Psychosomatic health

I can’t qualify posts like this enough; they are zeroth order thoughts that are more like armchair musings than they are even close to anything scientific.

Fairly often I come across a science news article about some study that correlates behavioral/mental health with some kind of somatic manifestation. There’s a term that many bandy around, “psychosomatic,” often used superciliously to indicate that one’s pains are somehow made up in one’s mind. However, pain is a very personal, neurological experience that makes objectivity difficult to assess. A recent headline I saw posits a link between loneliness in women and a higher incidence of breast cancer. To assume without qualification, for a moment, that there is a mechanism behind this observation, I was thinking about what that might be. The intuitive reaction is often to assume that it may be through inaction that the brain/body might not fight aggressive cancer cell growth if the loneliness is associated with less than optimal brain function (as is easy to imagine). However, what happens when we think about this in the opposite way? What if the mechanism of action is similar to programmed organism death? What if, in detecting a sub-optimal neurological state, the brain actively contributes to a condition that makes it susceptible to a parasitic toxin such as cancer cells?

This is all firmly in the domain of a gedanken experiment at the moment, but I bet there is some research that investigates several of these issues and potentially how they might work together to understand this. To extend these vagaries further, understanding may lead to effective treatment that could be as simple as “being social” or “making oneself happy.” Mechanism unknown, perhaps, these kinds of simple things may have far greater ramifications for our psychosomatic health (in the non-pejorative way).

Update: the original article will be published in a journal called Cancer Prevention Research.

Cocktail party concerto

Zeroth order thought.

Someone recently was listening to a Brahms’ Violin Concerto and asked the question of whether or not the soloist could be picked out amid the background of the much louder orchestra, when playing tutti. I recently had seen some evidence presented by a speaker that might apply here, and I thought it would be fun to consider the two things together.

Essentially, the problem is the following. You have a very large contingent of instruments playing, with a single instrument playing something different. We would like to hear the soloist, either in harmony with the other parts or distinctly, above the din. Now, there is clearly an absolute volume at which the soloist would be drowned out — we believe this from experience. However, assuming we’re not at that level yet, there’s sort of an interesting problem of differentiating a very loud section of instruments that is often of equal timbre but of higher amplitude.

The cocktail party problem refers to the observation that, when in a crowded room in which people are speaking, by looking at a single speaker, one can often hear what she is saying, despite the distracting, surrounding noise. So the concerto problem is similar in this regard.

Dream job post

I was looking at job offerings and came across a post for a neurobiophysicist. Because it contained both the roots “neuro” and “biophysics,” my interest was piqued, since I’ve never heard of anyone referred to by this compounded title. It turns out that I’m apparently training to be a neurobiophysicist, and despite the location of the position (San Antonio, Texas), I am very intrigued by the job post. My hope is that other academic places are looking for someone who fits this description (emphasis added):

We seek to recruit a vigorous academic with an established track-record of research in the broad area of the physics of neurobiological processes. This might be anyone from a “patch clamper” to a computational biophysicist to someone studying signal transduction mechanisms but not necessarily limited to these examples. Trinity University has secured funds to endow the position and provide continuing funding for the position after our HHMI award expires. This professorship will be at the level of full professor. The position comes with a dedicated half-time support staff position, discretionary funds, and a reduced teaching load (6 rather than 9 contact hours per semester).

This position will be an appointment in the Department of Physics, providing some support to introductory physics courses, a course in biophysics (suitable for both physics and neuroscience majors), and a course in the faculty member’s specific area of expertise (specifically supporting neuroscience majors) in the teaching repertoire. In addition, the person hired in the position will maintain a vigorous research program that will engage undergraduate student researchers. The neuroscience major requires independent research as a culminating experience for the degree. Further details about the position are available in the position announcement.

The position will also work with faculty from biology, chemistry, and psychology in supporting the neuroscience major. This interdisciplinary major was established in 2005 with funding from our 2004 HHMI award. Through a combination and retirement and this HHMI-funded neurobiophysicist position, the neuroscience program will be expanded and redefined. In addition to this position, we will be hiring a cognitive scientist in Psychology and an animal behaviorist in Biology to support the neuroscience program.

It’s so accurately describes what I am interested in (and doing) because of aspects of combining computational approaches with electrophysiological techniques. Additionally, the appointment is in a physics department (my undergraduate degree is in physics), with teaching duties in both neuroscience and physics. I’m pretty far from doing a job search, but here’s to hoping positions like this become more common by the time I graduate!

The career plan (revised)

When I began my graduate educational journey, which started with a dead end job working in IT for a small business, I didn’t really have a good concept of how things might unfold. I did all of my applications with the not-so-professional guidance of close friends who had as much experience as I did in successfully applying to neuroscience graduate schools (none). When I was fortunate enough to have a choice between graduate programs, I had the luxury of choosing a research direction, between the genetically minded approach of one school to the computationally minded approach of the other.

I ultimately chose the computational direction because while clearly genetics is relevant now and will be forever relevant, I think that the coming of age of computational neuroscience is now. Genetics presents an ethical minefield that threatens to ultimately slow the scope of its reach (for better or worse), and I think this will delay the onset of the true age of genetic “understanding” with respect to nervous systems.

Additionally, another branch point came during rotations in my first years in graduate school, when I had some experimental opportunities to weigh against learning computational techniques. In the end I chose the mathematics, though, because I feel like these are techniques that are far more difficult to learn on one’s own. The anecdotal evidence for this has been realized time and again by experimentalists who are struggling to learn modeling techniques in a meaningful way. In contrast, I’ve seen a number of computational folks who have made the transition to experimental techniques seamlessly.

My plan thus far was to continue developing my computational techniques throughout grad school and then transition into a postdoctoral position and learn some electrophysiology. However, a phenomenal opportunity arose in which I might be able to do in vivo electrophysiology now, as part of my dissertation. I jumped at it, almost without proper or deliberate consideration. It sounds reasonable enough to me, though it obviously is accompanied by a bit of trepidation, considering the magnitude of this change.

I already have experience in biophysical modeling, and I’m working on some data analysis techniques which will serve me well. If I do electrophysiology now, then I can devote my postdoctoral appointment to different computational techniques, or perhaps a more mathematical project. Since the landscape of computational neuro is in a sense just as large as experimental neuro, there are far more techniques I have no experience with but am interested in learning. Once again, at this stage it’s unclear which approach will be more fruitful, but here’s to the uncertainty of the journey!

Observations on making music

My Mom and Dad valued music, and while my brother somehow managed to escape from our childhood without learning an instrument, I learned the violin. I wasn’t always eager about it and sometimes outright did not enjoy it, which was probably just a product of hard work and sacrifice to get better at it.

It’s been quite a long time since I have taken lessons, and my music teacher has since passed away, but I’m thankful that I can, at any time, pick up a violin and have a repertoire of songs that I can play. The delay since the last time I played almost doesn’t seem to matter. I don’t often need the sheet music in order to play difficult pieces I haven’t played in years. I find that when I don’t think about the piece and just let my motor memory perform, I get much better results than if I think consciously about the piece.

In fact, often I’ll be playing a piece I knew well once, and the moment I try to think of what note or shift might come next, I have to stop completely. I don’t understand the neural mechanism behind this.

Now my music theory is pretty poor, as I always avoided it as a kid, and I regret that now. And while I don’t have perfect pitch, I have pretty good relative pitch, which means that I can’t name a note randomly, but as soon as I have a known reference point, I can half-step and whole-step my way around just fine. I am sure that there are memory systems associated with the semantic ability to name notes that requires a highly functioning auditory ability to discern notes. It’s something like being able to name colors, irrespective of the subtleties of the shade. These are subtleties of sound.

On a side but possibly related note, I love my music and my jazz music, and I have a really hard time naming tunes with no words, even for songs that I can sing from memory completely. I wonder if this semantic naming system is associated with perfect pitch abilities. Additionally, this kind of evidence suggests to me that the recall for the actual tune (even for tunes I haven’t played but only heard) is very much separate from a lot of the metadata knowledge associated with it, such as the name. The performer might fit into this category, except it’s hard for me to dissociate since I can usually, based on style, make a good guess at the composer, even when I can’t name the exact tune.

I also learned a few other things today. While at a friend’s house this weekend, the subject of Christmas music came up, and I got out the old violin and started to pluck out a few tunes on it. I found myself doing something very strange. I can sing or hum tunes and avoid inciting thrown tomatoes. However, I find that when I’m searching on my violin for that first note, I often end up with a note that falls somewhere in the middle of the song, instead of the one I’m searching for!

That single note in the context of the song I’m thinking about is generally enough for me to play at least a few bars if not the entire rest of the simple song. I can even backtrack and find the first note and start from there, incorporating my new song phrases into a complete song.

I’ve always been fascinated by this apparent ability of mine to identify a song by as few a couple of notes. It really only takes one or two notes, in some kind of rhythmic and timbre (such as the instrument) context, to elicit a memory of the entire song. I can usually name songs I hear from one or two notes, in loud restaurants or on the radio, etc. It can be a pretty random recall of songs I don’t think I know well or haven’t heard in years. It’s annoying when it’s songs that are bad. I’m not sure how that works, either, though it’s not too far fetched to imagine a synfire chain-like string of activity that elicits a memory from a particular stimulus (the sound).

It took me only a few minutes to work out each one of the Christmas songs I was thinking about today. I am classically trained, which for me means that my improvisational skills are pretty poor, though I’m trying to work on them through exercises like transcription and fooling around with variations on a melody.

But I was searching for some sheet music and found a nice resource for Christmas carols, and it had easy transcriptions of a lot of songs, including several I hadn’t thought of. For the ones I had already worked out, the keys were sometimes different. Perhaps because of my reliance on relative rather than absolute pitch, the differences to me are more a matter of taste than of correctness.

For the songs that I was familiar with but didn’t think to work out earlier, playing just a few bars was more than enough to elicit the entire song’s playback in my head, and I was often able to play the rest of the song, without relying on the sheet music, at tempo! Granted, we’re not talking about anything at all complex, but it was pleasing to play the entire melody only to scroll down the sheet and find it was identically transcribed.

One final observation about music and memory. One major misconception about rhythm is that it’s completely dissociated from pitch. This is more often than not incorrect in real life, as different percussion instruments generally associated with strong rhythms (like drums) all have unique pitch qualities that probably are often overlooked (this is part of why sometimes cymbals are used instead of snares; they each have a harmonic quality). While rhythm technically does exist in the time domain, I find that I cannot dissociate certain rhythms (William Tell Overture, as an example) from their melodic content. If I tap out, with my fingers on a desk, that famous rhythm, I cannot hear just fingers tapping on a desk, since I was very young. Not even consciously. I always hear the melody riding on that rhythmic structure. This leads to the curious phenomenon of random songs entering my head when I hear clapping or doors shutting, etc.

The latter of these observations reminds me of those visual tests with the old and young woman, in which you naturally see one or the other. Once you are aware of the presence of the other, you cannot help but see it.

All of these curious observations just make me wonder about how our brains are capable of such breadth (numbers of songs) and depth (detailed knowledge about each). How is this memory stored, and where? How many listens does it take to learn a song and under what conditions is it most salient? We have hints at the answers to many of these questions, but many more remain. It’s a fascinating time to be in the field of neuroscience.

While I’m thinking about it, this all reminds me that I just recently started reading Oliver Sack’s book Musicophilia. It’s essentially a collection of case studies involving patients with peculiar relationships with music. From what I’ve read so far, I highly recommend it.

The linear approximation of SfN

I recently came back from the Society for Neuroscience (SfN) mega-conference that draws over 30 kNeuroscientists each year. That’s right, 3×10⁴. Compare this, of course, to the annual conference for the American Association for the Advancement of Science, which draws about 10.000, or 1/3 of the number. Yet neuroscience is a subset of science, so, something funny must be happening. Is there a conference larger than SfN? I’d do well to avoid it, I think.

I looked at SfN this year from the limited but functional perspective of a computational neuroscientist. For the experimental posters and talks I attended, I tried to think of ways in which they could be modeled. On my mind notably is modeling of calcium wave propagation dynamics, which is an active area of imaging and may benefit from computational approaches. Surely there is work being done on this currently.

For the computational presentations, I tried to learn about various techniques that I have not yet been exposed to. There was a lot of basic Hodgkin-Huxley type modeling, along with a bunch of compartmental modeling. But what caught my eye was the use of stochastic processes in creating a framework for studying spiking neurons. At its best it may provide a mathematically rigorous description for certain experimentally observable data.

The most fruitful conversation I had was with a fellow grad student who had access to some very unique electrocorticogram (ECoG) data in auditory cortex of humans and had done some preliminary frequency analyses during natural, meaningful sounds (such as a spoken sentence). There is a songbird analog to this idea using natural, complex sounds that constitutes a very active area of research for auditory physiologists and others.

One interesting (to me) observation about the conference was the inelastic collision between vendors, industry, funding sources, institutions, and scientists. A lot of toys (swag) were doled out. I asked an NIH person about computational neuroscience support, and while they financially support efforts of collaborative science that include computation (a very reasonable approach, especially considering the NIH mission), they do not have but a handful of computational neuroscientists staffed at NIH.

One guy whose job was to connect industries with researchers said, when asked about the perceived value of computational neuroscientists in industry, “None yet. But keep asking.” I don’t know how indicative of greater industry he represents, but this means to me that the value of computational approaches may not have quite received full understanding in that community.

There seems to be a lot of convincing to be done.

The conference was not overwhelming as cautioned because I think I put enough (perhaps too many?) constraints on my activities. If I tried to do too much, I felt that I would have learned nothing. But focusing on learning a few things I know about and a few things I’m interested in learning pertaining to my own work, and the conference was very useful.

Functional Symmetry

I am intrigued with conservation and symmetry and general optimization properties as they pertain to biological systems. Well, perhaps simplicity also comes into play, in considerations of general aesthetics.

But there’s something profoundly beautiful about observations of the old biology tenet, “form follows function.” In an investigation of ion channels in biological membranes, particularly of neurons, one often notices a few things. First, a cell’s membrane is rather impermeable to ionic “stuff.” So in order for transport of ions, for instance, across a membrane, there have to be holes in the membrane.

As it turns out, there are many types of holes, including these ion channels, which are often selective for a particular ion. By and large, only that ion can go through its channel, and there are various ways of regulating this. Within the inside of the channels, called the pore (center of figure above), there are at least two factors involved in selectivity of ions – on one hand there is the physical size of the pore, and there is also what’s called a selectivity filter. A selectivity filter is essentially a binding site for the ion that enables it to continue along its pathway from one side of the membrane to the other. (There’s generally some sort of weak hydrogen-type bonding that occurs at that site, I think.)

So consider this – let’s say the ion, say a K⁺, is moving from the inside to the outside of the cell across a membrane. The selectivity would be such that you’d only want K⁺ leaving through the channel and not Ca²⁺ or Na⁺. But in the reverse direction – from out to in – the selectivity ought to be equally efficient.

This implies that the observed transverse symmetry in the protein channel is necessary for the bidirectional selectivity either by size of the ion or using a chemical segment as a filter! Quite fascinating, if you ask me!

“Magnetic Resonance Spectroscopy Identifies Neural Progenitor Cells in the Live Human Brain”

Using widely available techniques, Manganas et al. have been able to characterize a marker to quantify the presence of neural progenitor cells (NPCs) in rodent and human brains, in vivo. They are actually utilizing much of the same technique that takes two distinct functional forms; at the end of the day it’s all nuclear magnetic resonance imaging, but the commonly used NMR in organic chemistry is employed for characterizing chemicals, while the more familiar (to the neurosciences) magnetic resonance imaging (MRI) is utilized for visualization of the brains in vivo. According to Manganas et al., “Our findings thus open the possibility of investigating the role of NPCs and neurogenesis in a wide variety of human brain disorders.”

In regard to the plasticity of adult brains, though known for several years now, it often takes a long time for dogma in sciences to change. It certainly does not occur on the scale of developmental processes, and it’s largely a coping mechanism to compensate for various injuries. However, the evidence for dynamism within the adult brain is clear — it’s actually hard to imagine proper neural function within our daily lives without it, as every new “thing” we learn has got to have a neural correlate.

The article appears in this week’s issue of Science Magazine.

Brainbow imaging

Admittedly, molecular biology reads somewhat like the Post-Modern Prometheus, complete with green glowing rabbits and the like. The power to manipulate genetics has been harnessed in many ways in the last couple of decades to give tremendous insight into the complex world of biology. Recently, the Lichtman lab at Harvard University published their results regarding yet another new technology that may just live up to the promise.

The method harnesses a well known Cre/lox strategy of genetic recombination with a clever alteration, which enables the system to naturally create a large variety of permutations that are manifested as distinct fluorescent colors. Thus each cell will express its own color and can be viewed much more distinctly in the absolute mess of cells that constitutes this organ.

Where modern cellular imaging methods, such as fluorescence or electron microscopy, and “ancient” (but still employed) techniques of Golgi silver staining and Nissl staining are often plagued by either problems of sparseness or completeness in their ability to help visualize neurons, by the very nature of this technique, cells can be followed and more or less uniquely identified.

The article can be found in the journal Nature.

Synfire chains, supersynapses, and STDP

An idea proposed in the early 80s called synfire chains proposes that self-assembling networks of neurons can form in which small clusters or groups of neurons communicate with each other, possibly in a sequential linear fashion, more than other neurons. This concept is somewhat interesting if it can play a role in explaining how small networks of neurons spontaneously choose to communicate selectively with each other and possibly form cortical ensembles.

In a paper entitled “Development of Neural Circuitry for Precise Temporal Sequences through Spontaneous Activity, Axon Remodeling, and Synaptic Plasticity” by Jun and Jin, they explore the formation of these synfire chains with a computational model that utilizes leaky integrate and fire neurons along with two activity dependent learning rules. The more well known of these rules is the Hebbian type spike timing dependent plasticity (STDP), a phenomenological model that takes into account the relative timing between pre- and postsynaptic pairs of spikes.

Additionally, employing various thresholds of synaptic potentiation, the second rule they employ caps the number and strength of axonal remodeling — that is to say that when their neurons make strong connections with certain synapses (so-called supersynapses), they prune away other, weaker connections. It is this additional contribution that allows their model to spontaneously exhibit the synfire chain like behavior.

The article appeared in Aug. 2007 online issue of PLoS One, which is a progressive online journal offering free access. You can grab the article at http://www.plosone.org/.