Connectionistic models of boolean category representation

Several distinct connectionistic/neural representations capable of computing arbitrary Boolean functions are described and discussed in terms of possible tradeoffs between time, space, and expressive clarity. It is suggested that the ability of a threshold logic unit (TLU) to represent prototypical groupings has significant advantages for representing real world categories. Upper and lower bounds on the number of nodes needed for Boolean completeness are demonstrated. The necessary number of nodes is shown to increase exponentially with the number of input features, the exact rate of increase depending on the representation scheme. In addition, in non-recurrent networks, connection weights are shown to increase exponentially with a linear reduction in the number of nodes below approximately 2^d. This result suggests that optimum memory efficiency may require unacceptable learning time. Finally, two possible extensions to deal with non-Boolean values are considered.     

Vibrational mode frequency calculations of chlorophyll-<Emphasis Type="Italic">d</Emphasis> for assessing (P740<Superscript>+</Superscript>-P740) FTIR difference spectra obtained using photosystem I particles from <Emphasis Type="Italic">Acaryochloris marina</Emphasis>

Acaryochloris marina is an oxygen-evolving organism that utilizes chlorophyll-d for light induced photochemistry. In photosystem I particles from Acaryochloris marina, the primary electron donor is called P740, and it is thought that P740 consist of two chlorophyll-d molecules. (P740⁺-P740) FTIR difference spectra have been produced, and vibrational features that are specific to chlorophyll-d (and not chlorophyll-a) were observed, supporting the idea that P740 consists chlorophyll-d molecules. Although bands in the (P740⁺-P740) FTIR difference spectra were assigned specifically to chlorophyll-d, how these bands shifted, and how their intensities changed, upon cation formation was never considered. Without this information it is difficult to draw unambiguous conclusions from the FTIR difference spectra. To gain a more detailed understanding of cation induced shifting of bands associated with vibrational modes of P740 we have used density functional theory to calculate the vibrational properties of a chlorophyll-d model in the neutral, cation and anion states. These calculations are shown to be of considerable use in interpreting bands in (P740⁺-P740) FTIR difference spectra. Our calculations predict that the 3¹ formyl C–H mode of chlorophyll-d upshifts/downshifts upon cation/anion formation, respectively. The mode intensity also decreases/increases upon cation/anion formation, respectively. The cation induced bandshift of the 3¹ formyl C–H mode of chlorophyll-d is also strongly dependant on the dielectric environment of the chlorophyll-d molecules. With this new knowledge we reassess the interpretation of bands that were assigned to 3¹ formyl C–H modes of chlorophyll-d in (P740⁺-P740) FTIR difference spectra. Considering our calculations in combination with (P740⁺-P740) FTIR DS we find that the most likely conclusions are that P740 is a dimeric Chl-d species, in an environment of low effective dielectric constant (∼2–8). In the P740⁺ state, charge is asymmetrically distributed over the two Chl-d pigments in a roughly 60:40 ratio. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Proof of the Feldman–Karlin conjecture on the maximum number of equilibria in an evolutionary system

Feldman and Karlin conjectured that the number of isolated fixed points for deterministic models of viability selection and recombination among n possible haplotypes has an upper bound of 2ⁿ−1. Here a proof is provided. The upper bound of 3ⁿ⁻¹ obtained by Lyubich et al. (2001) using Bézout’s Theorem (1779) is reduced here to 2ⁿ through a change of representation that reduces the third-order polynomials to second order. A further reduction to 2ⁿ−1 is obtained using the homogeneous representation of the system, which yields always one solution ‘at infinity’. While the original conjecture was made for systems of selection and recombination, the results here generalize to viability selection with any arbitrary system of bi-parental transmission, which includes recombination and mutation as special cases. An example is constructed of a mutation-selection system that has 2ⁿ−1 fixed points given any n, which shows that 2ⁿ−1 is the sharpest possible upper bound that can be found for the general space of selection and transmission coefficients.     

The strengths and limitations of using biolayer interferometry to monitor equilibrium titrations of biomolecules

Every method used to quantify biomolecular interactions has its own strengths and limitations. To quantify protein‐DNA binding affinities, nitrocellulose filter binding assays with ³²P‐labeled DNA quantify K_d values from 10^?12 to 10^?8 M but have several technical limitations. Here, we considered the suitability of biolayer interferometry (BLI), which monitors association and dissociation of a soluble macromolecule to an immobilized species; the ratio k_off/k_on determines K_d. However, for lactose repressor protein (LacI) and an engineered repressor protein (“LLhF”) binding immobilized DNA, complicated kinetic curves precluded this analysis. Thus, we determined whether the amplitude of the BLI signal at equilibrium related linearly to the fraction of protein bound to DNA. A key question was the effective concentration of immobilized DNA. Equilibrium titration experiments with DNA concentrations below K_d (equilibrium binding regime) must be analyzed differently than those with DNA near or above K_d (stoichiometric binding regime). For ForteBio streptavidin tips, the most frequent effective DNA concentration was ~2 × 10^?9 M. Although variation occurred among different lots of sensor tips, binding events with K_d ≥ 10^?8 M should reliably be in the equilibrium binding regime. We also observed effects from multi‐valent interactions: Tetrameric LacI bound two immobilized DNAs whereas dimeric LLhF did not. We next used BLI to quantify the amount of inducer sugars required to allosterically diminish protein‐DNA binding and to assess the affinity of fructose‐1‐kinase for the DNA‐LLhF complex. Overall, when experimental design corresponded with appropriate data interpretation, BLI was convenient and reliable for monitoring equilibrium titrations and thereby quantifying a variety of binding interactions.     

Requirement for the location of both appropriate and irrelevant H-2 antigens on the same stimulator cell for unspecific DNA-synthesis inhibition by the H-2-antigen-primed,specific suppressor T cells

Specific suppressor T cells (SSTC), primed in vivo with H-2 antigens, have been shown previously to inhibit DNA synthesis in the one-way, three-cell mixed lymphocyte reaction (MLR) provided that (a) the stimulator cells bear the priming H-2 antigens, and (b) the responder cells possessIC+S regions homologous to those of the SSTC. Anti-B10.A BlO.A(2R) SSTC (anti-D^d) and anti-A.AL A.TL SSTC (anti-K^k) are shown here to be able to inhibit the DNA synthesis triggered in MLR, not only by the corresponding antigens, D^d and K^k, respectively, but also by irrelevant, third-party H-2 and Mls products provided that the corresponding and third-party antigens are presented on the same stimulator cell. If stimulatorH-2 regions, whose products interact with SSTC and responders, are located on different stimulator cells within the particular MLR, SSTC activity is not elicited. Participation of cytotoxic T lymphocytes in DNA-synthesis suppression is ruled out. Direct contact or location of the inhibited responder cell very close to SSTC is considered to be required for the development of SSTC activity.     

Distribution of Gammarus lacustris Sars (Amphipoda, Gammaridae) in Lake Shira (Khakasia, Siberia) and laboratory study of its growth characteristics

Spatial distribution of Gammarus lacustris in Lake Shira and growth of young specimens that feed on lake biota were studied. The amphipods have been shown to inhabit the littoral, sublittoral and the upper aphytal zones of the lake on stony-sandy soil and silted sand. The young and adult individuals stay apart. The young live in submerged or semi-submerged vegetation in the littoral, the adults in the sublittoral and upper aphytal zones. Maximum density of amphipods was encountered in the areas influenced by human activity.The feeding experiments revealed that the lake plankton is a more important food source for G. lacustris than any other food species. The specific growth rate measured was 0.039 d^–1, with a length increment 0.095 mm d^–1.     

Electrophysiology of the marine diatom Coscinodiscus wailesii IV: types of non-linear current-voltage-time relationships recorded with single saw-tooth voltage-clamp experiments

Electrophysiological states of the marine diatom Coscinodiscus wailesii are known to change spontaneously in the temporal range of seconds. In order to assess the genuine current-voltage-time relationships of individual states in less than a second, voltage-clamp experiments have been carried out using single sweeps of saw-tooth shaped command voltages. This method is introduced with model calculations. Plotting the results in current-voltage coordinates provides convenient access to several electrophysiological entities, such as absence of drift (smoothly closed IV loops), membrane capacitance (by I jump at sign reversal of dV/dt), and ohmic conductances (in linear regions of the current-voltage relationship), as well as equilibrium voltage (internal intersection of capacitance-corrected, 8-shaped tracings) and coarse gating kinetics (rise or fall of capacitance-corrected I at sign reversal of dV/dt) of a voltage-sensitive ion conductance. From electrophysiological measurements with double-barreled glass-microelectrodes on C. wailesii, several distinct types of current-voltage loops are presented. Most of the data, including recordings from electrical excitation, can be interpreted as temporal relaxations of voltage-sensitive conductances for K⁺ and Cl⁻. A more detailed analysis of the effect of tetraethylammonium (TEA⁺) shows that 10 and 20 mM TEA⁺ inhibit the K⁺ conductance in C. wailesii only by up to about 20% but predominantly via a K⁺ outward rectifier. Received: 23 December 1998 / Revised version: 1 June 1999 / Accepted: 1 June 1999     

An investigation of the dynamics of spermine bound to duplex and quadruplex DNA by <Superscript>13</Superscript>C NMR spectroscopy

A detailed analysis of the ¹³C relaxation of ¹³C-labelled spermine bound to duplex and quadruplex DNA is presented. T₁, T₂ and heteronuclear NOE data were collected at four ¹³C frequencies (75.4, 125.7, 150.9 and 201.2 MHz). The data were analyzed in terms of a frequency-dependent order parameter, S ²(ω), to estimate the generalized order parameter and the contributions to the relaxation from different motional frequencies in the picosecond–nanosecond timescale and from any exchange processes that may be occurring on the microsecond–millisecond timescale. The relaxation data was surprisingly similar for spermine bound to two different duplexes and a linear parallel quadruplex. Analysis of the relaxation data from these complexes confirmed the conclusions of previous studies that the dominant motion of spermine is independent of the macroscopic tumbling of the DNA and has an effective correlation time of ∼50 ps. In contrast, spermine bound to a folded antiparallel quadruplex had faster relaxation rates, especially R ₂. As with the other complexes, a fast internal motion of the order of 50 ps makes a substantial contribution to the relaxation. The generalized order parameter for spermine bound to duplex DNA and the linear quadruplex is small but is larger for spermine bound to the folded quadruplex. In the latter case, there is evidence for exchange between at least two populations of spermine occurring on the microsecond–millisecond timescale. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

On Relative Precision in Stratified Sampling for a Mean with Fixed Cost

This paper considers the extrema of the relative precision (R. P.), or the ratio of the variance of optimum allocation to that of proportional allocation in stratified sampling for a mean (=Y) or total (=Y) when there is a fixed cost C=c_o+σn_hc_h over L strata. The upper bound of R. P. is shown to be equal to unity, though the lower bound can be negative while approaching zero if the finite population size N→∞. Numerical results (cf. Table) when the upper bound of R. P. = 1 is attained are included for the cases L≦4 with various combinations of {W_h}, {c_h} and a fixed cost C; these give the appropriate sample sizes (=n_p) determined for proportional sampling in the cases of finite populations of sizes N=5000, 10,000 respectively.     

Ein automatentheoretischer Zugang zu Lernprozessen

Summary Learning processes are described in the behavioristic black-box conception by relations between input and output variables. In this sense it is shown, how linear learning models (Bush and Mosteller [2]) may be conceived as stochastic automata. Some T-maze experiments with rats illustrate this approach.     

Integrated tank cultivation of salmonids and Gracilaria chilensis (Gracilariales,Rhodophyta)

In order to study the N and P balances in the Driss I reservoir, we measured concentrations in the water of these elements, their sedimentation rates, and their input by the river Inaouen and output through the dam. Supply and loss of N and P were calculated from samples collected every 48 hours and were for N 15 and 27. 5 mg m⁻² d⁻¹ and for P 33.6 and 1.1 mg m⁻² d⁻¹ respectively. The sedimentation rates, determined using sediment traps, were high, about 75 mg m⁻² d⁻¹ for N and 34 mg m⁻² d⁻¹ for P. Phosphate input came from point sources and was mainly in particulate form. A large fraction of the phosphate sedimented in the form of calcium bound phosphate, with some iron bound phosphate. Bioavailability was, however, low, as the particulate phosphate disappeared rapidly from the epilimnion because of the high sedimentation rate. The ratio Tot-N/Tot-P of the sedimented particulate matter varied between 0.05 and 7.74 and depended on primary production and watergate management.     

Characterization of <Emphasis Type="SmallCaps">d</Emphasis>-tagatose-3-epimerase from <Emphasis Type="Italic">Rhodobacter sphaeroides</Emphasis> that converts <Emphasis Type="SmallCaps">d</Emphasis>-fructose into <Emphasis Type="SmallCaps">d-</Emphasis>psicose

A non-characterized gene, previously proposed as the d-tagatose-3-epimerase gene from Rhodobacter sphaeroides, was cloned and expressed in Escherichia coli. Its molecular mass was estimated to be 64 kDa with two identical subunits. The enzyme specificity was highest with d-fructose and decreased for other substrates in the order: d-tagatose, d-psicose, d-ribulose, d-xylulose and d-sorbose. Its activity was maximal at pH 9 and 40°C while being enhanced by Mn²⁺. At pH 9 and 40°C, 118 g d-psicose l⁻¹ was produced from 700 g d-fructose l⁻¹ after 3 h. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Growth control in cultured 3T3 fibroblasts : III. Binding interactions of a growth inhibitory activity with target cells

A Fibroblast Growth Regulator (FGR), which inhibits the growth and division of proliferating fibroblasts, has been isolated from medium conditioned by exposure to density-inhibited 3T3 cells. The binding of this FGR to its target cells has been analysed using [³⁵S]methionine-labeled inhibitor. The key features of this interaction are (a) there are approx. 3–4 × 10⁵ binding sites/cell; (b) more ligand is bound at 37 than at 4 °C, possibly due to internalization of the molecule at the higher temperature; (c) on a per cell basis, approximately the same amount of ligand is bound at high cell density as at low cell density; (d) there is no evidence of target cell heterogeneity in the binding of the inhibitor; and (e) the binding can be inhibited partially by calf serum. The results suggest that modulation of cell surface receptor binding by serum and growth inhibitory factors may be an important feature in growth regulation.     

Equilibrium dialysis studies of polyamine binding to DNA

Binding constants and binding site sizes for the interactions of the polyamines spermine (+4), spermidine (+3), and putrecine (+2) with helical DNA have been determined as a function of ionic conditions and temperature by equilibrium dialysis using ¹⁴C-labeled polyamines. In addition, competition equilibrium dialysis has been used to determine binding parameters for the divalent cations putrescine and Mg²⁺ from the competitive effect of these ions on the binding of spermine or spermidine. In all cases, the logarithm of the binding constant (log K_obs) varies linearly with the logarithm of the monovalent salt concentration; the slopes d log K_obs/d log[NaCl] are proportional to the valence of the ligand, and values of the extrapolated binding constants at 1M NaCl obtained from the intercepts are small (of order 1–10M^?1). In those cases examined, K_obs is insensitive to temperature; the free energy of binding is predominantly entropic. Consequently, polymines as DNA-binding ligands behave analogously to the oligolysìnes investigated previously [cf. Record, Lohman & de Haseth (1976) J. Mol. Biol. 107 , 145–158; Lohman, de Haseth & Record (1980) Biochemistry 19 , 3522–3530]. The interactions of these oligocations with DNA are predominantly electrostatic and are driven by the release of thermodynamically bound electrolyte ions from the vicinity of the DNA. The extent to which these oligocations are localized at individual phosphate binding sites or delocalized on the DNA molecule is currently not known.     

A Stock Market Forecasting Model Combining Two-Directional Two-Dimensional Principal Component Analysis and Radial Basis Function Neural Network

In this paper, we propose and implement a hybrid model combining two-directional two-dimensional principal component analysis ((2D)²PCA) and a Radial Basis Function Neural Network (RBFNN) to forecast stock market behavior. First, 36 stock market technical variables are selected as the input features, and a sliding window is used to obtain the input data of the model. Next, (2D)²PCA is utilized to reduce the dimension of the data and extract its intrinsic features. Finally, an RBFNN accepts the data processed by (2D)²PCA to forecast the next day''s stock price or movement. The proposed model is used on the Shanghai stock market index, and the experiments show that the model achieves a good level of fitness. The proposed model is then compared with one that uses the traditional dimension reduction method principal component analysis (PCA) and independent component analysis (ICA). The empirical results show that the proposed model outperforms the PCA-based model, as well as alternative models based on ICA and on the multilayer perceptron.     

Effect of Supplemental Electron Donors on the Microbial Reduction of Fe(III), Sulfate, and CO2 in Coal Mining–Impacted Freshwater Lake Sediments

Abstract In acidic mining-impacted lake sediments, the microbial reduction of Fe(III) is the dominant electron-accepting process, whereas the reduction of sulfate seems to be restricted to a narrow sediment zone of elevated pH and lower amounts of total and reactive iron. To evaluate the microbial heterogeneity and the commensal interactions of the microbial community, the flow of supplemental carbon and reductant was evaluated in four different zones of the sediment in anoxic microcosms at the in situ temperature of 12°C. Substrate consumption, product formation, and the potential to reduce Fe(III) and sulfate were similar with both upper and lower sediment zones. In the upper acidic iron-rich sediment zone, the rate of Fe(II) formation 204 nmol ml⁻¹ d⁻¹ was enhanced to 833 nmol ml⁻¹ d⁻¹ and 462 nmol ml⁻¹ d⁻¹ by supplemental glucose and H₂, respectively. Supplemental lactate and acetate were not consumed under acidic conditions and decreased the rate of Fe(II) formation to 130 nmol ml⁻¹ d⁻¹ and 52 nmol ml⁻¹ d⁻¹, respectively. When the pH of the upper sediment increased above pH 5, acetate-dependent reduction of sulfate was initiated even though the pool of Fe(III) was not depleted. In deeper sediment zones with elevated pH, the rapid consumption of acetate was always coincident to a decrease in the concentration of sulfate and soluble Fe(II), indicating the formation of Fe(II) sulfides. Although the reduction of Fe(III) was still an ongoing process in deeper sediment zones, the formation of Fe(II) was only slightly enhanced by the consumption of glucose or cellobiose, but not by H₂ or acetate. H₂-utilizing acetogens seemed to be involved in the consumption of H₂. These collective results indicated (i) that the reduction of Fe(III) predominated over the reduction of sulfate as long as the sediment remained acidic and carbon-limited, and (ii) that the sulfate-reducing microbiota in this heterogeneous sediment were better adapted to the geochemical gradients present than were other neutrophilic dissimilatory Fe(III) reducers. Received: 17 February 2000; Accepted: 22 June 2000; Online Publication: 28 August 2000     

Properties of the promoter region of theT18 d (T13 c )Tla gene

The T18^d (formerlyT13 ^c) gene of BALB/c mice belongs to the category ofTla genes which is expressed by both thymocytes and TL⁺ T-cell leukemias. To elucidate the regulation of T18^d, different restriction fragments of the 5' flanking region between -457 and⁺ 146 were linked to the chloramphenicol acetyltransferase (CAT) gene and transfected into TL⁺ and TL- cells. By comparison of transiently expresssedCAT activity among cells transfected with differentCAT constructs, the results suggest that determination of TL⁺ vs TL phenotypes is located within the region -105 to - 33, and that an element essential to T18 d expression resides within the region - 33 to + 54. Putative DNA-binding factors characterizing particular cell types and displaying selective affinity for particular TI 8^d restriction fragments were identified by electrophoretic mobility shift assays with nuclear extracts (NEs). Two factors (or complexes) which bound specifically to the T18^d fragment-105 to - 33 were expressed preferentially in TL⁺ cells and thus may be involved in determining the tissue-selective expression of TI8^d. The close proximity of negative and positive cis-acting elements within the promoter region is consistent with regulation of T18^d gene expression by a variety of trans-acting factors whose production is attuned to development and differentiation. The data provided may serve as a guide to study the regulation of other categories of Tla genes that are normally silent in thymocytes but may become expressed by leukemia cells.     

Quantification of the Role of Acclimation Temperature in Temperature Tolerance of Fishes

The relative effect of acclimation temperature on temperature tolerance was estimated from a geometrical partitioning of the temperature tolerance polygon of a fish species into three distinct zones relative to four key tolerance temperatures. This approach yields a middle tolerance zone which is independent of acclimation temperature bounded by upper and lower acclimation dependent zones. Acclimation dependent and independent temperature tolerance zones can be quantified by either areal or linear methods. Both methods were applied to quantify the effect of acclimation temperature in 21 species of temperate fishes for which temperature tolerance polygons were available. Temperature tolerance polygon areas of these 21 species ranged from 468 to 1380°C² and are linearly related (r ²=0.93, p<0.001) to ultimate incipient upper lethal temperatures. Although areal and linear partitioning methods yielded similar acclimation independent and dependent tolerances, estimates from the areal method incorporates additional information concerning the shape of the temperature tolerance polygon, in particular lower and upper lethal temperature plateaus. Mean combined acclimation dependent and independent tolerance areas of these 21 species were not different, indicating that acclimation effectively doubles the temperature tolerance polygon. Mean lower acclimation dependent area was nearly three times greater than mean upper acclimation dependent area, suggesting that acclimation plays a larger role in tolerance of low rather than high temperatures. Among these 21 species, temperature tolerance of brook charr and sheepshead minnow were the least and most affected by acclimation temperature, respectively.     

A neural network model for limb trajectory formation

This paper deals with the problem of representing and generating unconstrained aiming movements of a limb by means of a neural network architecture. The network produced time trajectories of a limb from a starting posture toward targets specified by sensory stimuli. Thus the network performed a sensory-motor transformation. The experimenters trained the network using a bell-shaped velocity profile on the trajectories. This type of profile is characteristic of most movements performed by biological systems. We investigated the generalization capabilities of the network as well as its internal organization. Experiments performed during learning and on the trained network showed that: (i) the task could be learned by a three-layer sequential network; (ii) the network successfully generalized in trajectory space and adjusted the velocity profiles properly; (iii) the same task could not be learned by a linear network; (iv) after learning, the internal connections became organized into inhibitory and excitatory zones and encoded the main features of the training set; (v) the model was robust to noise on the input signals; (vi) the network exhibited attractor-dynamics properties; (vii) the network was able to solve the motorequivalence problem. A key feature of this work is the fact that the neural network was coupled to a mechanical model of a limb in which muscles are represented as springs. With this representation the model solved the problem of motor redundancy.     

Bounds on the total particle number for species governed by reaction-diffusion equations in the infinite spatial domain

Differential inequality methods are developed for establishing upper and lower bounds on the total particle numberN(t)=∫θ(x,t) d³ x associated with solutions to nonlinear reaction-diffusion equations of the form ∂θ/∂t=D∇²θ+fθ-gθ ⁿ⁺¹ , whereD(>0),n(>0),f andg are constant parameters. If finite in a neighborhood oft=0,N(t) is bounded below for allt≥0 by a certain derived function oft for equations withg≥0. An upper bound onN(t) is obtained for equations withn=1,f<0 andg<0. These results provide general preservation and extinction criteria for the total particle number.