A Primer on Bayesian Inference for Biophysical Systems

Bayesian inference is a powerful statistical paradigm that has gained popularity in many fields of science, but adoption has been somewhat slower in biophysics. Here, I provide an accessible tutorial on the use of Bayesian methods by focusing on example applications that will be familiar to biophysicists. I first discuss the goals of Bayesian inference and show simple examples of posterior inference using conjugate priors. I then describe Markov chain Monte Carlo sampling and, in particular, discuss Gibbs sampling and Metropolis random walk algorithms with reference to detailed examples. These Bayesian methods (with the aid of Markov chain Monte Carlo sampling) provide a generalizable way of rigorously addressing parameter inference and identifiability for arbitrarily complicated models.     

Crystal Structure of the Nipah Virus Phosphoprotein Tetramerization Domain

The Nipah virus phosphoprotein (P) is multimeric and tethers the viral polymerase to the nucleocapsid. We present the crystal structure of the multimerization domain of Nipah virus P: a long, parallel, tetrameric, coiled coil with a small, α-helical cap structure. Across the paramyxoviruses, these domains share little sequence identity yet are similar in length and structural organization, suggesting a common requirement for scaffolding or spatial organization of the functions of P in the virus life cycle.     

Determination of parameter identifiability in nonlinear biophysical models: A Bayesian approach

A major goal of biophysics is to understand the physical mechanisms of biological molecules and systems. Mechanistic models are evaluated based on their ability to explain carefully controlled experiments. By fitting models to data, biophysical parameters that cannot be measured directly can be estimated from experimentation. However, it might be the case that many different combinations of model parameters can explain the observations equally well. In these cases, the model parameters are not identifiable: the experimentation has not provided sufficient constraining power to enable unique estimation of their true values. We demonstrate that this pitfall is present even in simple biophysical models. We investigate the underlying causes of parameter non-identifiability and discuss straightforward methods for determining when parameters of simple models can be inferred accurately. However, for models of even modest complexity, more general tools are required to diagnose parameter non-identifiability. We present a method based in Bayesian inference that can be used to establish the reliability of parameter estimates, as well as yield accurate quantification of parameter confidence.     

The Extracellular and Transmembrane Domains of the γC and Interleukin (IL)-13 Receptor α1 Chains, Not Their Cytoplasmic Domains, Dictate the Nature of Signaling Responses to IL-4 and IL-13

Previously, we demonstrated that the γC subunit of type I IL-4 receptor was required for robust tyrosine phosphorylation of the downstream adapter protein, IRS-2, correlating with the expression of genes (ArgI, Retnla, and Chi3l3) characteristic of alternatively activated macrophages. We located an I4R-like motif (IRS-2 docking sequence) in the γC cytoplasmic domain but not in the IL-13Rα1. Thus, we predicted that the γC tail directed enhanced IRS-2 phosphorylation. To test this, IL-4 signaling responses were examined in a mutant of the key I4R motif tyrosine residue (Y325F) and different γC truncation mutants (γ285, γ308, γ318, γ323, and γFULL LENGTH (FL)) co-expressed in L-cells or CHO cells with wild-type (WT) IL-4Rα. Surprisingly, IRS-1 phosphorylation was not diminished in Y325F L-cell mutants suggesting Tyr-325 was not required for the robust insulin receptor substrate response. IRS-2, STAT6, and JAK3 phosphorylation was observed in CHO cells expressing γ323 and γFL but not in γ318 and γ285 mutants. In addition, when CHO cells expressed γ318, γ323, or γFL with IL-2Rβ, IL-2 induced phospho-STAT5 only in the γ323 and γFL clones. Our data suggest that a smaller (5 amino acid) interval than previously determined is necessary for JAK3 activation/γC-mediated signaling in response to IL-4 and IL-2. Chimeric receptor chains of the γC tail fused to the IL-13Rα1 extracellular and transmembrane domain did not elicit robust IRS-2 phosphorylation in response to IL-13 suggesting that the extracellular/transmembrane domains of the IL-4/IL-13 receptor, not the cytoplasmic domains, control signaling efficiency. Understanding this pathway fully will lead to rational drug design for allergic disease.     

The genetic network controlling the Arabidopsis transcriptional response to Pseudomonas syringae pv. maculicola: roles of major regulators and the phytotoxin coronatine

Expression profiling of wild-type plants and mutants with defects in key components of the defense signaling network was used to model the Arabidopsis network 24 h after infection by Pseudomonas syringae pv. maculicola ES4326. Results using the Affymetrix ATH1 array revealed that expression levels of most pathogen-responsive genes were affected by mutations in coi1, ein2, npr1, pad4, or sid2. These five mutations defined a small number of different expression patterns displayed by the majority of pathogen-responsive genes. P. syringae pv. tomato strain DC3000 elicited a much weaker salicylic acid (SA) response than ES4326. Additional mutants were profiled using a custom array. Profiles of pbs3 and ndr1 revealed major effects of these mutations and allowed PBS3 and NDR1 to be placed between the EDS1/PAD4 node and the SA synthesis node in the defense network. Comparison of coi1, dde2, and jar1 profiles showed that many genes were affected by coi1 but very few were affected by dde2 or jar1. Profiles of coi1 plants infected with ES4326 were very similar to those of wild-type plants infected with bacteria unable to produce the phytotoxin coronatine, indicating that, essentially, all COI1-dependent gene expression changes in this system are caused by coronatine.     

Kallikreins are associated with secondary progressive multiple sclerosis and promote neurodegeneration

Tissue kallikrein KLK1 and the kallikrein-related peptidases KLK2-15 are a subfamily of serine proteases that have defined or proposed roles in a range of central nervous system (CNS) and non-CNS pathologies. To further understand their potential activity in multiple sclerosis (MS), serum levels of KLK1, 6, 7, 8 and 10 were determined in 35 MS patients and 62 controls by quantitative fluorometric ELISA. Serum levels were then correlated with Expanded Disability Status Scale (EDSS) scores determined at the time of serological sampling or at last clinical follow-up. Serum levels of KLK1 and KLK6 were elevated in MS patients (p相似文献   

Stress induced experimental colitis

Stress induces chemical changes in the central nervous system which alters the biochemistry and physiology of the digestive tract. The present study determines arachidonic acid oxidation and damage in the colon following stress. Ten rats were stressed by the cold-restraint method; ten were controls. Stress induced 0.5 +/- 0.7 (S.D.) mucosal erosions whereas controls had none. Subepithelial hemorrhage and erosions occurred only in the proximal two-thirds of the colon. Prostaglandin E(2) synthesis was increased after stress compared to the control (381 +/- 130 vs. 1610 +/- 372 ng/g/min). Leukotriene C(4) synthesis also increased after stress (4217 +/- 994 vs. 11300 +/- 1662 ng/g/min). Synthesis of prostaglandin E(2) increased (r = 0.9381) with leukotriene C(4). The response of the colon to stress is less severe than that in the stomach and may be related to regional regulation of prostaglandin and leukotriene synthesis.     

Plasma catecholamine levels during temperature biofeedback training in normal subjects

Thirty-nine normal volunteers of both sexes were randomly assigned to receive 8 sessions of temperature biofeedback or autogenic training to increase finger temperature. Temperature biofeedback subjects produced significant elevations in finger temperature during training, whereas those who received autogenic training did not. Temperature feedback subjects had significantly higher heart rates and diastolic blood pressures during training compared to autogenic subjects. There were no significant changes or group differences in plasma catecholamine levels. These data do not support the hypothesis that feedback-induced vasodilation is accompanied by decreased sympathetic activation in normal populations, when only temperature biofeedback is employed.Supported by research grant No. HL-30604 from NHLBI. Dr. Angela McGuady served as Action Editor for this paper.     

Nucleotide sequence of the Acinetobacter calcoaceticus trpGDC gene cluster

A plasmid library of Acinetobacter calcoaceticus HindIII fragments was constructed, and clones that complemented an Escherichia coli pabA mutant were selected. Plasmids containing a 3.9-kb fragment of A. calcoaceticus DNA that also complemented E. coli trpD and trpC-(trpF+) mutants were obtained. We infer that complementation of E. coli pabA mutants was the result of the expression of the amphibolic anthranilate- synthase/p-aminobenzoate-synthase glutamine-amidotransferase gene and that the plasmid insert carried the entire trpGDC gene cluster. In E. coli minicells, the plasmid insert directed the synthesis of polypeptides of 44,000, 33,000, and 20,000 daltons, molecular masses that are consistent with the reported molecular masses of phosphoribosylanthranilate transferase, indoleglycerol-phosphate synthase, and anthranilate-synthase component II, respectively. A 3,105- bp nucleotide sequence was determined. Comparison of the A. calcoaceticus trpGDC sequences with other known trp gene sequences has allowed insight into (1) the evolution of the amphibolic trpG gene, (2) varied strategies for coordinate expression of trp genes, and (3) mechanisms of gene fusions in the trp operon.