Wednesday, December 26, 2012

Haven't we had this conversation before?

Update December 26th, 2012: The Mermaid's Tale also covers this story.

Isn't it supposed be Boxing Day?  I guess I'll always be a stranger in a strange land.

Joe Pickrell is tweeting that "researchers" at the University of Connecticut are going to sequence the genome of Adam Lanza to "discover" biological clues to extreme violence.  A journalist at the New York Times has actually written an article about it and it has been published in the "Science" section of yesterday's paper.

Hello? 

People have been studying psychopathy for years. For example, Robert Hare has devoted his life to its study.  Other researchers, too numerous to mention, have soldiered on, bravely looking at the interplay between drugs, illicit and otherwise, alcohol, stress and mental health.

I have been listening to the discussion, off and on, on NPR, to the public reaction to the Newtown gunman.  Once again, we discuss the cause:  the media, gun control, and mental health.

The mental health discussion caught my attention.  One women mentioned something to the effect that approximately one percent of the population struggles with schizophrenia, which is approximately the case.  She then made the statement that she had never met anyone with schizophrenia and proceeded to discuss schizophrenia in very simplistic terms, as if to say that if we could just round up all the people struggling with schizophrenia, put them in group homes, then we'd all be A-OK.  No need at all to do something about those assault weapons.  Finally, someone stepped in to say that many people who have schizophrenia have successful lives and are, in fact, heroic in their struggle with the disease.

Still, I found the conversation a reflection of our profound ignorance of mental health issues.

Moreover, the discussions about "gun control" are, in my opinion, ridiculous.  It will be very unAmerican of me to say so, but as far as I am concerned, there are a few reasons why you might want to have a rifle:  you need to get rid of the crows; maybe you're hiking in grizzly bear territory (outside of a national, state or provincial park); you're a hunter . . . but a .22 would probably do the job, maybe a little bear spray .  .  . if you wanted more, you could try a bow.

However, beyond an arms race, there is no place for automatic assault weapons in our cities.  Oh, right, the Second Amendment.  Been to any federal, state or city office lately?  They're defended to the teeth.  It's always nice to have to state your case to an official through bullet proof glass.  Makes for really cozy communication.

As to the "researchers" at the University of Connecticut, I am sympathetic to the plight of the many families.  I am sure they are desperate for answers.  However, the genome of the Newtown gunman will not provide any.  That is not to say that genomics will not eventually tell us something about schizophrenia, psychopathy and other diseases of the mind.

I am profoundly sad to say that we will likely be having this conversation again.

Saturday, December 22, 2012

Related Blogs sidebar moved up

I've moved the list of blogs I follow up near to the top of the right sidebar.  There's some great holiday reading there.  Thanks to all the bloggers for their terrific efforts ... makes me feel like all those years toiling away in communication semiconductor land were worth it.

Sunday, December 16, 2012

Seeing Ghosts

Six rainbows across Norway (link).  
While rainbows are often associated with magic and pots of gold, they are actually caused by refraction and reflection. An analogy occurs in population sampling, where diverged and remerged sub populations can cause the effect of seeing "ghosts", or images, in a sampled population.  This can cause several artefacts.  One possible artefact is the appearance of migration between nonadjacent populations. Another artefact can be that the apparent level of migration can be more than two orders of magnitude smaller than the actual level, depending on the number of waves of migration allowed by the model.

Seeing ghosts: the effect of unsampled populations on
migration rates estimated for sampled populations


Montgomery Slatkin
Department of Integrative Biology, University of California, Berkeley, CA 94720–3140, USA
29 September 2004
(link)

Abstract
In 2004, the term ‘ghost population’ was introduced to summarize the effect of unsampled subpopulations that exchange migrants with other subpopulations that have been sampled.  Estimated long-term migration rates among populations sampled will be affected by ghost populations. Although it would be convenient to be able to define an apparent migration matrix among sampled populations that incorporate the exchange of migrants with ghost populations, no such matrix can be defined in a way that predicts all features of the coalescent process for the true migration matrix. This paper shows that if the underlying migration matrix is symmetric, it is possible to define an apparent migration matrix among sampled subpopulations that predicts the same within-population and between-population homozygosities among sampled populations as is predicted by the true migration matrix. Application of this method shows that there is no simple relationship between true and apparent migration rates, nor is there a way to place an upper bound on the effect of ghost populations.  In general, ghost populations can create the appearance of migration between subpopulations that do not actually exchange migrants. Comparison with published results from the application of the program, MIGRATE, shows that the apparent migration rates inferred with that program in a three subpopulation model differ from those based on pairwise homozygosities.  The apparent migration matrix determined by the method described in this paper probably represents the upper bound on the effect of ghost populations.migrate, shows that the apparent migration rates inferred with that program in a three- subpopulation model differ from those based on pairwise homozygosities.  The apparent migration matrix determined by the method described in this paper probably represents the upper bound on the effect of ghost populations.

Wednesday, December 12, 2012

Efficient moment-based inference of admixture parameters and sources of gene flow




Luis, you have to check out this new MixMapper paper.
The method employs a concept called scaffolding within a moment method.

There's an interesting discussion about the employment of f2 distances versus genetic drift D statistics, which I'm sure, you, Luis, will want to read. (See "Expressing branch lengths in drift units."). There's also a good discussion of ascertainment bias in the "Data Set" section.

Also, unsurprisingly, it appears that it is actually quite difficult to find populations that have little admixture: "Despite the focus of the HGDP on isolated populations, most of the 53 HGDP groups exhibit signs of admixture detectable by the 3-population test, as has been noted previously (Patterson et al., 2012)." And in the end, these researchers can find only 20 populations that are potentially unadmixed.

Regarding Europeans, there are virtually no unadmixed populations. But, on the bright side, Luis, apparently among Sardinians and Basques, there are 20-25% “ancient northern Eurasian” allele frequencies.

Other very interesting results are that the Daur, Hezhen, Oroqen, and Yakut—are likely descended from admixtures between native North Asian populations and East Asian populations related to Japanese.

Also interesting: "we found that Han Chinese have an optimal placement as an approximately equal mixture of two ancestral East Asian populations, one related to modern Dai (likely more southerly) and one related to modern Japanese (likely more northerly), corroborating a previous finding of admixture in Han populations between northern and southern clusters in a large-scale analysis of East Asia (HUGO Pan-Asian SNP Consortium, 2009)."

And corroborating various mtDNA, y-DNA, ADMIXTURE results: "Mozabite, Bedouin, Palestinian, and Druze, in decreasing order of African ancestry, are all optimally represented as a mixture between an admixed western Eurasian population (not necessarily European) related to Sardinian and an African population (Table 3)." 

I have a friend who is a Berber who is proud of her partly African ancestry . . . she told me so, not even knowing anything about these genetic studies.

Oh, and check this out: "Orcadian to an ancestor of Americans." Score one for the "Across Atlantic Ice" hypothesis. I'll have to add Dennis' book to my book list.

Way to go, MIT!

Here's the paper:

Efficient moment-based inference of admixture parameters and sources of gene flow
Mark Lipson, Po-Ru Loh, Alex Levin, David Reich, Nick Patterson, and Bonnie Berger
http://arxiv.org/abs/1212.2555


"[We] introduce MixMapper, a new computational tool that fits admixture trees by solving systems of moment equations involving the pairwise distance statistic f2 (Reich et al., 2009; Patterson et al., 2012), which is the average squared allele frequency difference between two populations. The theoretical expectation of f2 can be calculated in terms of branch lengths and mixture fractions of an admixture tree and then compared to empirical data. MixMapper can be thought of as a generalization of the qpgraph package (Patterson et al., 2012), which takes as input genotype data, along with a proposed arrangement of admixed and unadmixed populations, and returns branch lengths and mixture fractions that produce the best fit to allele frequency moment statistics measured on the data. MixMapper, by contrast, performs the fitting in two stages, first constructing an unadmixed scaffold tree via neighbor-joining and then automatically optimizing the placement of admixed populations onto this initial tree. Thus, no topological relationships among populations need to be specified in advance.

"Our method is similar in spirit to the independently developed TreeMix method (Pickrell and Pritchard, 2012). Like MixMapper, TreeMix builds admixture trees from second moments of allele frequency divergences, although it does so via a composite likelihood maximization approach made tractable with a multivariate normal approximation. Procedurally, TreeMix is structured in a “top-down” fashion, whereby a full set of populations is initially fit as an unadmixed tree, and gene flow edges are added sequentially to account for the greatest errors in the fit (Pickrell and Pritchard, 2012). This format makes TreeMix well-suited to handling very large trees: the entire fitting process is automated and can include arbitrarily many admixture events simultaneously. In contrast, MixMapper is designed as an interactive tool to maximize flexibility and precision with a “bottom-up” approach, beginning with a carefully screened unadmixed scaffold tree to which admixed populations are added with best-fitting parameter solutions.

"We use MixMapper to model the ancestral relationships among 52 populations from the CEPH-Human Genome Diversity Cell Line Panel (HGDP) (Rosenberg et al., 2002; Li et al., 2008) using recently published data from a new, specially ascertained SNP array designed for population genetics applications (Keinan et al., 2007; Patterson et al., 2012). Previous studies of these populations have built simple phylogenetic trees (Li et al., 2008; Sir´en et al., 2011), identified a substantial number of admixed populations with likely ancestors (Patterson et al., 2012), and constructed a large-scale admixture tree (Pickrell and Pritchard, 2012). Here, we add an additional level of quantitative detail, obtaining best-fit admixture parameters and bootstrap error estimates for 30 HGDP populations, of which 20 are admixed. The results include, most notably, a significant admixture event in the history of all sampled European populations (Patterson et al., 2012), among them Sardinians and Basques."

Tuesday, December 11, 2012

First of His Tribe to this Fair Vale

                                            Loch Tay

'What aspect bore the man who roved or fled,
 First of his tribe to this fair vale -
 What hopes came with him?'
                                                            - Wordsworth -

Lindores Abbey and Its Burgh of Newburgh:  Their History and Annals
Alexander Laing, George Seton, Anthony Hamilton
Chapter 1:  Prehistoric, page 1.
1876

It needs no evidence to prove that men who navigated our shores and rivers, in canoes hollowed out of single trees, had made but little progress in the constructive arts.  About sixty years ago two canoes, so made, were found in the bed of the Tay, opposite Lindores Abbey, the largest was twenty-eight feet long, and was quite entire. [1]

Another relic, telling of a condition and aspect of country widely different from the present, was discovered in the neighbourhood of Newburgh, in the end of the last century.  In draining what was called the Session Loch, at Mugdrum, the skull of a 'Great Ox,' Bos primigenius, or Urus, was found.  So huge was this skull, that even in that unscientific age the people flocked to see it.  Dr. Fleming, in his 'History of British Animals,' records that it was 27 1/2 inches in length. [2]  He says nothing of the kind of strata in which it was found, for geologists to build deductions on; but the cutting was carried through a great ridge of sand and river gravel, and the head was discovered at a considerable depth below the surface.  The Urus was little inferior to the elephant in size; one skull measured by Cuvier gave the proportions of the animal to be 12 feet in length and 6 1/2 in height.  Other skeletons have been found of much greater magnitude, affording indubitable evidence of the gigantic size of these wild denizens of the ancient Scottish forests.

The wild ox was a favourite object of the chase among our barbarian forefathers, and it was counted a great feat for a young man to bring home the horns of a Urus; they edged the finest of these horns with silver, and used them as drinking cups at great festive gatherings.[3]

[1]  These canoes were taken out of the Cruive bank opposite Lindores Abbey.  They were cut up and used for lintels in the erection of granaries at the west shore of Newburgh.  The largest canoe ever found in Scotland was 36 feet long and 4 feet wide - it was discovered at Carron - Wilson's Prehistoric Annals, Ed. 1851., p. 32.  Out of a list of about fifty ancient canoes, recorded as having been discovered in the west of Europe, only three are mentioned as larger than the largest found at Cruive bank.  There is one which was found in the Rhône, preserved in the museum at Lyons, 41 feet long. - Figuier's Primitive Man, p. 17.

[2]  Wilson's Prehistoric Annals, Ed. 1851., p. 23.

[3]  The representation of hunting scenes, on so many of the 'Sculptured Stones of Scotland', of which Mugdrum Cross is an instance, is enduring evidence of the importance of the chase among our forefathers.

Sunday, December 9, 2012

Sensor fusion in dynamical sytems

DREAM Seminar: Sensor fusion in dynamical systems - applications and research challenges
Seminar: DREAMS | December 11 | 4:10-5 p.m. | Soda Hall, Wozniak Lounge

Thomas Schön, Linköping University, Sweden
Electrical Engineering and Computer Sciences (EECS)

Abstract: Sensor fusion refers to the problem of computing state estimates using measurements from several different, often complementary, sensors. The strategy is explained and (perhaps more importantly) illustrated using four different industrial/research applications, very briefly introduced below. Guided partly by these applications we will highlight key directions for future research within the area of sensor fusion. Given that the number of available sensors is skyrocketing this technology is likely to become even more important in the future. The four applications are; 1. Real-time pose estimation and autonomous landing of the helicopter (using inertial sensors and a camera). 2. Pose estimation of a helicopter using an already existing map (a processed version of an aerial photograph of the operational area), inertial sensors and a camera. 3. Vehicle motion and road surface estimation (using inertial sensors, steering wheel sensor and an infrared camera). 4. Indoor pose estimation of a human body (using inertial sensors and ultra-wideband).

Bio: Thomas B. Schön is an Associate Professor with the Division of Automatic Control at Linköping University (Linköping, Sweden). He received the BSc degree in Business Administration and Economics in Jan. 2001, the MSc degree in Applied Physics and Electrical Engineering in Sep. 2001 and the PhD degree in Automatic Control in Feb. 2006, all from Linköping University. He has held visiting positions with the University of Cambridge (UK) and the University of Newcastle (Australia). He is a Senior member of the IEEE. He received the best teacher award at the Institute of Technology, Linköping University in 2009. Schön's main research interest is nonlinear inference problems, especially within the context of dynamical systems, solved using probabilistic methods. He is active within the fields of machine learning, signal processing and automatic control. He pursue both basic research and applied research, where the latter is typically carried out in collaboration with industry. More information about his research can be found from his home page: users.isy.liu.se/rt/schon

broman@eecs.berkeley.edu, 510-460-0280

Machine Understanding of Live Drum Performances

Dissertation Talk

Seminar: Departmental | December 10 | 11 a.m.-12 p.m. | 373 Soda Hall

Eric Battenberg
Electrical Engineering and Computer Sciences (EECS)

This talk will cover machine listening techniques for the automated real-time analysis of live drum performances. Onset detection, drum detection, beat tracking, and drum pattern analysis are combined into a system that provides rhythmic information useful in performance analysis, synchronization, and retrieval. The talk will focus on the drum detection and pattern analysis components of the the system.

For drum detection, a gamma mixture model is used to compute multiple spectral templates per drum onto which onset events can be decomposed using a technique based on non-negative matrix factorization. Unlike classification-based approaches to drum detection, this approach provides amplitude information which is invaluable in the analysis of rhythm.

The drum pattern analysis component uses a generatively pre-trained deep neural network in order to estimate high-level rhythmic information. The network is tested with beat alignment tasks, including downbeat detection, and significantly reduces alignment errors compared with a commonly used pattern correlation method.

ericb@eecs.berkeley.edu

Friday, December 7, 2012

Ghana, Shine Bright

Ghanaian President John Dramani Mahama addresses his nation on parliamentary elections today in Ghana.
(link)

". . . On such a historic occasion, it is worth reminding ourselves that whatever our political differences are, Ghana's stable institutions, democratic culture and the fortitude of its people have at each election, collectively risen to the occasion and made us proud as a nation."

"Fellow Ghanaians, an election is a contest between competing policy visions and must never set families, ethnic groups and religions against one another."

"Out of this contest of ideas shall emerge a President and leader whose character embodies and reflects our collective aspirations as a nation towards peace, unity and accelerated socio-economic development."

Thursday, December 6, 2012

UBC Green College: 2012/13 Seminars on the Evolution of Religion, Cooperation and Morality

University of British Columbia
(link)

The evolutionary and cognitive sciences have recently experienced an explosion of work on religion, cooperation, and morality, and in particular on their interrelationships. The emerging framework promises to re-energize these long-languishing topics by bringing a fully interdisciplinary approach to these topics that synthesizes the integrative rigor and precision of the evolutionary sciences with the depth of history and ethnography. The series will feature both leading researchers on these topics from Vancouver and experts from across the globe.

Wednesday, December 5, 2012

La majestueuse Méditerranée

Désert des Agriates, La Corse  (link)

de Un soir de veillee en Provence, par Jean-Michel Oprendek

"Et, à deux pas de là, intemporelle dans son bleu manteau d’écume, la majestueuse Méditerranée, la mare nostrum, ce berceau de peuples ardents, de cités, et de cultures, sculptait sans fin le sable, polissait les galets, au gré des vents du large qui, se mariant aux nôtres, portaient de mille rives tous les parfums de la vie, notre vie."

A Four Degree World is Catastrophic

Former Irish President Mary Robinson:
Climate Change the Biggest Human Rights Issue of Our Time

(link)


Amy Goodman of Democracy Now! interviews
Mary Robinson at the Climate Conference in DOHA

"Unfortunately, there is a lot of position taking.  It's almost like a trade negotiation, rather than an urgent meeting about staying below 2 degrees celsius, getting urgent commitments on emissions, on adaptation, on transfer of technologies and on finance.  It's really very frustrating for those who come hoping that the negotiators will take their responsibilities."


What are we waiting for, assembled in the forum?
      The barbarians are due here today.

Why isn’t anything going on in the senate?
Why are the senators sitting there without legislating?
      Because the barbarians are coming today.
      What’s the point of senators making laws now?
      Once the barbarians are here, they’ll do the legislating.

Why did our emperor get up so early,
and why is he sitting enthroned at the city’s main gate,
in state, wearing the crown?
      Because the barbarians are coming today
      and the emperor’s waiting to receive their leader.
      He’s even got a scroll to give him,
      loaded with titles, with imposing names.

Why have our two consuls and praetors come out today
wearing their embroidered, their scarlet togas?
Why have they put on bracelets with so many amethysts,
rings sparkling with magnificent emeralds?
Why are they carrying elegant canes
beautifully worked in silver and gold?
      Because the barbarians are coming today
      and things like that dazzle the barbarians.

Why don’t our distinguished orators turn up as usual
to make their speeches, say what they have to say?
      Because the barbarians are coming today
      and they’re bored by rhetoric and public speaking.

Why this sudden bewilderment, this confusion?
(How serious people’s faces have become.)
Why are the streets and squares emptying so rapidly,
everyone going home lost in thought?
      Because night has fallen and the barbarians haven't come.
      And some of our men just in from the border say
      there are no barbarians any longer.

Now what’s going to happen to us without barbarians?
Those people were a kind of solution.



Waiting for the Barbarians, C.P. Cavafy: Collected Poems
Translated by Edmund Keeley and Philip Sherrard
Translation Copyright © 1975, 1992 by Edmund Keeley and Philip Sherrard
Reproduced with permission of Princeton University Press

Monday, December 3, 2012

The Adaptive Landscape in Evolutionary Biology

(Link)

"The 'Adaptive Landscape' has been a central concept in population genetics and evolutionary biology since this powerful metaphor was first formulated by Sewall Wright in 1932. Eighty years later, it has become a central framework in evolutionary quantitative genetics, selection studies in natural populations, and in studies of ecological speciation and adaptive radiations. Recently, the simple concept of adaptive landscapes in two dimensions (genes or traits) has been criticized and several new and more sophisticated versions of the original adaptive landscape evolutionary model have been developed in response. No published volume has yet critically discussed the past, present state, and future prospect of the adaptive landscape in evolutionary biology. This volume brings together prominent historians of science, philosophers, ecologists, and evolutionary biologists, with the aim of discussing the state of the art of the Adaptive Landscape from several different perspectives."

Purchase:
Amazon (link)