Local Affine Research Articles

Convex formulation of confidence level optimization of DG affine reactive power controllers in distribution grids

Volatile productions and consumptions generate a stochastic behavior of distribution grids and make its supervision difficult to achieve. Usually, the Distributed Generators reactive powers are adjusted to perform decentralized voltage control. Industrial controllers are generally equipped with a local affine feedback law, which settings are tuned at early stage using local data. A centralized and more efficient tuning method should aim to maximize the probability that all the node voltages of distribution grids remain within prescribed bounds. When the characteristics of the stochastic power forecasts are known, the centralized algorithm allows to update the settings on a regular time basis. However, the method requires to solve stochastic optimization problem. Assuming that stochastic variables have Gaussian distributions, a procedure is given which guarantees the convergence of the stochastic optimization. Convex problems drastically reduce the difficulty and the computational time required to reach the global minimum, compared to nonconvex optimal power flow problems. The linear controllers with optimized parameters are compared to traditional control laws using simulations of a real distribution grid model. The results show that the algorithm is reliable and moreover fast enough. Hence, the proposed method can be used to update periodically the control parameters.

Enhanced Catalysis from Multienzyme Cascades Assembled on a DNA Origami Triangle.

Developing reliable methods of constructing cell-free multienzyme biocatalytic systems is a milestone goal of synthetic biology. It would enable overcoming the limitations of current cell-based systems, which suffer from the presence of competing pathways, toxicity, and inefficient access to extracellular reactants and removal of products. DNA nanostructures have been suggested as ideal scaffolds for assembling sequential enzymatic cascades in close enough proximity to potentially allow for exploiting of channeling effects; however, initial demonstrations have provided somewhat contradictory results toward confirming this phenomenon. In this work, a three-enzyme sequential cascade was realized by site-specifically immobilizing DNA-conjugated amylase, maltase, and glucokinase on a self-assembled DNA origami triangle. The kinetics of seven different enzyme configurations were evaluated experimentally and compared to simulations of optimized activity. A 30-fold increase in the pathway's kinetic activity was observed for enzymes assembled to the DNA. Detailed kinetic analysis suggests that this catalytic enhancement originated from increased enzyme stability and a localized DNA surface affinity or hydration layer effect and not from a directed enzyme-to-enzyme channeling mechanism. Nevertheless, the approach used to construct this pathway still shows promise toward improving other more elaborate multienzymatic cascades and could potentially allow for the custom synthesis of complex (bio)molecules that cannot be realized with conventional organic chemistry approaches.

Developing a coaching identity in interdisciplinary STEM partnerships

ABSTRACTWe presented one new coach’s development during a K-12 STEM integration initiative in order to explore how coaches develop professional identities for coaching. Madison, a former mathematics educator, partnered with science teachers for 1 year in order to co-develop and implement STEM curriculum. Transcripts of coaching conversations, written reflection data, assignments from a graduate-level coaching course, and interviews were analyzed using three-phased qualitative methods. Findings suggest that a new coach approximated practice similar to new teachers; relied on existing professional identities when constructing new coaching identities; and the local, global, and institutional affinity groups for STEM coaching both informed and disrupted coaching identity development. The process by which new coaches develop contextually dependent and discipline-specific identities for coaching has implications for the success of STEM coaching efforts.

Local affine selections of convex set-valued functions

Abstract It is well known that not every convex set-valued function admits an affine selection. One could ask whether there exists at least one local affine selection. The answer is positive in the finite-dimensional case. The main part of the note comprises two examples of the non-existence of local affine selections of convex set-valued functions defined on certain infinite-dimensional Banach spaces.

A method for local approximation of a planar deformation field

Abstract We present a method of approximation of a deformation field based on the local affine transformations constructed based on n nearest neighbors with respect to points of adopted grid. The local affine transformations are weighted by means of inverse distance squared between each grid point and observed points (nearest neighbors). This work uses a deformation gradient, although it is possible to use a displacement gradient instead – the two approaches are equivalent. To decompose the deformation gradient into components related to rigid motions (rotations, translations are excluded from the deformation gradient through differentiation process) and deformations, we used a polar decomposition and decomposition into a sum of symmetric and an anti-symmetric matrices (tensors). We discuss the results from both decompositions. Calibration of a local affine transformations model (i.e., number of nearest neighbors) is performed on observed points and is carried out in a cross-validation procedure. Verification of the method was conducted on simulated data-grids subjected to known (functionally generated) deformations, hence, known in every point of a study area.

Node-Sensitive Graph Fusion via Topo-Correlation for Image Retrieval

Various kinds of features prove to be effective for content-based image retrieval. However, due to the diversity of image contents, a descriptor may achieve impressive performance on specific images while becoming invalid on others. Although some efforts have been made to combine features as complementary counterparts, proper weighting scheme is still a challenge for fast and accurate retrieval. In this paper, we propose an effective fusion method, termed as Topo-correlation (Topo), where the importance of each feature is measured by cross-view correlations on local affinity graphs. Specifically, the weights of similarities are node-sensitive as well as modality-sensitive, thus boosting the results of good cues while depressing adverse factors for individual images. By estimating the consensus of similarity scores with regard to a query-driven criterion, the weighted graphs are generated efficiently with low computational complexity. Extensive experimental results on four benchmarks demonstrate the superiority of the proposed approach over the state-of-the-art methods.

Rare Human Missense Variants can affect the Function of Disease-Relevant Proteins by Loss and Gain of Peroxisomal Targeting Motifs.

Single nucleotide variants (SNVs) resulting in amino acid substitutions (i.e., missense variants) can affect protein localization by changing or creating new targeting signals. Here, we studied the potential of naturally occurring SNVs from the Genome Aggregation Database (gnomAD) to result in the loss of an existing peroxisomal targeting signal 1 (PTS1) or gain of a novel PTS1 leading to mistargeting of cytosolic proteins to peroxisomes. Filtering down from 32,985 SNVs resulting in missense mutations within the C-terminal tripeptide of 23,064 human proteins, based on gene annotation data and computational prediction, we selected six SNVs for experimental testing of loss of function (LoF) of the PTS1 motif and five SNVs in cytosolic proteins for gain in PTS1-mediated peroxisome import (GoF). Experimental verification by immunofluorescence microscopy for subcellular localization and FRET affinity measurements for interaction with the receptor PEX5 demonstrated that five of the six predicted LoF SNVs resulted in loss of the PTS1 motif while three of five predicted GoF SNVs resulted in de novo PTS1 generation. Overall, we showed that a complementary approach incorporating bioinformatics methods and experimental testing was successful in identifying SNVs capable of altering peroxisome protein import, which may have implications in human disease.

Abstract 1137: Unraveling spatially-dependent interactions of tumor-associated macrophage in the tumor microenvironment

Abstract Tumor-infiltrating lymphocytes (TIL) are a strong prognostic factor in cancer patient outcomes. This includes immune checkpoint blockade (ICB) of the PD-1/PD-L1 axis, where patients with a higher abundance of pre-treatment TILs respond more frequently. However, tumors often elaborate mechanisms to exclude TILs and/or suppress their function. Identifying negative regulators of TILs is thus of great importance to improve patient outcomes and increase the patient population that may benefit from ICB. Tumor-associated macrophages (TAM) are a known modulator of TIL activity but the interaction between the two cell types is typically characterized with in vitro experiments and spatially-agnostic sequencing, ignoring location-specific factors contributing to the interaction’s net effects. Interrogating the TAM-TIL interactions in an intact tumor microenvironment (TME) will uncover novel mechanisms responsible for TAM function and may be key to reversing TIL immunosuppression. Here we leverage bulk RNA-sequencing with histocytometry, a multiplex quantitative tissue imaging method, to spatially-resolve the TAM-TIL interactions in intact human metastatic melanoma microenvironments. Bulk sequencing distinguishes tumor samples by high and low enrichment of lymphocyte gene sets, with additional sub-stratification by macrophage enrichment. Modular repertoire analysis implicates an interferon response in the differential lymphocyte enrichment, but does not address macrophage-dependent effects. Drawing from physical chemistry concepts, our spatial analyses of cell-cell interactions from histocytometry reveals distinct TAM populations potentially regulating TIL activity. Phagocytic TAMs in the tumor and non-phagocytic TAMs in the stroma physically contact T-cells, with the former interaction strongly correlated with lymphocyte enrichment and T-cell infiltration. Local and regional TAM phagocytosis affinities estimated by the Langmuir adsorption model further detail the nature of this interaction within the tumor, as well as the role of macrophage phagocytosis in regulating TILs. We demonstrate linear models integrating bulk sequencing and spatial metrics to quantify the relationships between genetics, cell-cell interactions, and T-cell invasiveness. Our work unravels important TAM interactions that shape the TME which have not been previously appreciated, providing novel insight into the forces driving T-cell exclusion and revealing new TAM biology to explore further. Citation Format: Victor Wang, Jan Martinek, Hannah M. Brookes, Kyung In Kim, Karolina Palucka, Jeff Chuang. Unraveling spatially-dependent interactions of tumor-associated macrophage in the tumor microenvironment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 1137.

LAM: Locality affine-invariant feature matching

False match removal is a crucial and fundamental task in photogrammetry and computer vision. This paper proposes a robust and efficient mismatch-removal algorithm based on the concepts of local barycentric coordinate (LBC) and matching coordinate matrices (MCMs), called locality affine-invariant matching (LAM). LAM is suitable for both rigid and nonrigid image matching problems. We define a novel LBC system based on area ratios, which is invariant to local affine transformations. We also present the MCMs based on the coordinates of matches, whose degeneracy is able to indicate the correctness of correspondences. Our LAM method first builds a mathematical model based on the LBCs to extract good matches that preserve local neighborhood structures. Then, LAM constructs local MCMs using the extracted reliable correspondences and identifies the correctness for the remaining matches via minimizing the rank of the MCMs. LAM has linear space and linearithmic time complexities. Extensive experiments on both rigid and nonrigid real datasets demonstrate the power of the proposed method; i.e., LAM is more robust to complex transformations compared to other methods and is two orders of magnitude faster than RANSAC under low inlier rates. The source code of the proposed LAM method will be publicly available in http://www.escience.cn/people/lijiayuan/index.html.

The transformation of the affine velocityand its application to a rotating disc

The aim of the article is to find a transformation that links the local affine velocity of a non-rigid body in the laboratory inertial reference frame $ S $ with the centro-affine velocity of motion of this body in the accompanying accelerated frame $ k $. This paper is based on the kinematics of a continuous medium and the generalized Lorentz transformation. In this paper we show the 3D transformation of velocity linking the reference system $ S $ and the reference system $ k $, which moves without rotation. Wherein the motion of various points of the rigid system $ k $ is inhomogeneous. Using these formulas, we obtain the desired direct and inverse transformation of the local affine velocity. Important special cases of this transformation are considered. They are the motion of particles in a uniform force field and the precession of Thomas. As an example of using the transformation of affine velocity in $ S $, accelerated rotation of the disk was considered and the local angular velocity and the magnitude of the deformation of its points were calculated. Wherein, the calculated stretching coefficient is consistent with the known one, and the formula found for the angular velocity is more general than the earlier result obtained for uniform rotation of the disk.

Impact study of data locality on task-based applications through the Heteroprio scheduler.

The task-based approach has emerged as a viable way to effectively use modern heterogeneous computing nodes. It allows the development of parallel applications with an abstraction of the hardware by delegating task distribution and load balancing to a dynamic scheduler. In this organization, the scheduler is the most critical component that solves the DAG scheduling problem in order to select the right processing unit for the computation of each task. In this work, we extend our Heteroprio scheduler that was originally created to execute the fast multipole method on multi-GPUs nodes. We improve Heteroprio by taking into account data locality during task distribution. The main principle is to use different task-lists for the different memory nodes and to investigate how locality affinity between the tasks and the different memory nodes can be evaluated without looking at the tasks’ dependencies. We evaluate the benefit of our method on two linear algebra applications and a stencil code. We show that simple heuristics can provide significant performance improvement and cut by more than half the total memory transfer of an execution.

Structure regularized sparse coding for data representation

Sparse coding (SC) exhibits impressive performance in many practical applications. However, in the unsupervised scenario, most of the conventional SC methods fail to fully take advantage of the structure of the data. Actually, the structure of the data, especially the global structure that is an implicit prior knowledge, is vital for data analysis. In this paper, we propose a novel method called structure regularized sparse coding (SRSC) for the sparse representation of the data in the unsupervised scenario. In contrast with the other SC methods, a distinct feature of SRSC is that it takes into consideration both the local and global structure of the data and fully exploits the latent category information in the data. By using the local affinity matrix that captures the local structure, we first build a global affinity matrix to encode the global structure of the data. The global affinity matrix fully carries the latent category information that is beneficial to obtain the discriminating representation of the data, Then, we define the optimization model of SRSC and develop a two-step iterative optimization algorithm based on the alternating direction method of multipliers (ADMM) to solve it. The experimental results validate that the proposed method is effective and can achieve better performance over its counterparts.

BALG: An alternative for fast and robust feature matching

Robust feature matcher is usually inapplicable to real time computer vision applications as its high computational complexity, while fast feature matcher is short of robustness to geometric transformations. By studying the sampling pattern of binary descriptor and local geometric statistics of features, this paper proposes a fast and robust feature matching method with binary affine invariant descriptor and local geometric consistency check (BALG). The sampling pattern is adaptively adjusted according local affine transformation that can be effectively estimated from the local intensity moments. The affine sampling pattern improves the affine invariance of binary descriptor while enable fast processing. Furthermore, candidate matches are sorted into local groups along different orientations with sequence order constraint, and then followed by local geometric consistency check. False matches are efficiently filtered out with high recall. Extensive experiments on four publicly benchmark datasets prove the proposed method to be an alternative for time critical feature matcher.

Comparison between Echinococcus granulosus sensu stricto (G1) and E. canadensis (G6) mitochondrial genes (cox1 and nad1) and their related protein models using experimental and bioinformatics analysis

BackgroundCystic echinococcosis (CE) as a zoonotic parasitic disease, remains a health challenge in many parts of the world. There are different species of Echinococcus granulosus sensu lato with different pathogenicity and host preferences.Different procedures have been applied for characterization of Echinococcus taxa in which two mitochondrial genes, cox1 and nad1 have been used more common. They have been able to differentiate E. granulosus sensu stricto and E. canadensis species in different hosts. The affinity of E. granulosus sensu stricto and E. canadensis species for localizing different organs seems to be different. To what such affinity and related pathogenicity could be related, is not known, so far. Bioinformatics analysis may be helpful to interpret such difference by investigating the genes and their related protein models between different species infecting human and animals. The current work was designed to study the differences between E. granulosus s.s. and E. canadensis species mitochondrial genes (cox1 and nad1) and related protein models of CE cysts by experimental and bioinformatics analysis. Materials and methodsDifferent human and animal CE cysts were collected and their DNA was extracted and sequenced based on their cox1 and nad1 genes. In order to determine the E. granulosus s.s. and E. canadensis species of the samples, BLAST analysis was performed on sequenced genes. Three sequences were selected for analysis and were deposited in GenBank. Moreover, the sequence number of KT988116.1 which belonged to E. canadensis from our already deposited in GenBank was also selected. Alignment and phylogenetic analysis were performed on the sequences using BioEdit and MEGA7 software. The raw sequences of translated proteins belonged to the mentioned genes were obtained from Protein database in NCBI. The secondary structure was determined by PSIPRED Protein Sequence Analysis Workbench. The tertiary models of COX1 and NAD1 proteins in both genotypes were constructed using Modeler 9.12 software and their physicochemical features were computed using ProtParam tool in ExPASY server. ResultsBLAST analysis on sequenced genes showed that the samples belonged to E. granulosus s.s. and E. canadensis species. These sequences were deposited in GenBank with accession numbers: JN579173.1, KF437811.1, and KY924632.1.The results showed that proteins of COX1 of E. granulosus s.s., COX1of E. canadensis, NAD1of E. granulosus s.s. and NAD1of E. canadensis species, consisted of 135, 122, 120 and 124 amino acids, respectively. The aligned sequences of translated proteins belonged to COX1 and NAD1 enzymes in E. granulosus s.s. and E. canadensis species were different; such that alignment COX1 sequence between E. granulosus s.s. and E. canadensis species showed that amino acids were different in 6 positions. This difference for NAD1 sequences were different in 19 positions. The secondary structure determined by PSIPRED showed differences in coil, strand and helix chains in COX1 and NAD1 proteins in E. granulosus s.s. and E. canadensis species. Comparison between three-dimensional structures (3D) of COX1 protein model in E. granulosus s.s. and E. canadensis species demonstrated an additional helix with two conserved iron binding sites in the COX1 protein of E. granulosus s.s. species. ConclusionE. granulosus s.s. and E. canadensis species differences are reflected in two important proteins: COX1 and NAD1. These differences are demonstrable in the 3D structure of proteins of both strains. So, the present study is adding to our understanding of the difference in molecular sequences between the E. granulosus s.s. (G1) and E. canadensis (G6) which may be used for interpreting the difference between the pathogenicity and localization affinity in these two important helminthic zoonosis.

Shape based local affine invariant texture characteristics for fabric image retrieval

The rapid growth of fabric images needs fast retrieval for related applications, such as fashion design. The goal of fabric image retrieval is to retrieve and rank relevant fabrics from a large scale fabric database to visually assist users’ online shopping process in e-commerce. Most of existing solutions to this issue are not invariant with respect to 2D similarity or affine transformations, much less to 3D transformations of textured surface. In this paper, we propose a new search method with a shape based local affine invariant texture characteristics. By employing topographic map to represent fabric images, which is a complete, multi-scale and contrast invariant representation, the proposed method first obtains a tree of shapes from the topographic map. Then, a group of statistics is applied on these shapes to acquire a set of features that are invariant to 3D transformations. We finally represent these features combing relations between shapes, and based on the representation the similarity of pairs of fabric images can be estimated. To evaluate the performance of our algorithm, we conducted a series of experiments on a real-world fabric image dataset, and compared the proposed method with other previous ones. Experimental results demonstrate that the time of the proposed method spending in searching is less than 1 second, and meanwhile a higher PR score than others is obtained.

Blended Emotion in-the-Wild: Multi-label Facial Expression Recognition Using Crowdsourced Annotations and Deep Locality Feature Learning

Comprehending different categories of facial expressions plays a great role in the design of computational model analyzing human perceived and affective state. Authoritative studies have revealed that facial expressions in human daily life are in multiple or co-occurring mental states. However, due to the lack of valid datasets, most previous studies are still restricted to basic emotions with single label. In this paper, we present a novel multi-label facial expression database, RAF-ML, along with a new deep learning algorithm, to address this problem. Specifically, a crowdsourcing annotation of 1.2 million labels from 315 participants was implemented to identify the multi-label expressions collected from social network, then EM algorithm was designed to filter out unreliable labels. For all we know, RAF-ML is the first database in the wild that provides with crowdsourced cognition for multi-label expressions. Focusing on the ambiguity and continuity of blended expressions, we propose a new deep manifold learning network, called Deep Bi-Manifold CNN, to learn the discriminative feature for multi-label expressions by jointly preserving the local affinity of deep features and the manifold structures of emotion labels. Furthermore, a deep domain adaption method is leveraged to extend the deep manifold features learned from RAF-ML to other expression databases under various imaging conditions and cultures. Extensive experiments on the RAF-ML and other diverse databases (JAFFE, CK$$+$$+, SFEW and MMI) show that the deep manifold feature is not only superior in multi-label expression recognition in the wild, but also captures the elemental and generic components that are effective for a wide range of expression recognition tasks.

Plastic deformation of a permanently bonded network: Stress relaxation by pleats.

We show that a flat two dimensional network of connected vertices, when stretched, may deform plastically by producing "pleats", system spanning linear structures with width comparable to the lattice spacing, where the network overlaps on itself. To understand the pleating process, we introduce an external field that couples to local non-affine displacements, i.e., those displacements of neighbouring vertices that cannot be represented as a local affine strain. We obtain both zero and finite temperature phase diagrams in the strain-field plane. Pleats occur here as a result of an equilibrium first-order transition from the homogeneous network to a heterogeneous phase where stress is localised within pleats and eliminated elsewhere. We show that in the thermodynamic limit, the un-pleated state is always metastable at vanishing field for infinitesimal strain. Plastic deformation of the initially homogeneous network is akin to the decay of a metastable phase via a dynamical transition. We make predictions concerning local stress distributions and thermal effects associated with pleats which may be observable in suitable experimental systems.

3D Human Action Recognition Using a Single Depth Feature and Locality-Constrained Affine Subspace Coding

This paper addresses the problem of recognizing human actions from depth videos. We propose a depth-based local descriptor and affine subspace coding representation with locality-constrained affine subspace coding (LASC) for 3D action recognition. First, each depth video sequence is divided into a set of subsequences (i.e., multi-scale sub-actions) based on the normalized motion energy vector. Next, depth motion map-based gradient local auto-correlation features are employed to capture the shape information and motion cues of each sub-action. In order to obtain discriminative and compact representation, we extract the local high-order information of the depth video using LASC. Through experiments, we show that the use of LASC exhibits better performance compared with existing methods such as locality-constrained linear coding. We compared LASC with the state-of-the-art methods based on similar principle, using features extracted from a single modality, on four datasets, and with those using multiple features or nonlinear recognition machines. The results on four datasets clearly show the effectiveness of the proposed method.

Non-rigid point cloud registration based lung motion estimation using tangent-plane distance.

Accurate estimation of motion field in respiration-correlated 4DCT images, is a precondition for the analysis of patient-specific breathing dynamics and subsequent image-supported treatment planning. However, the lung motion estimation often suffers from the sliding motion. In this paper, a novel lung motion method based on the non-rigid registration of point clouds is proposed, and the tangent-plane distance is used to represent the distance term, which describes the difference between two point clouds. Local affine transformation model is used to express the non-rigid deformation of the lung motion. The final objective function is expressed in the Frobenius norm formation, and matrix optimization scheme is carried out to find out the optimal transformation parameters that minimize the objective function. A key advantage of our proposed method is that it alleviates the requirement that the source point cloud and the reference point cloud should be in one-to-one corresponding relationship, and the requirement is difficult to be satisfied in practical application. Furthermore, the proposed method takes the sliding motion of the lung into consideration and improves the registration accuracy by reducing the constraint of the motion along the tangent direction. Non-rigid registration experiments are carried out to validate the performance of the proposed method using popi-model data. The results demonstrate that the proposed method outperforms the traditional method with about 20% accuracy increase.

Monitoring ecological change inUKwoodlands and rivers: An exploration of the relational geographies of citizen science

The adoption of citizen science methodologies by environmental organisations and ecologists entrusts the task of collecting ecological data to non‐experts operating at a local scale. This presents individuals and communities with opportunities to monitor ecological change and contribute to local environmental management. Little is known about why volunteers choose to participate in burgeoning contemporary citizen science research initiatives. The aim of this paper is thus to explore volunteer motivation for involvement in two environmental citizen science initiatives, based in the United Kingdom. It contributes to understandings of the socio‐geographical influences that act on participation in environmental citizen science. It is proposed within this paper that affective connections with local geographies provide a conceptual framework for understanding citizen science motivations. The paper discusses the main themes emerging from site‐based, in‐depth interviews with 22 citizen science participants in variousUKlocations. The study revealed that early affective bonds formed with ecological spaces endured throughout life courses, while citizen science participation offered a way of remaining connected to local environments. The paper reflects on the endurance of affective environmental bonds and their manifestation within the expressed motivations for citizen science participation, which emerged as fulfilling a compulsion to observe ecological surroundings, a desire to participate in environmental research and a commitment to protecting local environments. The paper proposes that citizen science participation offers a framework to connect to and protect local and global affinity spaces, while assisting in monitoring global environmental change.