Effects of low-chloride solutions on action potentials of sheep cadiac purkinje fibers

The rapid repolarization during phase 1 of the action potential of sheep cardiac purkinje fibers has been attributed to a time- and voltage-dependent chloride current. In part, this conclusion was based on experiments that showed a substantial slowing of phase 1 when larger, presumably impermeant, anions were substituted for chloride in tyrode’s solution. We have re- examined the electrical effects of low-chloride solutions. We recorded action potentials of sheep cardiac purkinje fibers in normal tyrode’s solution and in low-chloride solutions made by substituting sodium propionate, acetylglycinate, methylsulfate, or methanesulfonate for the NaCl of Tyrode’s solution. Total calcium was adjusted to keep calcium ion activity of test solutions equal to that of control solutions. Propionate gave qualitatively variable results in preliminary experiments; it was not tested further. Low-chloride solutions made with the other anions gave much more consistent results: phase 1 and the notch that often occurs between phases 1 and 2 were usually unaffected, and the action potential duration usually increased. The only apparent change in the resting potential was a transient 3-6 mV depolarization when low-chloride solution was first admitted to the chamber, and a symmetrical transient hyperpolarization when chloride was returned to normal. If a time- and voltage-dependent chloride current exists in sheep cardiac purkinje fibers, our results suggest that it plays little role in generating phase 1 of the action potential.     

Bayesian Computation Emerges in Generic Cortical Microcircuits through Spike-Timing-Dependent Plasticity

The principles by which networks of neurons compute, and how spike-timing dependent plasticity (STDP) of synaptic weights generates and maintains their computational function, are unknown. Preceding work has shown that soft winner-take-all (WTA) circuits, where pyramidal neurons inhibit each other via interneurons, are a common motif of cortical microcircuits. We show through theoretical analysis and computer simulations that Bayesian computation is induced in these network motifs through STDP in combination with activity-dependent changes in the excitability of neurons. The fundamental components of this emergent Bayesian computation are priors that result from adaptation of neuronal excitability and implicit generative models for hidden causes that are created in the synaptic weights through STDP. In fact, a surprising result is that STDP is able to approximate a powerful principle for fitting such implicit generative models to high-dimensional spike inputs: Expectation Maximization. Our results suggest that the experimentally observed spontaneous activity and trial-to-trial variability of cortical neurons are essential features of their information processing capability, since their functional role is to represent probability distributions rather than static neural codes. Furthermore it suggests networks of Bayesian computation modules as a new model for distributed information processing in the cortex.     

Using co-occurrence network structure to extract synonymous gene and protein names from MEDLINE abstracts

Background  

Text-mining can assist biomedical researchers in reducing information overload by extracting useful knowledge from large collections of text. We developed a novel text-mining method based on analyzing the network structure created by symbol co-occurrences as a way to extend the capabilities of knowledge extraction. The method was applied to the task of automatic gene and protein name synonym extraction.     

An IGF-I promoter polymorphism modifies the relationships between birth weight and risk factors for cardiovascular disease and diabetes at age 36

BACKGROUND: Biochemical testing for pheochromocytoma by measurement of fractionated plasma metanephrines is limited by false positive rates of up to 18% in people without known genetic predisposition to the disease. The plasma normetanephrine fraction is responsible for most false positives and plasma normetanephrine increases with age. The objective of this study was to determine if we could improve the specificity of fractionated plasma measurements, by statistically adjusting for age. METHODS: An age-adjusted metanephrine score was derived using logistic regression from 343 subjects (including 33 people with pheochromocytoma) who underwent fractionated plasma metanephrine measurements as part of investigations for suspected pheochromocytoma at Mayo Clinic Rochester (derivation set). The performance of the age-adjusted score was validated in a dataset of 158 subjects (including patients 23 with pheochromocytoma) that underwent measurements of fractionated plasma metanephrines at Mayo Clinic the following year (validation dataset). None of the participants in the validation dataset had known genetic predisposition to pheochromocytoma. RESULTS: The sensitivity of the age-adjusted metanephrine score was the same as that of traditional interpretation of fractionated plasma metanephrine measurements, yielding a sensitivity of 100% (23/23, 95% confidence interval [CI] 85.7%, 100%). However, the false positive rate with traditional interpretation of fractionated plasma metanephrine measurements was 16.3% (22/135, 95% CI, 11.0%, 23.4%) and that of the age-adjusted score was significantly lower at 3.0% (4/135, 95% CI, 1.2%, 7.4%) (p < 0.001 using McNemar's test). CONCLUSION: An adjustment for age in the interpretation of results of fractionated plasma metanephrines may significantly decrease false positives when using this test to exclude sporadic pheochromocytoma. Such improvements in false positive rate may result in savings of expenditures related to confirmatory imaging.     

Genetic affinities of inbred mouse strains of uncertain origin

Phylogenetic analyses of genetic data arising from 144 gene loci are used to describe the interrelationships among 24 widely used inbred strains of mice. An unordered-parsimony analysis gives a cladogram that is virtually identical to the known genealogy of the mouse strains. A loss-parsimony analysis is used to evaluate the hypothesis that the observed patterns of genetic divergence among these 24 strains can be explained by the segregation of residual heterozygosity arising from a small population of highly heterozygous mice. The loss-parsimony cladogram is very similar to both the unordered-parsimony cladogram and the known genealogy of the mice. The phylogenetic analyses of these 144 loci are integrated with data on the type and origin of the Y chromosome. Inclusion of the Y-chromosome data provides additional insights into the genetic composition of several of the original stocks used to produce the current inbred strains of mice. Ten strains of uncertain origin are contained in these analyses, including AKR, BUB, CE, I, NZB, P, RF, SJL, ST, and SWR. SJL is hypothesized to have been derived from the same Swiss albino stock previously used to produce SWR. The BUB strain appears to have had a complex origin and shows closest genetic similarity to SWR and ST. AKR and RF are shown to be closely related, while the I strain shows greatest genetic similarity to DBA/2 for the 144 loci. However, I and DBA possess different types of Y chromosome. The NZB strain shows genetic similarity to several stocks of both U.S. and European origins. The power of the genetic data used in these analyses reiterates that inbred strains of mice can be a valuable paradigm for studies in evolutionary biology.      

The role of disjunction and postglacial population expansion on phylogeographical history and genetic diversity of the circumboreal plant Chamaedaphne calyculata

In the last decade a number of studies has illustrated quite different phylogeographical patterns amongst plants with a northern present‐day geographical distribution, spanning the entire circumboreal region and/or circumarctic region and southern mountains. These works, employing several marker systems, have brought to light the complex evolutionary histories of this group. Here I focus on one circumboreal plant species, Chamaedaphne calyculata (leatherleaf), to unravel its phylogeographical history and patterns of genetic diversity across its geographical range. A survey of 29 populations with combined analyses of chloroplast DNA (cpDNA), internal transcribed spacer (ITS) and AFLP markers revealed structuring into two groups: Eurasian/north‐western North American, and north‐eastern North American. The present geographical distribution of C. calyculata has resulted from colonization from two putative refugial areas: east Beringia and south‐eastern North America. The variation of chloroplast DNA (cpDNA) and ITS sequences strongly indicated that the evolutionary histories of the Eurasian/north‐western North American and the north‐eastern North American populations were independent of each other because of a geographical disjunction in the distribution area and ice‐sheet history between north‐eastern and north‐western North America. Mismatch analysis using ITS confirmed that the present‐day population structure is the result of rapid expansion, probably since the last glacial maximum. The AFLP data revealed low genetic diversity of C. calyculata (P = 19.5%, H = 0.085) over the whole geographical range, and there was no evidence of loss of genetic diversity within populations in the continuous range, either at the margins or in formerly glaciated and nonglaciated regions. © 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, 105 , 761–775.     

Neural dynamics as sampling: a model for stochastic computation in recurrent networks of spiking neurons

The organization of computations in networks of spiking neurons in the brain is still largely unknown, in particular in view of the inherently stochastic features of their firing activity and the experimentally observed trial-to-trial variability of neural systems in the brain. In principle there exists a powerful computational framework for stochastic computations, probabilistic inference by sampling, which can explain a large number of macroscopic experimental data in neuroscience and cognitive science. But it has turned out to be surprisingly difficult to create a link between these abstract models for stochastic computations and more detailed models of the dynamics of networks of spiking neurons. Here we create such a link and show that under some conditions the stochastic firing activity of networks of spiking neurons can be interpreted as probabilistic inference via Markov chain Monte Carlo (MCMC) sampling. Since common methods for MCMC sampling in distributed systems, such as Gibbs sampling, are inconsistent with the dynamics of spiking neurons, we introduce a different approach based on non-reversible Markov chains that is able to reflect inherent temporal processes of spiking neuronal activity through a suitable choice of random variables. We propose a neural network model and show by a rigorous theoretical analysis that its neural activity implements MCMC sampling of a given distribution, both for the case of discrete and continuous time. This provides a step towards closing the gap between abstract functional models of cortical computation and more detailed models of networks of spiking neurons.     

Probabilistic inference in general graphical models through sampling in stochastic networks of spiking neurons

An important open problem of computational neuroscience is the generic organization of computations in networks of neurons in the brain. We show here through rigorous theoretical analysis that inherent stochastic features of spiking neurons, in combination with simple nonlinear computational operations in specific network motifs and dendritic arbors, enable networks of spiking neurons to carry out probabilistic inference through sampling in general graphical models. In particular, it enables them to carry out probabilistic inference in Bayesian networks with converging arrows ("explaining away") and with undirected loops, that occur in many real-world tasks. Ubiquitous stochastic features of networks of spiking neurons, such as trial-to-trial variability and spontaneous activity, are necessary ingredients of the underlying computational organization. We demonstrate through computer simulations that this approach can be scaled up to neural emulations of probabilistic inference in fairly large graphical models, yielding some of the most complex computations that have been carried out so far in networks of spiking neurons.