Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Parallel Algorithms”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 289 records · Page 16Linked to original sources
PubMed1988

Bicarbonate secretion modulates ammonium absorption in rat distal colon in vivo.

Although the mammalian colon is thought to absorb large quantities of total ammonia, principally in the form of NH3, quantitative support for this hypothesis is lacking. In rat distal colon, we observed that NH3 was approximately 400 times more permeant than NH+4. In addition, colonic HCO-3 secretion influenced total ammonia (NH3 plus NH+4) absorption; that is, alteration of HCO-3 secretion caused a parallel change in total ammonia absorption. Perfusion with total ammonia also caused net HCO-3 secretion to switch to net absorption, and, in the setting of preexisting HCO-3 absorption, perfusate containing total ammonia enhanced HCO-3 absorption. These events suggest that colonic HCO-3 secretion titrates luminal NH+4 to NH3, permitting NH3 to diffuse from the lumen, while HCO-3 is titrated to carbon dioxide and also diffuses from the lumen. In support of titration of NH+4 and HCO-3, the magnitude of induced HCO-3 absorption approximated total ammonia absorption. This titration relationship suggests that, in kinetic studies, total ammonia absorption will be limited by a fixed rate of HCO-3 secretion. A model was developed that simulated these events.

Algorithms↗
PubMed2026

SAIGE-GPU: accelerating genome- and phenome-wide association studies using GPUs.

MOTIVATION: Genome-wide association studies (GWAS) at biobank scale are computationally intensive, especially for admixed populations requiring robust statistical models. SAIGE is a widely used method for generalized linear mixed-model GWAS but is limited by its CPU-based implementation, making phenome-wide association studies impractical for many research groups. RESULTS: We developed SAIGE-GPU, a GPU-accelerated version of SAIGE that replaces CPU-intensive matrix operations with GPU-optimized kernels. The core innovation is distributing genetic relationship matrix calculations across GPUs and communication layers. Applied to 2068 phenotypes from 635 969 participants in the Million Veteran Program, including diverse and admixed populations, SAIGE-GPU achieved a 5-fold speedup in mixed model fitting on supercomputing infrastructure and cloud platforms. We further optimized the variant association testing step through multi-core and multi-trait parallelization. Deployed on Google Cloud Platform and Azure, the method provided substantial cost and time savings. AVAILABILITY AND IMPLEMENTATION: Source code and binaries are available for download at https://github.com/saigegit/SAIGE/tree/SAIGE-GPU-1.3.3. A code snapshot is archived at Zenodo for reproducibility (DOI: [10.5281/zenodo.17642591]). SAIGE-GPU is available in a containerized format for use across HPC and cloud environments and is implemented in R/C++ and runs on Linux systems.

Genome-Wide Association Study↗
PubMed1988

Inhibition of porcine pepsin by two substrate analogues containing statine. The effect of histidine at the P2 subsite on the inhibition of aspartic proteinases.

Two new inhibitors, 4 and 5, of the aspartic proteinase porcine pepsin were synthesized. These compounds, which span the P4-P'3 binding subsites of the enzyme, were derived by replacing the Nph-Phe dipeptidyl unit of a good pepsin substrate, H2N-Phe-Gly-His-Nph-Phe-Ala-Phe-OMe (3), with statine [(3S,4S)-4-amino-3-hydroxy-6-methylheptanoic acid, Sta]. Hexapeptide 5, H2N-Phe-Gly-Val-(S,S)-Sta-Ala-Phe-OMe, is an extremely potent inhibitor of pepsin with a Ki value less than 1 nM. This result is consistent with the proposal that statine functions as a bioisosteric replacement for a substrate dipeptidyl unit. Compound 4, which contains His at P2, is 2 orders of magnitude less active than the valine analogue 5 (Ki = 150 nM). The factor for the decrease in binding to pepsin effected by replacement of Val by His at P2 parallels the ratio of protonated vs unprotonated imidazole group in peptide 4 at pH 4, according to the Henderson-Hasselbach equation. This result suggests that a positively charged side chain at P2 is undesirable for maximum pepsin inhibition. Kinetic constants for several known inhibitors of pepsin and renin are presented that demonstrate that the effect of His incorporation at P2 on pepsin inhibition depends upon the peptide sequence and that the effect is considerably different for renin inhibitors. We further suggest that the high selectivity of potent renin inhibitors known to be only weak pepsin and cathepsin D inhibitors is due in part to the extent of histidine protonation at P2 arising from pH differences in the inhibition kinetics assay of renin (neutral conditions) compared to other aspartic proteinases (acid pH 2-4).

Algorithms↗
PubMed1987

A model for the spatio-temporal organization of DNA replication in mammalian cells.

The spatio-temporal organization of chromosomal DNA replication was analyzed using a model based on a "DNA unit" (or decondensation unit) hypothesis. The model is an extension of the fork movement theory of Huberman & Riggs (1968) and can account for a partially deterministic and partially stochastic order of DNA replication in chromosomes. It presumes that each chromosome is composed of DNA units that are arranged in sequence and that are replicated in parallel. A deterministic wave of chromatin decondensation propagates along the DNA unit continuously and progressively providing a field for the random activation of replication origin. Assignment of replication times to DNA compartments by a Monte Carlo method was programmed based on the model and the program was used to stimulate DNA synthesis rate curves that can be measured by the method of Dolbeare et al. (1983, 1985). The shape of the curve is shown to constrain possible parameter values of the model, which include the rate of fork movement, the fraction of chromatin that is decondensed at the start of S-phase, the initial number of origins activated, the rate at which new origins are activated, etc. The chromosomal organization that controls the molecular level of DNA replication is briefly reviewed and its relevance to the model is also discussed.

Algorithms↗
PubMed1983

Conformation of glucagon in a lipid-water interphase by 1H nuclear magnetic resonance.

A determination of the spatial structure of the polypeptide hormone glucagon bound to perdeuterated dodecylphosphocholine micelles is described. A map of distance constraints between individually assigned hydrogen atoms of the polypeptide chain was obtained from two-dimensional nuclear Overhauser enhancement spectroscopy. These data were used as the input for a distance geometry algorithm for computing conformations that would be compatible with the experiments. In the region from residues 5 to 29 the mobility of the polypeptide backbone and most of the amino acid side-chains was found to be essentially restricted to the overall rotational tumbling of the micelles. The secondary structure in this region includes three turns of irregular alpha-helix in the segment of residues 17 to 29 near the C terminus, a stretch of extended polypeptide chain from residues 14 to 17, an alpha-helix-like turn formed by the residues 10 to 14 and another extended region from residues 5 to 10. In the N-terminal tetrapeptide H-His-Ser-Gln-Gly- the two terminal residues are highly mobile, indicating that they extend into the aqueous phase, and the mobility of the residues Gln3 and Gly4 appears to be only partially restricted by the binding to the micelle. The absence of long range nuclear Overhauser effects between the peptide segments 5-9 and 11-29, and between 5-16 and 19-29 shows that the polypeptide chain does not fold back on itself and hence that micelle-bound glucagon does not adopt a globular tertiary structure. Previously it was shown that the polypeptide backbone of glucagon is located close to and runs roughly parallel to the micelle surface. Combination of these observations suggests that the overall spatial arrangement of the glucagon polypeptide chain in a lipid-water interphase is largely determined by the topology of the lipid support, in the present case the curvature of the dodecylphosphocholine micelles. The tertiary structure is further characterized by the formation of two hydrophobic patches by the side-chains of Phe6, Tyr10 and Leu14, and the side-chains of Ala19, Phe22, Val23, Trp25 and Leu26, respectively.

Amino Acid Sequence↗
PubMed1990

Bifurcation analysis of nonlinear retinal horizontal cell models. II. Network properties.

1. We have previously presented a model of horizontal-cell soma isolated from fish retina. The model consists of a synaptic conductance representing input from photoreceptors in parallel with voltage-dependent membrane currents. Membrane-current models are based on I-V curves measured in isolated fish horizontal cells. Bifurcation theory was used to analyze model properties. The major findings of this study were 1) the inward Ca2+ current must be inactivated to account for horizontal-cell resting potentials and hyperpolarizing responses to light stimuli in a background of dark, and 2) the synaptic conductance controls the bifurcation structure of the model, with bistable behavior occurring at small and monostable behavior occurring at larger values of the synaptic conductance. The synaptic conductance at the point of transition from bistable to monostable behavior corresponds to the activation of as few as 100 synaptic channels. Thus tonic synaptic input from photoreceptors and inactivation of the inward Ca2+ current act to "linearize" responses of isolated horizontal-cell models. 2. The model described in this paper extends these analyses to large networks of horizontal cells in which each cell is coupled resistively to its nearest neighbors and is modeled with the use of the full complement of nonlinear membrane currents. Network responses to arbitrary patterns of conductance change (simulating inputs from photoreceptors), current-, or voltage-clamp stimuli are computed using the Newton iteration. The Newton descent direction is computed using either conjugate gradient (CG) or preconditioned CG algorithms. 3. An analysis of network stability properties is performed. Network I-V curves are computed by voltage-clamping the center node and computing the current required to maintain the clamp voltage. Computations are performed on networks of model cells in which the Ca2+ current is fully activated and the synaptic conductance is zero, thus making each cell as nonlinear as possible. Coupling conductance values slightly greater than 100 pS provide a current shunt sufficient to prevent the generation of Ca2+ action potentials in the network. This coupling conductance corresponds to the conductance of as few as two gap-junction channels and is more than two orders of magnitude less than the coupling known to exist between pairs of cultured horizontal cells.(ABSTRACT TRUNCATED AT 400 WORDS)

Animals↗
PubMed1989

A parallel implementation of the ALOPEX process.

Optimization techniques have found many applications in science, engineering, and industry. In all applications, the best value of a "cost function" is sought in a well-defined domain; this cost function in general depends on many parameters. An iterative optimization technique has been developed (ALOPEX) that uses feedback in order to optimize the response of a system. The cost function for this process is problem dependent and therefore quite flexible. The method has been applied successfully to different optimization problems such as pattern recognition, receptive field studies in the visual system of animals, curve fitting, etc. We present two special purpose hardware implementations for ALOPEX. The first method takes time O(logN + logm) and uses O(mN2) processing elements. The second method takes O(logN + m) time and uses O(N2) processing elements. Our basic architecture is a binary tree with N2 leaves (equal to the length of the vectors) and therefore had depth O(logN). Different implications of the two approaches will be discussed including similarities with the biological visual process.

Algorithms↗
PubMed1992

Droop: a rapidly computable descriptor of local minimum tissue temperature during conductive interstitial hyperthermia.

Although the goal of local hyperthermia therapy for cancer is to elevate the temperature of a tumour to cytotoxic levels, without the presence of 'cold spots', varying blood flow has made the achievement of consistent, therapeutic temperature distributions extraordinarily difficult. The paper presents a novel approach to estimating local minimum tumour temperatures during conductive interstitial hyperthermia which facilitates identification and elimination of cold spots. Conductive interstitial hyperthermia is modelled mathematically for a parallel array of implanted, electrically heated catheters which warms the treated tissue by thermal conduction and blood perfusion. Computer simulations employing the bioheat transfer equation reveal a predictive relationship between implanted catheter temperature, catheter power, implantation geometry and local minimum tumour temperature. Formulation of this relationship in terms of a parameter named 'droop' allows estimation of local minimum intratumoural temperatures from individual catheter temperature and power. Computer simulations are also performed to determine the sensitivity of the droop-based estimator to variations in properties of the tissue and catheters. Generally, variations in geometry or thermal properties of about 10 per cent cause estimation errors of less than 1 degree C in magnitude. These results suggest that online estimates of thermal 'droop' may provide a practical route to more consistent control of intratumoural minimum temperature during conductive interstitial heat therapy.

Algorithms↗
PubMed1990

Energetics of the structure and chain tilting of antiparallel beta-barrels in proteins.

The preferred structural pattern of antiparallel beta-barrels in proteins, described as the right-handed tilting of the peptide strands with respect to the axis of the barrel, is accounted for in terms of intra- and interchain interaction energies. It is related to the preference of beta-sheets for right-handed twisting. Conformational energy computations have been carried out on three eight-stranded antiparallel beta-barrels composed of six-residue strands, in which L-Val and Gly alternate, and having a right-handed, a left-handed, or no tilt. After energy minimization, the relative energies of these structures were 0.0, 8.6, and 46.1 kcal/mol, respectively; i.e., the right-tilted beta-barrel is favored energetically, in agreement with anti-parallel beta-barrels observed in proteins. Tilting of the barrel is favored, relative to the nontilted structure, by both intra- and interstrand interactions, because tilting allows better packing of the bulky side chains. On the other hand, the energy difference between the left- and right-tilted barrels arises essentially from intrachain interactions. This is a consequence of the preference of beta-sheets for a right-handed twist. Space limitations inside the barrel are satisfied if there is an alternation of bulky residues and residues with small or no side chain (preferably Gly) in neighboring positions on adjacent strands. Such a pattern is seen frequently in antiparallel beta-barrels of globular proteins. The computations indicate that a structure with Val...Gly pairs can be accommodated in a beta-barrel with no distortion.

Algorithms↗
PubMed1989

Three-dimensional image reconstruction from complete projections.

Three-dimensional medical image reconstruction for both transmission and emission tomography has traditionally decomposed the problem into a set of two-dimensional reconstructions on parallel transverse sections. There is, however, increasing interest in reconstructing projection data directly in three dimensions. For emission tomography in particular, such a reconstruction procedure would clearly make more efficient use of the available photon flux. In the past few years, a number of authors have studied the problems associated with full three-dimensional reconstruction, especially in the case of positron tomography where three-dimensional reconstruction is likely to offer the greatest benefits. While most approaches follow that of filtered backprojection, the relationship between the various filters that have been proposed is far from evident. This paper clarifies this relationship by analysing and generalising the different classes of published filters and establishes the properties and characteristics of a general solution to the three-dimensional reconstruction problem. Some guidelines are suggested for the choice of an appropriate filter in a given situation.

Algorithms↗
PubMed1995

The effect of inhomogeneous sample susceptibility on measured diffusion anisotropy using NMR imaging.

Water diffusion measurements in white matter of freshly excised pig spinal cord and in parenchyma of fresh celery (excluding the fibers along the edge of the stalk) were performed using NMR at 200 MHz. In white matter of pig spinal cord, the measured diffusion coefficient is anisotropic and independent of sample orientation with respect to the magnetic field. In celery parenchyma, diffusion is isotropic and independent of orientation in the magnetic field when using a diffusion sequence that gives results independent of self-induced magnetic-field gradients. However, when the standard diffusion pulse sequence that gives results dependent upon self-induced magnetic-field gradients is used, diffusion in celery appears isotropic when the stalk is oriented parallel to the magnetic field but anisotropic when oriented perpendicular. Susceptibility variations leading to anisotropic self-induced magnetic-field gradients approximately 3 kHz/cm in magnitude when the celery is oriented perpendicular to the magnetic field can explain this apparent anisotropic diffusion. A study of the apparent diffusion coefficient (ADC) in celery as a function of diffusion times ranging from 8 to 22 ms indicates that the motion is at most only slightly restricted. Therefore, although the effect is not seen in all types of samples, one must be aware that self-induced gradients may affect the ADC and may cause isotropic diffusion to appear anisotropic. In addition, NMR experiments that change diffusion-sensitizing gradient timings to study restricted diffusion change the effects of the self-induced gradients as well as the effect of barriers on the ADC, complicating interpretation.

Algorithms↗
PubMed1988

Effect of Ca2+ on cross-bridge turnover kinetics in skinned single rabbit psoas fibers: implications for regulation of muscle contraction.

The effect of Ca2+ upon the rate constant of force redevelopment following a period of isotonic shortening with immediate restretch to the starting sarcomere length was studied in rabbit psoas fibers at 5 degrees C. Control experiments support the assumption that the rate constant of force redevelopment represents isometric cross-bridge turnover kinetics (fapp + gapp), where fapp and gapp are the rate constants characterizing the transitions from the non-force-generating states to the force-generating states and back to the non-force-generating states, respectively. Parallel measurements of the rate constant of force redevelopment and of force, stiffness, and fiber ATPase during isometric contraction allow the effect of Ca2+ upon fapp and gapp to be determined. Analysis reveals that Ca2+ has a marked effect upon fapp, while gapp remains approximately unchanged. Furthermore, in the range above 25-30% of maximum Ca2+ activation, regulation of force, stiffness, and ATPase is mediated through changes in fapp. Below this range, however, it cannot be ruled out that, in addition, cross-bridges are also switched in and out of the turnover process ("recruitment"). As a consequence of regulation through turnover kinetics, both Ca2+ sensitivity and the slope of force-pCa (-log[Ca2+]) relations are shown to be affected by the ratio fapp/gapp, which may represent an important mechanism of modulation of contractile function in addition to modulation through changes within the regulatory protein system.

Adenosine Triphosphatases↗
PubMed2025

Columba: fast approximate pattern matching with optimized search schemes.

MOTIVATION: Aligning sequencing reads to reference genomes is a fundamental task in bioinformatics. Aligners can be classified as lossy or lossless: lossy aligners prioritize speed by reporting only one or a few high-scoring alignments, whereas lossless aligners output all optimal alignments, ensuring completeness and sensitivity. RESULTS: This paper introduces Columba, a high-performance lossless aligner tailored for Illumina sequencing data. Columba processes single or paired-end reads in FASTQ format and outputs alignments in SAM format. By utilizing advanced search schemes and bit-parallel alignment techniques, Columba achieves exceptional speed. Columba is available in two variants. The first, based on the bidirectional FM-index, prioritizes speed. The second, Columba RLC, uses run-length compression using a bidirectional move structure, significantly reducing memory usage for large, repetitive datasets like pan-genomes. Benchmarks on the human genome, as well as bacterial and human pan-genome datasets, demonstrate that Columba is much faster than existing lossless aligners and even competitive with lossy tools. We integrated Columba into the OptiType HLA genotyping pipeline, where it substantially reduced computational time while maintaining accuracy. These results position Columba as a versatile, state-of-the-art tool for high-sensitivity genomic analyses. AVAILABILITY AND IMPLEMENTATION: The source code of Columba is available at https://github.com/biointec/columba under AGPL license. Scripts to reproduce the benchmarks and analyses are available at https://doi.org/10.5281/zenodo.15849246.

Software↗
PubMed1989

3-D superposition for radiotherapy treatment planning using fast Fourier transforms.

Currently used radiotherapy treatment planning algorithms based on effective path length or scatter function methods do not model electron ranging from photon interaction sites. The superposition (or convolution) technique does model this effect, which is especially important at higher (linear accelerator) energies since the electron range is significant. Another advantage of this method is that it is conceptually simple and models the physical processes directly, rather than using empirically derived methods. A major disadvantage of superposition lies in the large amount of computer time required to generate a plan, especially in three dimensions. To help solve this problem, superposition using an invariant dose spread array (kernel) can be achieved by performing a convolution in Fourier space using fast Fourier transforms (FFTs). A method for 3 dimensional calculation of dose using FFTs is presented. Dose spread arrays are calculated using the EGS Monte Carlo code, and convolved with the TERMA (total energy released per unit mass). In both cases a 10 MV nominal beam energy is modelled by a 10 component spectrum, which is compared to the result obtained using monochromatic energy only (3.0 MeV at the surface). The FFT technique is shown to be significantly faster than standard convolution for medium to large TERMA and dose spread array sizes. The method is shown to be highly accurate for small fields in homogeneous media. For larger fields the central axis depth dose is accurate but the profile shape in the penumbral region becomes slightly distorted. This is because photons incident near the beam edges are not parallel to the cartesian coordinate system used as the convolution framework. However, this effect is sufficiently small to indicate that the convolution method is suitable for use in routine treatment planning.

Fourier Analysis↗
PubMed2026

PanForest: predicting genes in genomes using random forests.

MOTIVATION: The presence or absence of some genes in a genome can influence whether other genes are likely to be present or absent. Understanding these gene co-occurrence and avoidance patterns reveals fundamental principles of genome organization, with applications ranging from evolutionary reconstruction to rational design of synthetic genomes. RESULTS: PanForest, presented here, uses random forest classifiers to predict the presence and absence of genes in genomes from the set of other genes present. Performance statistics output by PanForest reveal how predictable each gene's presence or absence is, based on the presence or absence of other genes in the genome. Further, PanForest produces statistics indicating the importance of each gene in predicting the presence or absence of each other gene. The PanForest software can run serially or in parallel, thereby facilitating the analysis of pangenomes at Network of Life scale.A pangenome of 12 741 accessory genes in 1000 Escherichia coli genomes was analysed in around 5 h using eight processors. To demonstrate PanForest's utility, we present a case study and show that certain genes associated with resistance to antimicrobial drugs reliably predict the presence or absence of other genes associated with resistance to the same drug. Further, we highlight several associations between those genes and others not known to be associated with antimicrobial resistance (AMR), or associated with resistance to other drugs. We envisage PanForest's use in studies from multiple disciplines concerning the dynamics of gene distributions in pangenomes ranging from biomedical science and synthetic biology to molecular ecology. AVAILABILITY AND IMPLEMENTATION: The software if freely available with a full manual and can be found with at www.github.com/alanbeavan/PanForest DOI: https://doi.org/10.5281/zenodo.17865482.

Software↗
PubMed1992

Adaptive rate pacing controlled by the right ventricular preejection interval: clinical experience with a physiological pacing system.

In the Precept pacing system, the right ventricular intracardiac impedance waveform is used to evaluate either of two indicators of metabolic demand relative right ventricular stroke volume and preejection interval (PEI). PEI is known to reliably parallel contractility changes, which is reflective of physical and emotional stress. The stability and dynamic behavior of PEI were tested in ten patients with a Precept pacing system under various forms of exercise and during postural changes. Although significant patient-to-patient variability of the sensor values was observed, reflecting individual physiological differences, the chronic stability of PEI was excellent in the total device experience of 147 months. In all patients, PEI shortened significantly during bicycle ergometry from a mean value of 137.7 +/- 17.8 (range 96-162) to a mean value of 103.0 +/- 21.6 (range 92-109) (P less than 0.05). Low level bicycle exercise of short duration resulted in a prompt decrease in PEI and increase in pacing rate in all patients. There were no uniform postural responses overall, although some posture related rate changes were observed in two patients. We conclude that the first generation of a PEI based pacing system holds promise for adaptive rate pacing.

Aged↗
PubMed1992

Artificial neural network classification of Drosophila courtship song mutants.

Courtship songs produced by Drosophila males--wild-type, plus the cacophony and dissonance behavioral mutants--were examined with the aid of newly developed strategies for adaptive acoustic analysis and classification. This system used several techniques involving artificial neural networks (a.k.a. parallel distributed processing), including learned vector quantization of signals and non-linear adaption (back-propagation) of data analysis. "Pulse" song from several individual wild-type and mutant males were first vector-quantized according to their frequency spectra. The accumulated quantized data of this kind, for a given song, were then used to "teach" or adapt a multiple-layered feedforward artificial neural network, which classified that song according to its original genotype. Results are presented on the performance of the final adapted system when faced with novel test data and on acoustic features the system decides upon for predicting the song-mutant genotype in question. The potential applications and extensions of this new system are discussed, including how it could be used to screen for courtship mutants, search novel behavior patterns or cause-and-effect relationships associated with reproduction, compress these kinds of data for digital storage, and analyze Drosophila behavior beyond the case of courtship song.

Algorithms↗
PubMedDate not supplied

Superposition on a multicomputer system.

Superposition (convolution using a noninvariant kernel) has been shown to be a highly promising technique for use in calculating dose distributions in radiotherapy treatment planning. However, one major difficulty that currently prevents use in routine planning is the computational effort required to perform the calculation in three dimensions. To help solve this problem the superposition technique has been implemented on a parallel processor multicomputer in order to examine the performance characteristics of such a system. Up to eight elements have been connected in a pipeline (linear array), and tree networks of three and seven processors have also been constructed (using INMOS T800 transputers). The significant results obtained with these networks are: (1) Both topologies provide near-linear speedup with increasing processor number (8 processors provide 7.81 times the computing power of a single processor when using an optimal communication packet size); (2) increasing communication packet size from 1 voxel to an optimum of approximately 40 voxels significantly reduces communication overhead per processor. Overhead per processor for a 7-element linear array is 6.9% when using 1-voxel packets, but only 1.8% when using 40-voxel packets; (3) the topology of the network has some effect on communication overhead: Arranging 7 processors in a 1-2-4 binary tree reduces overhead to 80.1% of that encountered using a 7-element linear array (with packet size of 1 voxel).

Algorithms↗