Musings May 2012 to ?? (current posts)

   Introduction (separate page).
This page:
Current posts -- 2012 (May -??)
   New items (Posted since current e-mail; they will be announced in next e-mail, but feel free... !
      * If you would like to get an e-mail announcement of the new posts each week, see the Contact information at the top or bottom of this page.)
   May 16 (Current e-mail)
   May 9    May 2

      Older items are on the archive page.

New items

Posted since current e-mail; they will be announced in next e-mail, but feel free...

A mouse carrying a serotonin-transport gene that contributes to human autism

May 18, 2012

Autism is a major disease, which affects brain function. Little is understood about its underlying basis. It is likely that there is a genetic component to autism. Several genes show some association with autism, but there is no simple one gene-one disease relationship. Autism (or, more broadly, autism-spectrum disorders) is undoubtedly heterogeneous -- a family of diseases with various causes contributing.

An old finding is that some people with autism show alterations in metabolism of the neurotransmitter serotonin. Among the genes associated with autism is one that affects serotonin transport -- called SERT. In a current paper, a team of scientists has made a mouse carrying a particular SERT mutation that has been found in some cases of autism. The results are intriguing.

They introduce a mutation in the mouse SERT gene, corresponding to a mutation found in some human autism. The mutation is G56A, which means that the amino acid G (glycine, or Gly) at position 56 in the protein chain has been replaced by A (alanine, or Ala). They do various tests, biochemical and behavioral, comparing the wild type (Gly56) and the mutant (Ala56).

The two frames here show examples of their results for behavioral tests.

In part E (left), they measure the vocalizations when a mouse pup is removed from its mother (7 days after birth). You can see that the mutant mice (black bar, right) make fewer vocalizations. (The y-axis shows the number of vocalizations observed during a defined test period, which is 5 minutes. The * indicates that the result tests as statistically significant.)

In part F (right), they use the "three-chamber Crawley sociability test". In this test, one chamber (left side, as labeled on the graph) has a novel inanimate object and one chamber (right) has a novel mouse. The central chamber of the test box is a "neutral" reference point -- and the common entry to both boxes of novelty. They measure the amount of time a test mouse spends in each chamber. The results are that the wild type mice spend much more time checking out the novel mouse than the novel "object". In contrast, the mouse with the altered SERT gene spends about the same amount of time with each -- actually about the same time in each of the three chambers..

These are parts of Figure 2 from the article.

Both of the behavioral tests above are considered tests of sociability. In both cases, the mice with the mutant SERT gene are less sociable. Other tests they do are in agreement. That is, a mutation that is associated with some cases of autism in humans seems to result in some autism-like behavioral changes in mice.

What are we to make of this? Let's be cautious. It is an interesting experimental finding in a model system. Let's not draw any big conclusions beyond that for now. Nevertheless, one wonders whether the shortage of brain serotonin caused by the mutation studied here is somehow part of the autism disease process, at least in some cases. (The particular mutation studied here leads to increased serotonin in the blood, but decreased serotonin in the brain.) As so often, further work is needed.

News story: Novel Mouse Model for Autism Yields Clues to a 50-Year-Old Mystery. (ScienceDaily, March 20, 2012.)

The article: Autism gene variant causes hyperserotonemia, serotonin receptor hypersensitivity, social impairment and repetitive behavior. (J Veenstra-VanderWeele et al, PNAS 109:5469, April 3, 2012.)

More on autism: Early detection of autism (June 14, 2011).

May 16, 2012 (Current e-mail)

Baseball and violins

A recent post was about an unusual way to make a violin: Spiders and violins (May 4, 2012). It reminded me of an earlier news story about another unusual violin. I think I distributed it privately to some of you, but it now seems worth noting it in Musings for the record.

The Iceman's blood

May 14, 2012

A red blood cell. 5300 years old. It is from Oetzi (or Ötzi), the Iceman. The image here is by atomic force microscopy.

This is Figure 1e from the paper. The full figure compares Oetzi's red blood cells with those of modern humans, observed by the same methods.

What's this all about? Oetzi is the man whose frozen body was discovered in the Alps in 1991. The body was dated to about 5300 years ago; it is remarkably well preserved. Study of Oetzi has become quite an active field.

Much is now known about Oetzi's diet -- and about how he died. He now stands as the oldest known murder victim. Oetzi's genome was recently sequenced. And here we have an examination of his red blood cells, found near the wound that presumably led to his death. They are the oldest known human blood cells -- by about 3000 years. They look remarkably like modern blood cells. In a sense, that is no surprise, but it is a thrill to see them.

Further examination of the blood cells provided evidence for hemoglobin -- and for the clotting protein fibrin.

Oetzi fascinates us. It was an accident that he died under circumstances that promoted good preservation -- as generally true with fossils. Now we have Oetzi, and we have modern technologies to study him. Oetzi is our window into an ancient era of humankind.

News story: Iceman Mummy: 5,000-Year-Old Red Blood Cells Discovered -- Oldest Blood Known to Modern Science. (ScienceDaily, May 2, 2012.)

The article, which is freely available: Preservation of 5300 year old red blood cells in the Iceman. (M Janko et al, J. R. Soc. Interface, in press.)

More about ancient humans...
* Inuk, a 4000 year old Saqqaq from Qeqertasussuk (March 1, 2010).
* The Siberian finger: a new human species? -- A follow-up in the story of Denisovan man (January 14, 2011).

More about ancient blood and ancient proteins...
* Mammoth hemoglobin (February 1, 2011).
* Dinosaur proteins (July 6, 2009).

Can French baboons learn to read English?

A strange story. I must admit that I am still not sure what is going on here. Let's have a look at what they did, and what they claim. It's especially important with something like this to be sure we look at what the article actually says, and minimize being biased by news coverage or headlines (including mine!). Keep an open mind.

Here is the basic experimental design. Baboons have a computer console. They are presented with a four letter item -- and asked to tell whether it is a "word" or a "non-word". If they are right, they get a food reward. It's a fairly straightforward and common type of experiment. You can see it in action in the video listed below; the video is worthwhile. The baboons are first trained on known words and non-words, and then tested on items they have not seen so far. They do rather well, making the right choice 75% of the time.

In one sense, this is impressive. The baboons are doing something -- and that is interesting. But what is it they are doing? I find the use of the term "word" confusing, perhaps even distracting. Above we noted the example of "beng" as a non-word. I agree it is a non-word -- but I don't really see why. It follows the general pattern of words. If someone invented a new word, beng, no one would object that it just doesn't seem like a word.

Now, the authors are clear to define what they mean by a word, in terms of bigrams. Is that a sufficient answer to my confusion? The baboons are recognizing patterns -- patterns that involve letters. Let's focus on patterns, not "words". Perhaps, but I sense that the authors think that what they are studying is relevant to the development of reading skills. They present some argument for this. As a non-expert in their field, I cannot really judge it.

No matter. They have an interesting experimental system. They will be pursuing it. As we get more information, we can all judge what the significance is.

Where is the dark matter?

May 11, 2012

Dark matter. Scientists estimate that most -- over 80% -- of the matter in the universe is invisible by our ordinary observational methods. We call it dark matter. If we can't see it, how do we know it is there? And what is it?

The first question is fairly straightforward. It is matter. It has mass -- and thus follows the law of gravity. If we observe a system of moving bodies, such as a galaxy, it must follow the law of gravity. From the observed motions, we can infer how much mass there must be. And, for large astronomical collections, that amount is typically far more than what we can account for by observation. To explain this discrepancy between the amount of mass inferred from the law of gravity and the amount observed, we invoke "dark matter" -- something that has mass but is not observed. This discrepancy, and hence the postulate of dark matter, dates back to observations by Fritz Zwicky in 1933; numerous observations support the general idea. As to the second question... the nature of dark matter remains unknown.

Now a team has estimated the amount of dark matter in a region of space near our Sun. Their basic approach is as before: calculate what gravity requires, and calculate what they can see. What is really new here is the thoroughness of their observations. The striking finding is that there is no finding. They don't find dark matter in this region. That is, the amount of mass required by gravity is fully accounted for by what they can see. Models of how our galaxy formed make predictions about the amount of dark matter that should be in this region; they don't see it.

What is the significance of this new finding (or non-finding)? Who knows. It is part of the continuing mystery of dark matter. Perhaps there is something wrong with this analysis. Perhaps the distribution of dark matter in our galaxy is different from what we expected. Or ??? Dark matter was originally postulated to address a discrepancy between theory and observation. This is another piece of the puzzle -- not one that leads to a solution, but one that must be addressed at some point. Their own final sentence in the paper is "Indeed, we believe that our results do not solve any problem, but pose important, new ones."

There is an interesting consequence of their finding... If it is really true that our local region is short of dark matter, that will hamper efforts to examine the nature of dark matter by earth-bound experiments. It's hard to find something that is invisible. It's even harder if it is not there.

What they do in the new work is to set an upper limit for how much dark matter is in this region of space. They give that limit as a density. They find that the amount of dark matter is less than 1 milli-solar-masses per cubic parsec. (The expected value is 5-13, in the same units.)

News stories:
* Serious Blow to Dark Matter Theories? New Study Finds Mysterious Lack of Dark Matter in Sun's Neighborhood. (ScienceDaily, April 18, 2012.)
* Has Dark Matter Gone Missing? (Science Now, April 19, 2012.) This story includes more discussion of alternative interpretations of the significance of the new work.

The article... Kinematical and chemical vertical structure of the Galactic thick disk II. A lack of dark matter in the solar neighborhood. (C Moni Bidin et al, Astrophysical Journal 751:30, May 20, 2012.) A preprint is freely available at the arXiv: copy at ArXiv.

More about gravity: The potato we call home: a study of the earth's gravity (May 3, 2011).

More about galaxies: LEDA 074886 (April 2, 2012).

May 9, 2012

Should you get a rabies vaccination before boarding an airliner?

This incident ended without harm, and it is easy to look on this story as odd. Yet the issues are important. What if ... ??? Remember, they still don't know the complete list of passengers. (And the little thing tied up one of the bathrooms for over half the flight!)

The news story is plenty, but if you do want more... (As usual, I try to include a scientific article, so you know the story is grounded in serious work. But most posts do not depend on reading the article.) The article, which is freely available: Rabies Risk Assessment of Exposures to a Bat on a Commercial Airliner -- United States, August 2011. (J Kazmierczak et al, Morbidity and Mortality Weekly Report (MMWR) 61:242, April 13, 2012.) As usual with MMWR, the main part of the article is followed by an "Editorial note", which is a plain-language overview, and puts the item in perspective. The Editorial Note can be a good place to start.

Fossil raindrops and the density of the ancient atmosphere

May 6, 2012

Fun. A recent article reports measurements of fossil raindrops -- more specifically, of the little "craters" left when a raindrop hits the ground. From the sizes of these fossil raindrops -- or fossil "rain-prints", if you like -- they make inferences about the density of the atmosphere when those raindrops fell -- 2.7 billion years ago.

There is a reason for wanting to know the density of the ancient atmosphere. In those days, the Sun was less bright than it is now, yet Earth temperature was not much different than now. Assuming those two statements really are true, there is a gap in our understanding. One way we could have a dim Sun and a warm Earth would be for there to be high levels of greenhouse gases. Some have suggested that the atmosphere back then might have had twice the density, or more, than our modern atmosphere. So, examining the ancient atmospheric density is of interest. And their approach is fun -- and logical. Whether it gives a good result is hard to tell, but let's look at the idea.

The heart of their work is finding geological samples that have rain-prints. They measure the sizes of the individual prints, and from a calibration curve determine the density of the air the drops must have fallen through. In some ways, that is fairly straightforward: rain drops fall far enough that they all reach a terminal velocity that is rather simply related to the density of the atmosphere.

Here is their calibration curve for size of rain-print versus the properties of the raindrop. They made this calibration curve by dropping water drops of known size onto a surface of the appropriate soil material, and measuring the size of the imprint. The y-axis is the size of the imprint -- the "raindrop crater area", in square millimeters. The x-axis is a measure of the momentum of the drop. Basic momentum is mass x velocity; they make it a bit more complex here, but it doesn't matter for now.

This is Figure 2 of the article.

As you can see there is a simple relationship between the momentum of the drop and the imprint it leaves. So, they can measure the size of a fossil rain-print and use this calibration curve to estimate the momentum of the raindrop that created it. Knowing the momentum gives them the velocity of the drop (but see next paragraph); knowing the velocity gives them the density of the air the drop fell through.

Simple. Logical. And cute. But it isn't quite that simple. The problem is that the momentum involves both mass and volume. For their calibration tests, they use drops of known size; for the fossil drops, they have to make some assumptions. Of course, the rain-print size also depends on the type of soil the rain hits. Again, they make assumptions. With all these assumptions, it is not clear how useful the result is. Nevertheless, the approach is interesting; perhaps such tests can be done with other geological samples of rain-prints, and over time a clearer picture may emerge.

For the record... With their assumptions, they estimate that the density of the atmosphere 2.7 billion years ago was not much different than ours -- perhaps even a little less. If this is valid, it argues against a thick atmosphere with lots of CO₂.

My comments about the assumptions in the paper and therefore its limitations are not intended as "criticism". The paper presents clearly what they did, including the assumptions I note here. They are quite clear about how their conclusions depend on the assumptions. This paper makes a contribution: it introduces a new type of measurement and provides some example of its use. That's a useful step. Single scientific papers typically do not make huge steps. Over time, we integrate what we learn from a wide variety of work. Looking at single papers lets us see science in progress. And this paper is fun and readable.

There is another neat part to this story. As background, they state, "The idea of using raindrop imprints as a proxy for air density was suggested by Lyell¹⁴ in 1851 but has hitherto been unexplored." So we have a current paper that points to a mid-19th century paper by one of the founders of modern geology. We'll look at what Lyell said in that reference #14 in the accompanying post, below.

News stories:
* Fossil Raindrop Impressions Imply Greenhouse Gases Loaded Early Atmosphere. (ScienceDaily, March 28, 2012.)
* Primeval Precipitation: What Fossil Imprints of Rain Reveal about Early Earth. (Scientific American, March 28, 2012.)
* Ancient Raindrops provide Insight into our Early Atmosphere. (Naked Scientists, April 2012.) An interview with one of the authors. Good overview of the work.

* News story accompanying the article: Geoscience: Fossil raindrops and ancient air. (W S Cassata & P R Renne, Nature 484:322, April 19, 2012.)
* The article: Air density 2.7 billion years ago limited to less than twice modern levels by fossil raindrop imprints. (S M Som et al, Nature 484:359, April 19, 2012.)

As noted, the post immediately following, Lyell on fossil rain-prints (May 6, 2012), is closely related. That post also includes crosslinks to related Musings posts -- related to both this post and that one.

Lyell on fossil rain-prints

This post is closely related to the one immediately above. That post refers to the paper listed here as an early example of the basic logic they use. The paper here stands on its own as a historic paper -- one that is very readable and enjoyable.

In this 1851 paper, Sir Charles Lyell examined and compared modern and ancient rain-prints.

One of the issues Lyell addressed was whether the cavities he observed might be due to air bubbles that had been trapped in mud, and then burst. He discussed this with "Mr. Faraday" and also did some experimental work. He concludes that rain-prints and burst air bubbles are easily distinguished.

In the final paragraph, Lyell notes "... it is satisfactory to obtain positive proofs of showers of rain, the drops of which resembled in their average size those which now fall from the clouds. From such data we may presume that the atmosphere of one of the remotest periods known in geology corresponded in density with that now investing the globe, ...". This is the tie to the new paper, in the accompanying post (directly above). The new paper reaches back further in time. Much further.

Spiders and violins

May 4, 2012

A Japanese physicist-violinist, Dr Shigeyoshi Osaki of Nara Medical University, has announced a new development in making violins. He employed a group of about 300 collaborators to help develop a new material for violin strings.

A member of the team.

This is from the BBC news story.
(I do not know that the specific individual shown here participated in the work.)

He collected the silk from spider webs, and twisted the silk filaments into long strings -- violin strings. He then studied the physical and musical qualities of the strings. His basic conclusion was that the spider-silk strings gave the violin a novel tone, one judged by some to be richer. The following analysis shows why.

The figure shows frequency spectra for spider and steel strings (left and right, respectively). The first peak is the fundamental tone, 293 Hz. You can see that the sound produced with the spider string (left) was richer in overtones.

This is Figure 3 parts a and b from the article. The full figure also includes the spectrum for gut strings.

Overall, this work provides an interesting combination of biology, physics and music. Will anything useful come of it? Who knows; new developments in silks (see below) will provide new options to try. In any case, it is a fun exploration.

News stories:
* Spider silk spun into violin strings. (BBC, March 4, 2012.)
* Spider silk spun into violin strings. (New Scientist, March 5, 2012.) Includes a short video comparing the sound of the violin with various types of strings.

Additional video: YouTube: Spider Silk Violin Strings. Audio only (but the background figure is cute). A one minute sample from the violin with strings made of spider silk.

The article: Spider Silk Violin Strings with a Unique Packing Structure Generate a Soft and Profound Timbre. (S Osaki, Physical Review Letters 108:154301, April 13, 2012.) The paper includes more about the physics of the strings, including the deformation caused by the twisting.

More on violins:
* The smallest violin (November 16, 2009).
* Added May 15, 2012. Baseball and violins (May 15, 2012).

Previous post on spider silk: Spider silk: Can you teach an old silkworm new tricks? -- Update (February 11, 2012). This could be directly relevant to the current post, by supplying a range of silks.

May 2, 2012

Blueprint of a seaweed (1843)

The blueprint process, as with other "early" photographic processes, depends on photosensitive chemicals. In this case, iron salts are used. Exposure of the iron-impregnated paper to sunlight causes it to turn blue -- except where the object is, protecting that area from the light. Thus, in this case, the area where the seaweed lies remains white.

The blueprint process was invented by John Herschel, in 1842. His friend Anna Atkins soon began to experiment with the process for making pictures of biological materials. The picture above, and the book it is from, are the results.

Why did the HIV vaccine work for some people? Follow-up

May 1, 2012

In 2009 we were intrigued by the results of a trial of an HIV vaccine. The vaccine gave about 30% protection against infection. That's low, but it is the best found for an HIV vaccine. In 2011 we noted a preliminary announcement that scientists had found differences between people who were protected and people who were not protected. (Links to the earlier posts are at the end.) We now have some details of that story, and it is indeed an interesting story.

It is important to emphasize that the claims here are more hypotheses than answers. They are based on statistical analyses of small data sets. The analyses suggest certain things; some of these things may make some sense, but they need to be tested.

Here is an example of one of their analyses.

The type of graph here is one that is common in dealing with diseases or treatments. The y-axis shows the incidence -- or probability. In this case, it is the probability of acquiring HIV -- on a scale of 0 to 1. The x-axis is time, shown here relative to a particular step in the vaccine trial.

The graph is "double". There is the main graph, with axis labels along side and bottom. And then there is an inner graph, often called an inset. This type of graph is used to allow one to see both the "big picture" (main graph) and a detail (inset). In this case, there is a special reason for doing that.

Start by looking at the main (big, outer) graph. Do you find the data points? They are all along the very bottom. The incidence of HIV in the trial is so low that it is almost invisible on a normal scale. Thus the main graph here actually shows no useful information -- and that is their point.

So, to the inset, with an expanded scale. The y-axis, showing the incidence, now runs from 0 to 0.008 -- still less than 1%. Now you can see curves -- four of them. There is a key to the curves across the top. Start with the black curve; this shows when HIV cases were first diagnosed for the placebo (unvaccinated group; the "control" or "reference"). Then, there are three curves for parts of the vaccinated group. They are for those who had low, medium, or high response to the vaccine -- as judged by one particular immune system criterion. You can see that one curve shows a distinctly lower incidence of HIV; check the legend and you will see that curve is for those with a high response. (If you check further, you will see that the three curves for the vaccinated group are actually in order -- low, medium, high -- but the difference between the first two is small.)

This is Figure 3A from the article.

Thus we see, from the graph above, that there is a correlation between a particular type of immune response and the apparent effectiveness of the vaccine. Higher response is correlated with less HIV infection. This may seem "logical" to you; caution, we have not said what the particular response is, and have not discussed the other responses. We do not know why this occurs. In particular, we do not know that the particular immune response measured here is the actual cause of the reduced infection rate. What we have is a correlation -- and that allows them to generate hypotheses that can be tested further.

The particular immune response tested here is labeled "IgG Antibodies Binding to V1V2". IgG is the major class of antibodies. V1 and V2 are two particular regions of the protein on the viral coat. It is intriguing that antibodies against this region might be particularly effective.

In the paper, they explore several immune responses, and test six of them in the same type of detail as shown above. One shows a reverse effect: more immune response is correlated with more HIV infection. That's "bad" -- backwards from what one wants. They suggest this might be due to a particular antibody interfering with the effectiveness of good antibodies. Four of the immune responses they measured show no effect. Thus it does seem that the response presented in the figure above is especially interesting at this point. Why do some people show more of this response? Is it because of innate differences between the people. Is it "random", due to how the immune system works? In either case, it is possible to develop a vaccine that would provide more of this response? Questions. for the future.

News stories:
* Possible clues found to why HIV vaccine showed modest protection. (Medical Xpress, April 4, 2012.)
* RV144 HIV Vaccine Trial Give Clues About Protection from HIV. (ScienceDaily, April 4, 2012.)

* Editorial accompanying the article: The Road to an Effective HIV Vaccine. (L R Baden & R Dolin, New England Journal of Medicine 366:1343, April 5, 2012.)
* The article: Immune-Correlates Analysis of an HIV-1 Vaccine Efficacy Trial. (B F Haynes et al, New England Journal of Medicine 366:1275, April 5, 2012.) The paper is rather difficult -- full of statistics. Those seriously interested may find the beginning and end of the article more accessible. Most readers will want to content themselves with the main ideas, from the post here or perhaps one of the news stories.

The vaccine trial:
* The original post was HIV vaccine trial -- and quibbling about statistics (November 2, 2009). It links to various follow-ups. The key one for now is the one listed below.
* Preliminary information about the analysis of individual responses: Why did the HIV vaccine work for some people? (September 27, 2011).

Also see a recent post on an unusual approach to making an HIV vaccine. A novel approach to providing immunity to HIV (March 12, 2012). This approach could be relevant if work such as described above leads to the desirability of providing a particular kind of antibody.

The underlying issue here is how people differ in medically relevant ways. The increasing role of genome information has been a major catalyst for this field. Several posts on personalized are listed at: Personalized medicine: Getting your genes checked (October 27, 2009).

Here is another post involving an immune response that works backwards from what we want. Why are HIV-infected people more susceptible to Salmonella infection? (May 21, 2010).