Reduction Of Several Orders Research Articles

Abstract 7410: PyVIPER: A fast and scalable toolkit for the identification of dysregulated proteins in tumor-derived single-cell RNA-seq data

Abstract Single-cell RNA sequencing (scRNA-seq) enabled the characterization of tumor microenvironment heterogeneity with unprecedented resolution. However, its low signal-to-noise ratio due to data sparsity hinders biomarkers quantitative measurements. To perform a more robust assessments of protein markers, we have previously introduced the use of the VIPER algorithm that estimates regulatory proteins activity using the expression of their downstream targets by leveraging lineage specific regulatory networks. We have shown that this approach compares favorably with both gene expression and protein abundance measurements to identify novel cell states corresponding to tumor subpopulations that are undetectable at the gene expression level, by inferring the mechanistic contribution of regulatory and signaling proteins to cellular phenotypes. VIPER-inferred cell states have been linked to critical functional roles, including immune-evasion and progression, disease heterogeneity and drug sensitivity in cancer. However, the ever-growing size of scRNA-seq datasets demands for scalable tools that simultaneously analyze tens or hundreds of thousands of cells in tumor contexts, such as tumor atlases and large clinical cohorts. We present PyVIPER, a fast, memory-efficient, and highly scalable Python framework that allows protein activity inference of large scRNA-seq datasets. PyVIPER is fully integrated with both AnnData and Scanpy frameworks for single-cell analysis and enables a seamless interface with cutting-edge machine learning libraries (e.g., sckit-learn and TensorFlow). PyVIPER provides multiple algorithms for enrichment analysis, novel modules for postprocessing and pathway analysis, and a set of procedures tailored for precision oncology studies in single cells, including OncoMatch, a validated platform to assess the fidelity of models (in vitro or animal) with human tumors for drug sensitivity elucidation. We benchmarked PyVIPER against several scRNA-seq cancer datasets obtaining runtimes reduction of several orders of magnitude (from hours to minutes) when compared to previous VIPER implementations. Taken together, PyVIPER scalability and its embedded toolkit for cancer data analysis address the need to efficiently extend protein activity analyses to large single-cell datasets, with the goal of providing cancer researchers with an effective platform to identify aberrant molecular mechanisms in cancer cells. Citation Format: Luca Zanella, Alexander L. Wang, Zizhao J. Lin, Lukas J. Vlahos, Miquel Anglada Girotto, Aziz Zafar, Andrea Califano, Alessandro Vasciaveo. PyVIPER: A fast and scalable toolkit for the identification of dysregulated proteins in tumor-derived single-cell RNA-seq data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7410.

Physics-driven proper orthogonal decomposition: A simulation methodology for partial differential equations

A simulation methodology derived from a learning algorithm based on Proper Orthogonal Decomposition (POD) is presented to solve partial differential equations (PDEs) for physical problems of interest. Using the developed methodology, a physical problem of interest is projected onto a functional space described by a set of basis functions (or POD modes) that are trained via the POD by solution data collected from direct numerical simulations (DNSs) of the PDE. The Galerkin projection of the PDE is then performed to account for physical principles guided by the PDE. The procedure to construct the physics-driven POD-Galerkin simulation methodology is presented in detail, together with demonstrations of POD-Galerkin simulations of dynamic thermal analysis on a microprocessor and the Schrödinger equation for a quantum nanostructure. The physics-driven methodology allows a reduction of several orders in degrees of freedom (DoF) while maintaining high accuracy. This leads to a drastic decrease in computational effort when compared with DNS. The major steps for implementing the methodology include:•Solution data collection from DNSs of the physical problem subjected to parametric variations of the system.•Calculations of POD modes and eigenvalues from the collected data using the method of snapshots.•Galerkin projection of the governing equation onto the POD space to derive the model.

TreeMerge: Efficient Generation of Minimal Hitting-Sets for Conflict Sets in Tree Structure for Model-Based Fault Diagnosis

For many high-tech fields such as space exploration, nuclear technology, and smart automobiles, it is vital to timely find faulty components of man-made devices to ensure safety. However, there is nearly no enough diagnostic experience accumulated in these new devices, and thus, it is hardly suitable to only apply the traditional expert/experience-based fault diagnosis approach. Thus, model-based diagnosis was proposed for efficient detection of faulty components; this approach explores the behavioral and structural information of the device to be diagnosed, and no experience is required. In model-based diagnosis, for a device to be diagnosed, minimal conflict sets of components are first generated, and all minimal hitting-sets for them will be derived as candidate diagnoses. Therefore, it is vital to efficiently generate all minimal hitting-sets to find the final diagnosis. Unfortunately, it is proven to be NP-hard when deriving all minimal hitting-sets for given minimal conflict sets. To improve the computing efficiency, in this article, we propose a novel approach called TreeMerge , which considers a special type of tree structure of minimal conflict sets of large sizes since structural information usually plays an important role in solving complex problems. Theoretically, compared with other algorithms, the time complexity of the new algorithm is greatly reduced, as the time complexity of the new algorithm becomes linear rather than quadratic . Furthermore, experimental results on multiple synthetic and benchmark examples show that the proposed TreeMerge algorithm is more efficient than many other state-of-the-art methods, with a reduction of several orders of magnitude runtime (seconds).

Symmetric Helmholtz Fermi-surface harmonics for an optimal representation of anisotropic quantities on the Fermi surface: Application to the electron-phonon problem

We outline a numerical procedure to incorporate the crystal symmetries in the Helmholtz Fermi-surface harmonics basis set, which are the solutions of the Helmholtz equation defined on the Fermi surface. This improvement allows for an optimal representation of anisotropic quantities defined on the Fermi surface in terms of few symmetric elements of the set. We demonstrate the general validity of our approach by identifying the fully symmetric Helmholtz Fermi-surface harmonics subset for several representative systems with different crystal structures, namely, FCC-Cu, HEX-MgB$_2$, and BCC-YH$_6$. Furthermore, we illustrate the potential of the method applied to the electron-phonon problem, showing that the anisotropic electron-phonon mass-enhancement parameter $\lambda_{\bf k}$ can be represented to high accuracy by a handful of coefficients. This works as an effective filter, paving the way for a reduction of several orders of magnitude in the computation of superconductivity, impurity problems, or any other Fermi surface dependent property of metals from first principles.

A microscopically motivated model for the swelling-induced drastic softening of hydrogen-bond dominated biopolymer networks

The penetration of water into rubber-like protein networks such as cross-linked resilin, which is found in insects, can lead to changes in stiffness that range over several orders of magnitude. This softening effect cannot be explained by the volumetric changes associated with pure swelling/deswelling used to describe networks with covalent bonds. Rather, this property stems from the reversible swelling-induced breaking of hydrogen cross-linking bonds that connect the chains in the network. This work presents a model for the swelling and the mechanical response of hydrogen-bond dominated biopolymer networks. It is shown that the penetration of water molecules into the network leads to the breaking of non-covalent cross-linking sites. In turn, the network experiences a reduction in the effective chain-density, an increase in entropy, and a consequent decrease in free energy, thus explaining the dramatic softening. Additionally, the breaking of hydrogen bonds alters the micro-structure and changes the quantitative elastic behavior of the network. The proposed model is found to be in excellent agreement with several experimental findings. The merit of the work is twofold in that it (1) accounts for the number and the strength of non-covalent cross-linking bonds, thus explaining the drastic reduction in stiffness upon water uptake, and (2) provides a method to characterize the micro-structural evolution of hydrogen-bond dominated networks. Consequently, the model can be used as a micro-structural design-guide to program the response of synthetic polymers. Statement of SignificanceHydrogen-bond dominated biopolymer networks are found in insects and have a unique structure that allows a dramatic reduction of several orders of magnitude in stiffness upon hydration. Understanding the micro-structure of such networks is key in the fabrication of new biomimetic polymers with tunable mechanical properties. This work introduces a microscopically motivated model that explains the dramatic reduction in stiffness and quantifies the influence of key micro-structural quantities on the overall response. The model is validated through several experimental findings. The insights from this work motivate further attempts at the fabrication of new biomimetic polymers and serve as a micro-structural design guide that enables the programming of the elastic swelling-induced response.

Parametric generation of conditional geological realizations using generative neural networks

Deep learning techniques are increasingly being considered for geological applications where—much like in computer vision—the challenges are characterized by high-dimensional spatial data dominated by multipoint statistics. In particular, a novel technique called generative adversarial networks has been recently studied for geological parametrization and synthesis, obtaining very impressive results that are at least qualitatively competitive with previous methods. The method obtains a neural network parametrization of the geology—so-called a generator—that is capable of reproducing very complex geological patterns with dimensionality reduction of several orders of magnitude. Subsequent works have addressed the conditioning task, i.e., using the generator to generate realizations honoring spatial observations (hard data). The current approaches, however, do not provide a parametrization of the conditional generation process. In this work, we propose a method to obtain a parametrization for direct generation of conditional realizations. The main idea is to simply extend the existing generator network by stacking a second inference network that learns to perform the conditioning. This inference network is a neural network trained to sample a posterior distribution derived using a Bayesian formulation of the conditioning task. The resulting extended neural network thus provides the conditional parametrization. Our method is assessed on a benchmark image of binary channelized subsurface, obtaining very promising results for a wide variety of conditioning configurations.

Experimental refutation of the deuterium permeability in vanadium, niobium and tantalum

Deuterium permeability through Nb, V, and Ta was measured in the temperature range between ∼300-500 °C and 100–1000 mbar. For the three materials, the permeation rate at 500 °C was between two and torders of magnitude below expected values which were obtained by combining diffusivity and solubility data and which have been traditionally accepted due to the insufficient experimental available data. Comparison with existing literature suggests a complex dependence of permeability with temperature, which helps explain this discrepancy. Given the important role of the surface condition on the permeability, additional deuterium permeation tests were carried out in vanadium for more realistic sample preparation. Results indicate that native oxides formed at room temperature introduce a permeation reduction of several orders of magnitude, and must be considered for practical applications.

Radiative heat transfer between core-shell nanoparticles

Radiative heat transfer in systems with core-shell nanoparticles may exhibit not only a combination of disparate physical properties of its components but also further enhanced properties that arise from the synergistic properties of the core and shell components. We study the thermal conductance between two core-shell nanoparticles (CSNPs). The contribution of electric and magnetic dipole moments to the thermal conductance depend sensitively on the core and shell materials, and adjustable by core size and shell thickness. We predict that the radiative heat transfer in a dimer of Au@SiO2 CSNPs (i.e., silica-coated gold nanoparticles) could be enhanced several order of magnitude compared to bare Au nanoparticles. However, the reduction of several orders of magnitude in the heat transfer is possible between SiO2@Au CSNPs (i.e., silica as a core and gold as a shell) than that of uncoated SiO2 nanoparticles.

Parallel Algorithms for Operations on Multi-Valued Decision Diagrams

Multi-valued Decision Diagrams (MDDs) have been extensively studied in the last ten years. Recently, efficient algorithms implementing operators such as reduction, union, intersection, difference, etc., have been designed. They directly deal with the graph structure of the MDD and a time reduction of several orders of magnitude in comparison to other existing algorithms have been observed. These operators have permitted a new look at MDDs, because extremely large MDDs can finally be manipulated as shown by the models used to solve complex application in music generation. However, MDDs become so large (50GB) that minutes are sometimes required to perform some operations. In order to accelerate the manipulation of MDDs, parallel algorithms are required. In this paper, we introduce such algorithms. We carefully design them in order to overcome inherent difficulties of the parallelization of sequential algorithms such as data dependencies, software lock-out, false sharing, or load balancing. As a result, we observe a speed-up , i.e. ratio between parallel and sequential runtimes, growing linearly with the number of cores.

Tracking at High Level Trigger in CMS

The trigger systems of the LHC detectors play a crucial role in determining the physics capabilities of experiments. A reduction of several orders of magnitude of the event rate is needed to reach values compatible with detector readout, offline storage and analysis capability. The CMS experiment has been designed with a two-level trigger system: the Level-1 Trigger (L1T), implemented on custom-designed electronics, and the High Level Trigger (HLT), a streamlined version of the CMS offline reconstruction software running on a computer farm. A software trigger system requires a trade-off between the complexity of the algorithms, the sustainable output rate, and the selection efficiency. With the computing power available during the 2012 data taking the maximum reconstruction time at HLT was about 200 ms per event, at the nominal L1T rate of 100 kHz. Track reconstruction algorithms are widely used in the HLT, for the reconstruction of the physics objects as well as in the identification of b-jets and lepton isolation. Reconstructed tracks are also used to distinguish the primary vertex, which identifies the hard interaction process, from the pileup ones. This task is particularly important in the LHC environment given the large number of interactions per bunch crossing: on average 25 in 2012, and expected to be around 40 in Run II. We will present the performance of HLT tracking algorithms, discussing its impact on CMS physics program, as well as new developments done towards the next data taking in 2015.

CMS Trigger Improvements Towards Run II

The trigger systems of the LHC detectors play a crucial role in determining the physics capabilities of the experiments. A reduction of several orders of magnitude of the event rate is needed to reach values compatible with detector readout, offline storage and analysis capability. The CMS experiment has been designed with a two-level trigger system: the Level-1 Trigger (L1), implemented on custom-designed electronics, and the High Level Trigger (HLT), a streamlined version of the CMS offline reconstruction software running on a computer farm. Both systems need to provide an efficient and fast selection of events, to keep the average write-out rate below 1 kHz. For Run II, the doubling of both the center of mass energy to 13 TeV and the collision rate to 40 MHz, will imply increased cross sections and out-of-time pile-up. We will present the improvements brought to both L1 and HLT strategies to meet those new challenges.

Muons in the CMS High Level Trigger System

The trigger systems of LHC detectors play a fundamental role in defining the physics capabilities of the experiments. A reduction of several orders of magnitude in the rate of collected events, with respect to the proton-proton bunch crossing rate generated by the LHC, is mandatory to cope with the limits imposed by the readout and storage system. An accurate and efficient online selection mechanism is thus required to fulfill the task keeping maximal the acceptance to physics signals. The CMS experiment operates using a two-level trigger system. Firstly a Level-1 Trigger (L1T) system, implemented using custom-designed electronics, is designed to reduce the event rate to a limit compatible to the CMS Data Acquisition (DAQ) capabilities. A High Level Trigger System (HLT) follows, aimed at further reducing the rate of collected events finally stored for analysis purposes. The latter consists of a streamlined version of the CMS offline reconstruction software and operates on a computer farm. It runs algorithms optimized to make a trade-off between computational complexity, rate reduction and high selection efficiency. With the computing power available in 2012 the maximum reconstruction time at HLT was about 200 ms per event, at the nominal L1T rate of 100 kHz. An efficient selection of muons at HLT, as well as an accurate measurement of their properties, such as transverse momentum and isolation, is fundamental for the CMS physics programme. The performance of the muon HLT for single and double muon triggers achieved in Run I will be presented. Results from new developments, aimed at improving the performance of the algorithms for the harsher scenarios of collisions per event (pile-up) and luminosity expected for Run II will also be discussed.

Performance of Tracking, b-tagging and Jet/MET reconstruction at the CMS High Level Trigger

The trigger systems of the LHC detectors play a crucial role in determining the physics capabilities of experiments. In 2015, the center-of-mass energy of proton-proton collisions will reach 13 TeV up to an unprecedented luminosity of 1 × 1034 cm−2s−1. A reduction of several orders of magnitude of the event rate is needed to reach values compatible with detector readout, offline storage and analysis capabilities. The CMS experiment has been designed with a two-level trigger system: the Level-1 Trigger (L1T), implemented on custom-designed electronics, and the High Level Trigger (HLT), a streamlined version of the offline reconstruction software running on a computer farm. A software trigger system requires a trade-off between the complexity of the algorithms, the sustainable output rate, and the selection efficiency. With the computing power available during the 2012 data taking the maximum reconstruction time at HLT was about 200 ms per event, at the nominal L1T rate of 100 kHz. Tracking algorithms are widely used in the HLT in the object reconstruction through particle-flow techniques as well as in the identification of b-jets and lepton isolation. Reconstructed tracks are also used to distinguish the primary vertex, which identifies the hard interaction process, from the pileup ones. This task is particularly important in the LHC environment given the large number of interactions per bunch crossing: on average 25 in 2012, and expected to be around 40 in Run II with a large contribution from out-of-time particles. In order to cope with tougher conditions the tracking and vertexing techniques used in 2012 have been largely improved in terms of timing and efficiency in order to keep the physics reach at the level of Run I conditions. We will present the performance of these newly developed algorithms, discussing their impact on the b-tagging performances as well as on the jet and missing transverse energy reconstruction.

Sterilization of bacteria suspensions and identification of radicals deposited during plasma treatment

Abstract In this paper we will present results for plasma sterilization of planktonic samples of two reference strains of bacteria, Pseudomonas aeruginosa ATCC 27853 and Enterococcus faecalis ATCC 29212. We have used a plasma needle as a source of non-equilibrium atmospheric plasma in all treatments. This device is already well characterized by OES, derivative probes and mass spectrometry. It was shown that power delivered to the plasma is bellow 2 W and that it produces the main radical oxygen and nitrogen species believed to be responsible for the sterilization process. Here we will only present results obtained by electron paramagnetic resonance which was used to detect the OH, H and NO species. Treatment time and power delivered to the plasma were found to have the strongest influence on sterilization. In all cases we have observed a reduction of several orders of magnitude in the concentration of bacteria and for the longest treatment time complete eradication. A more efficient sterilization was achieved in the case of gram negative bacteria.

AM1 Specific Reaction Parameters for Reactions of Hydroxide Ion with Halomethanes in Complex Environments: Development and Testing.

A specific reaction parameters (SRP) approach is used to reparameterize the semiempirical AM1 method to represent two SN2 reactions of OH(-) with halomethanes (CH3F and CH3Cl). The objective is to develop SRP models that are able to accurately reproduce ab initio potential energy surfaces but with a reduction of several orders of magnitude in computational cost. The developed models, labeled AM1-SRP, have been tested in gas and aqueous phases. The results show that the AM1-SRP models are able to predict a minimum reaction path in close agreement with high level ab initio calculations in gas phase. A QM/MM simulation method with the solutes represented by the AM1-SRP models is used to compute the free energy profiles of the reactions in aqueous phase. The AM1-SRP models predict reaction free energies and free energy barriers in good agreement with experimental values. The current study indicates that semiempirical models may be improved to accurately reproduce an ab initio minimum reaction path at a considerably lower cost. The SRP approach is expected to be useful in long time scale molecular dynamics simulations where it is very expensive to use ab initio methods, specially in condensed phases.

A dynamical model for generating Eurasian lithospheric stress and strain rate fields: Effect of rheology and cratons

For most continents, stress models driven by plate boundary forces have successfully reproduced the main characteristics of the stress field. However, Eurasia has remained a challenge due to its large areas of intraplate deformation. We present a set of three‐dimensional models of the upper mantle lithosphere system for a simplified geometry of the Eurasian plate where we try to match the first‐order characteristics of the stress and strain rate fields simultaneously. For typical elastic, viscous, or plastic rheologies, high stress levels are required in order to produce realistic convergence rates between India and Asia. Our models show robustly that such stresses are transmitted throughout most of the plate, dominating locally generated stresses even in distal regions such as Europe in a manner that is not compatible with observations. Cratons with roots that extend deep into the mantle are unable to provide a significant stress‐shielding effect unless the viscosity contrast between the asthenosphere and the underlying mantle is around 100 or greater. A damage rheology for the lithosphere with history‐dependent behavior and material softening by a viscosity reduction of several orders of magnitude is shown to eliminate this conundrum. Continental convergence at high velocity but low stress is facilitated by the formation of long‐lived shear zones similar to those observed north of the Himalayas. The low stress associated with the collision, together with the decoupling effect of the shear zones, causes the distal stress field in Europe to be controlled by the effects of the neighboring boundaries in agreement with observations.

Electrochemical pH-Stat for Microliter Fluid Specimens

pH-stating is a widely used technique for monitoring various (bio)chemical reactions involving generation or consumption of hydrogen/hydroxide ions. In this work a reagentless pH-stat has been developed where acid or base are generated by water electrolysis. Since the current that generates the reagents can be precisely measured even at low levels, very small sample volumes, in the 10-20 μL range, become accessible for pH-stating: a reduction of several orders of magnitude in specimen size. With proper current control the scheme may be calibration free by virtue of Faraday's law. Simplicity of reagent delivery and miniaturization of the accessible sample size render this approach advantageous with respect to conventional pH-stats, especially in the biomedical area. The electrochemical micro pH-stat is a derivative of the Rotating Sample System that has been conceived for a number of applications in very small samples.

Nanosecond Time‐Resolved Fluorescence Protease Assays

A background reduction of several orders of magnitude in fluorescence protease assays can be achieved by employing time-resolved fluorescence detection with a time gate to eliminate short-lived emission. The use of 2,3-diazabicyclo[2.2.2]oct-2-ene as a hydrophilic and biocompatible fluorophore allows such measurements to be conducted in the nanosecond time domain with a purely organic fluorophore for the first time. The method is applicable to assaying carboxypeptidase A activity on both the laboratory scale and in high-throughput screening.

Comparison of expanded austenite and expanded martensite formed after nitrogen PIII

The formation of expanded austenite is a well-established phenomenon after energetic nitrogen ion implantation into austenitic steel or nickel alloys, correlated with a very high surface hardness and wear resistance. Using high-voltage PIII implantation into martensitic stainless steels, it is also possible to form expanded martensite. In this work, results are compared for martensitic and austenitic stainless steel. Both groups of materials show a lattice expansion of the base material within the nitrogen diffusion zone of several micrometer at 400 °C. Nevertheless, the expanded martensite exhibits a significantly higher hardness, reaching up to 2000 HV, compared to 1000–1200 HV for expanded austenite. In both groups, a wear reduction of several orders of magnitude is observed. An attempt is made to explain the results within the framework of ion-induced effects.

Comparison of expanded austenite and expanded martensite formed after nitrogen PIII

The formation of expanded austenite is a well-established phenomenon after energetic nitrogen ion implantation into austenitic steel or nickel alloys , correlated with a very high surface hardness and wear resistance. Using high-voltage PIII implantation into martensitic stainless steels , it is also possible to form expanded martensite. In this work, results are compared for martensitic and austenitic stainless steel. Both groups of materials show a lattice expansion of the base material within the nitrogen diffusion zone of several micrometer at 400 °C. Nevertheless, the expanded martensite exhibits a significantly higher hardness, reaching up to 2000 HV, compared to 1000–1200 HV for expanded austenite. In both groups, a wear reduction of several orders of magnitude is observed. An attempt is made to explain the results within the framework of ion-induced effects.