epidemiology, 34 (3), 521-524 DOI: 10.1093/ije/dyh196[]
  6. I believe Sir Arthur Newsholme was important in instituting this in Britain[]
June 2nd, 2009

“Mus homunculus” in the lab?

Researchers have used the mouse extensively as a model organism to study the pathogenesis of human infections and found that it imperfectly recapitulates many aspects of infectious disease as seen in patients. 1

Mickey mouse evolution
Humanizing a mouse

That strikes a chord with me because I just sent off a grant application explaining that mice are not suitable models for viral immune evasion.  However, my application may show a failure of imagination (or courage), because what Coers et al.1 are driving toward is humanizing mice to make them better models for human disease, whereas I am merely proposing a different animal model.

What causes species specificity in pathogens?  That is, why is it that many pathogens infect humans very nicely, but don’t infect mice to any extent?  (And, of course, conversely, why do other pathogens cause disease in mice and not in humans.)

Chlamydia trachomatis in human cells
Chlamydia trachomatis in human cells

In some cases, a viral pathogen may simply be unable to get into the appropriate cell in the wrong species. An example is poliovirus, which normally doesn’t infect mice at all. But if you make a transgenic mouse2 that expresses the (human) poliovirus receptor3, then the virus infects mice, and causes disease in them, perfectly well.   In this case, the receptor is the critical determinant of species specificity.  As a natural example of the same concept, SARS virus at least partly adapted to infecting humans by modifying its receptor-binding protein4 to improve interaction with the human version of the protein.

But there are also lots of cases where the virus can get into cells from the other species, yet doesn’t manage to replicate well or cause disease.  I’ve talked about mouse cytomegalovirus (MCMV) and its inability to infect humans here; it turned out that MCMV can’t infect human cells well because its normal ability to disarm the programmed cell death (apoptosis) pathways only works against the mouse versions of the pathway.  There are similar stories with HIV and its primate-infecting cousins; these viruses are limited to infecting hosts in which they (the viruses) can eliminate the APOBEC retrovirus-destroying proteins.  And the poxvirus myxomavirus is at least partly restricted to infecting rabbits because it can only inactivate the interferon pathway in rabbit cells. 5

Mouse manYou may notice two things about these examples: First, the non-receptor examples are generally immune evasion stories.  That is, these viruses are often apparently restricted to infecting a limited number of species because their immune evasion arsenal is limited to those species; take away their immune evasion by putting them in the wrong species, and they’re enfeebled.  Second, these examples are viruses.  The reason for that is just that I’m used to dealing with the crisp, clean mountain air of virology, and I don’t usually descend into the fetid swamps of bacteriology.6

But it turns out that at least in some cases the principles seem to be the same.  The Coers et al. paper1 I cited at the top here makes some very familiar points: The receptor half of the story (“Colonization often relies on species-specific interactions of microbial ligands with host cell receptors“) applies to some bacterial pathogens (“Transgenic mice expressing human E-caherin in the small intestine, on the other hand, are susceptible to oral infections with L. monocytogenes and develop enteropathogenicity and systemic infections“).  And the immune evasion half also applies to some bacterial pathogens (“Additionally, host restriction may be caused by the failure of pathogens to deter immune assaults in the non-typical host“).

Even the nature of the immune evasion targets is familiar. Interferon pathways are frequent targets of bacterial immune evasion, as they are of viral immune evasion.  The details are different, in that the instances Coers et al. describe target a different branch of the interferon induction pathway, but the pattern is the same:

… the mouse-adapted strain Chlamydia muridarum, but not its close relative C. trachomatis, can specifically evade IRG-mediated7 host resistance … The divergent counterimmune mechanisms employed by the human pathogen C. trachomatis and the mouse-adapted pathogen C. muridarum clearly reflect the differences in the IFN? responses of their respective hosts. 1

They finally discuss the possibilities of “Mus homunculus”, humanized mice, tailored to each pathogen, that would make more authentic models of infectious disease.  “Though the creation of humanized mouse models for infectious disease will require substantial effort and resources, the long-term benefits of these new models would undoubtedly be enormous.1


  1. Coers, J., Starnbach, M., & Howard, J. (2009). Modeling Infectious Disease in Mice: Co-Adaptation and the Role of Host-Specific IFN? Responses PLoS Pathogens, 5 (5) DOI: 10.1371/journal.ppat.1000333[][][][][]
  2. Hi, Vincent![]
  3. Transgenic mice expressing a human poliovirus receptor: A new model for poliomyelitis.
    Ruibao Rena, Frank Costantinib, Edward J. Gorgaczc, James J. Leeb and Vincent R. Racanielloa
    Cell 63:353-362 (1990) []
  4. Li W, Zhang C, Sui J, Kuhn JH, Moore MJ, et al. (April 20, 2005) Receptor and viral determinants of SARS-coronavirus adaptation to human ACE2. EMBO J 24.:1634-43.
    Sheahan T, Rockx B, Donaldson E, Sims A, Pickles R, et al. (March 2008) Mechanisms of zoonotic severe acute respiratory syndrome coronavirus host range expansion in human airway epithelium. J Virol 82.:2274-85.[]
  5. Wang F , Ma Y , Barrett JW , Gao X , Loh J , Barton E , Virgin HW , McFadden G (2004) Disruption of Erk-dependent type I interferon induction breaks the myxoma virus species barrier. Nat Immunol 5: 1266-1274[]
  6. In other words, I don’t know much about bacteriology.[]
  7. IRG is part of an interferon-induction pathway[]