Co-location Patterns Research Articles

CONNEKT: Co-Located Nearest Neighbor Search using KNN Querying with K-D Tree

Data about entities or objects associated with geographical or location information could be called as spatial data. Spatial data helps in identifying and positioning anyone or anything globally anywhere across the world. Instances of various spatial features that are closely found together are called as spatial co-located patterns. So far, the spatial co-located patterns have been used only for knowledge discovery process but it would serve a wide variety of applications if analyzed intensively. One such application is to use co-location pattern mining for a context aware based search. Hence the main aim of this work is to extend the K-Nearest Neighbor (KNN) querying to co-located instances for context aware based querying or location-based services (LBS). For the above-said purpose, co-located nearest neighbor search algorithm namely “CONNEKT” is proposed. The co-located instances are mapped onto a K-dimensional tree (K-d tree) inorder to make the querying process efficient. The algorithm is analyzed using a hypothetical data set generated through QGIS.

Interdisciplinary-driven hypotheses on spatial associations of mixtures of industrial air pollutants with adverse birth outcomes

BackgroundAdverse birth outcomes (ABO) such as prematurity and small for gestational age confer a high risk of mortality and morbidity. ABO have been linked to air pollution; however, relationships with mixtures of industrial emissions are poorly understood. The exploration of relationships between ABO and mixtures is complex when hundreds of chemicals are analyzed simultaneously, requiring the use of novel approaches. ObjectiveWe aimed to generate robust hypotheses spatially linking mixtures and the occurrence of ABO using a spatial data mining algorithm and subsequent geographical and statistical analysis. The spatial data mining approach aimed to reduce data dimensionality and efficiently identify spatial associations between multiple chemicals and ABO. MethodsWe discovered co-location patterns of mixtures and ABO in Alberta, Canada (2006–2012). An ad-hoc spatial data mining algorithm allowed the extraction of primary co-location patterns of 136 chemicals released into the air by 6279 industrial facilities (National Pollutant Release Inventory), wind-patterns from 182 stations, and 333,247 singleton live births at the maternal postal code at delivery (Alberta Perinatal Health Program), from which we identified cases of preterm birth, small for gestational age, and low birth weight at term. We selected secondary patterns using a lift ratio metric from ABO and non-ABO impacted by the same mixture. The relevance of the secondary patterns was estimated using logistic models (adjusted by socioeconomic status and ABO-related maternal factors) and a geographic-based assignment of maternal exposure to the mixtures as calculated by kernel density. ResultsFrom 136 chemicals and three ABO, spatial data mining identified 1700 primary patterns from which five secondary patterns of three-chemical mixtures, including particulate matter, methyl-ethyl-ketone, xylene, carbon monoxide, 2-butoxyethanol, and n-butyl alcohol, were subsequently analyzed. The significance of the associations (odds ratio > 1) between the five mixtures and ABO provided statistical support for a new set of hypotheses. ConclusionThis study demonstrated that, in complex research settings, spatial data mining followed by pattern selection and geographic and statistical analyses can catalyze future research on associations between air pollutant mixtures and adverse birth outcomes.

Dynamic compartmentalization of purine nucleotide metabolic enzymes at leading edge in highly motile renal cell carcinoma

Compartmentalization is vital for biological systems at multiple levels, including biochemical reactions in metabolism. Organelle-based compartments such as mitochondria and peroxisomes sequester the responsible enzymes and increase the efficiency of metabolism while simultaneously protecting the cell from dangerous intermediates, such as radical oxygen species. Recent studies show intracellular nucleotides, such as ATP and GTP, are heterogeneously distributed in cells with high concentrations at the lamellipodial and filopodial projections, or leading edge. However, the intracellular distribution of purine nucleotide enzymes remains unclear. Here, we report the enhanced localization of GTP-biosynthetic enzymes, including inosine monophosphate dehydrogenase (IMPDH isotype 1 and 2), GMP synthase (GMPS), guanylate kinase (GUK1) and nucleoside diphosphate kinase-A (NDPK-A) at the leading edge in renal cell carcinoma cells. They show significant co-localization at the membrane subdomain, and their co-localization pattern at the membrane is distinct from that of the cell body. While other purine nucleotide biosynthetic enzymes also show significant localization at the leading edge, their co-localization pattern with IMPDH is divergent. In contrast, a key glycolytic enzyme, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), predominantly localized in the cytoplasm. Mechanistically, we found that plasma membrane localization of IMPDH isozymes requires active actin polymerization. Our results demonstrate the formation of a discrete metabolic compartment for localized purine biosynthesis at the leading edge, which may promote localized nucleotide metabolism for cell migration and metastasis in cancers.

Detecting urban fire high-risk regions using colocation pattern measures

Understanding the spatial patterns of fire incidents and identifying high-risk locations are important for urban fire safety management. Colocation pattern analysis involving geographic information system (GIS) intends to examine the spatial proximity of distinct categories of objects. Most existing studies are conducted from a global perspective and assumes that urban events taking place occur in an isotropic Euclidean space. However, the physical environment embedded in an urban space is virtually characterized by the layout of road network, and many valid scopes of patterns are restricted to local regions. In response to the above-mentioned limitations, this study investigates the spatial correlation patterns between types of fires and their neighbouring land-use facility types based on network distance using two colocation measures: the global colocation quotient (GCLQ) and the local colocation quotient (LCLQ). The global measure helps detect the overall colocation patterns across the entire study area, whereas a LCLQ can provide more detailed site-specific patterns and can capture the variability of colocation patterns across places. For illustration purposes, we use the two methods to analyse the spatial associations between 6 types of fires and 12 types of land-use facilities in the city of Nanjing, China. Experimental results show that our analysis offers valuable insights into fire risk estimation and fire service management.

Distributions and relationships of chemically defined enteroendocrine cells in the rat gastric mucosa.

This paper provides quantitative data on the distributions of enteroendocrine cells (EEC), defined by the hormones they contain, patterns of colocalisation between hormones and EEC relations to nerve fibres in the rat gastric mucosa. The rat stomach has three mucosal types: non-glandular stratified squamous epithelium of the fundus and esophageal groove, a region of oxyntic glands in the corpus, and pyloric glands of the antrum and pylorus. Ghrelin and histamine were both contained in closed cells, not contacting the lumen, and were most numerous in the corpus. Gastrin cells were confined to the antrum, and 5-hydroxytryptamine (5-HT) and somatostatin cells were more frequent in the antrum than the corpus. Most somatostatin cells had basal processes that in the antrum commonly contacted gastrin cells. Peptide YY (PYY) cells were rare and mainly in the antrum. The only numerous colocalisations were 5-HT and histamine, PYY and gastrin and gastrin and histamine in the antrum, but each of these populations was small. Peptide-containing nerve fibres were found in the mucosa. One of the most common types was vasoactive intestinal peptide (VIP) fibres. High-resolution analysis showed that ghrelin cells were closely and selectively approached by VIP fibres. In contrast, gastrin cells were not selectively innervated by VIP or CGRP fibres. The study indicates that there are distinct populations of gastric EEC and selective innervation of ghrelin cells. It also shows that, in contrast to EEC of the small intestine, the majority of EEC within the stomach contained only a single hormone.

Sa1530 – Celiac Disease Associated Liver Dysfunction: Study of the Histological Features and Co-Localisation Patterns of Iga and Igg Anti-Tissue Transglutaminase Antibody in Liver Biopsies

Sa1530 – Celiac Disease Associated Liver Dysfunction: Study of the Histological Features and Co-Localisation Patterns of Iga and Igg Anti-Tissue Transglutaminase Antibody in Liver Biopsies

A clique-based approach for co-location pattern mining

Co-location pattern mining refers to the task of discovering the group of features (geographic object types) whose instances (geographic objects) are frequently located close together in a geometric space. Current approaches on this topic adopt a prevalence threshold (a measure of a user's interest in a pattern) to generate prevalent co-location patterns. However, in practice, it is not easy to specify a suitable prevalence threshold. Thus, users have to repeatedly execute the program to find a suitable prevalence threshold. Besides, the efficiency of these approaches is limited because of the expensive cost of identifying row-instances of co-location patterns. In this paper, we propose a novel clique-based approach for discovering complete and correct prevalent co-location patterns. The proposed approach avoids identifying row-instances of co-location patterns thus making it much easier to find a proper prevalence threshold. First, two efficient schemas are designed to generate complete and correct cliques. Next, these cliques are transformed into a hash structure which is independent of the prevalence threshold. Finally, the prevalence of each co-location pattern is efficiently calculated using the hash structure. The experiments on both real and synthetic datasets show the efficiency and effectiveness of our proposed approaches.

HR23B pathology preferentially co-localizes with p62, pTDP-43 and poly-GA in C9ORF72-linked frontotemporal dementia and amyotrophic lateral sclerosis

Human homologue of yeast UV excision repair protein Rad23b (HR23B) inclusions are found in a number of neurodegenerative diseases, including frontotemporal dementia (FTD), Huntington’s disease (HD), spinocerebellar ataxia type 3 and 7 (SCA3/7), fragile X associated tremor/ataxia syndrome (FXTAS) and Parkinson’s disease (PD). Here, we describe HR23B pathology in C9ORF72 linked FTD and amyotrophic lateral sclerosis (ALS) cases. HR23B presented in neuropils, intranuclear inclusions and cytoplasmic and perinuclear inclusions and was predominantly found in cortices (frontal, temporal and motor), spinal cord and hippocampal dentate gyrus. HR23B co-localized with poly-GA-, pTDP-43- and p62-positive inclusions in frontal cortex and in hippocampal dentate gyrus, the latter showing higher co-localization percentages. HR23B binding partners XPC, 20S and ataxin-3, which are involved in nucleotide excision repair (NER) and the ubiquitin-proteasome system (UPS), did not show an aberrant distribution. However, C9ORF72 fibroblasts were more sensitive for UV-C damage than healthy control fibroblasts, even though all factors involved in NER localized normally to DNA damage and the efficiency of DNA repair was not reduced. HR23Bs other binding partner NGly1/PNGase, involved in ER-associated degradation (ERAD) of misfolded proteins, was not expressed in the majority of neurons in C9FTD/ALS brain sections compared to non-demented controls. Our results suggest a difference in HR23B aggregation and co-localization pattern with DPRs, pTDP-43 and p62 between different brain areas from C9FTD/ALS cases. We hypothesize that HR23B may play a role in C9ORF72 pathogenesis, possibly by aberrant ERAD functioning.

Place niche and its regional variability: Measuring spatial context patterns for points of interest with representation learning

In the built environment, places such as retail outlets and public sites are embedded in the spatial context formed by neighboring places. We define the sets of these symbiotic places in the proximity of a focal place as the place's “place niche”, which conceptually represents the features of the local environment. While current literature has focused on pairwise spatial colocation patterns, we represent the niche as an integrated feature for each type of place, and quantify the niches' variation across cities. Here, with point of interest (POI) data as an approximation of places in cities, we propose representation learning models to explore place niche patterns. The models generate two main outputs: first, distributed representations for place niche by POI category (e.g. Restaurant, Museum, Park) in a latent vector space, where close vectors represent similar niches; and second, conditional probabilities of POI appearance of each place type in the proximity of a focal POI. With a case study using Yelp data in four U.S. cities, we reveal spatial context patterns and find that some POI categories have more unique surroundings than others. We also demonstrate that niche patterns are strong indicators of the function of POI categories in Phoenix and Las Vegas, but not in Pittsburgh and Cleveland. Moreover, we find that niche patterns of more commercialized categories tend to have less regional variation than others, and the city-level niche-pattern changes for POI categories are generally similar only between certain city pairs. By exploring patterns for place niche, we not only produce geographical knowledge for business location choice and urban policymaking, but also demonstrate the potential and limitations of using spatial context patterns for GIScience tasks such as information retrieval and place recommendation.

Mining maximal sub-prevalent co-location patterns

Spatial prevalent co-location pattern mining is to discover interesting and potentially useful patterns from spatial data, and it plays an important role in identifying spatially correlated features in many domains, such as Earth science and Public transportation. Existing approaches in this field only take into account the clique instances where feature instances form a clique. However, they may neglect some important spatial correlations among features in practice. In this paper, we introduce star participation instances to measure the prevalence of co-location patterns such that spatially correlated instances which cannot form cliques will also be properly considered. Then we propose a new concept called sub-prevalent co-location patterns (SPCP) based on the star participation instances. Furthermore, two efficient algorithms -- the prefix-tree-based algorithm (PTBA) and the partition-based algorithm (PBA) -- are proposed to mine all the maximal sub-prevalent co-location patterns (MSPCP) in a spatial data set. PTBA uses a typical candidate generate-and-test way starting from candidates with the longest pattern-size, while PBA adopts a step-by-step manner starting from 3-size core patterns. We demonstrate the significance of our proposed new concepts as well as the efficiency of our algorithms through extensive experiments.

A Method of Mining Spatial High Utility Co-location Patterns Based on Feature Actual Participation Weight

Spatial high utility co-location patterns mining takes pattern utility participation index as interest measure, but it still lacks standard method for calculating pattern utility participation index. This paper presents some reasonable definitions of spatial high utility colocation patterns based on the datasets of instance with utility value, such as Feature Actual Participation Weight, Feature Utility Participation Index and Pattern Utility Participation Index and many more. A basic algorithm BWHMA(high utility co-location patterns mining based on feature actual participation weight) is proposed. For improving the efficiency of algorithm BWHMA, one pruning algorithm FUPRA(based on feature utility participation rate) and one optimization algorithm FRPWA(based on feature actual participation weight) are put forward. Finally, a large number of comparative experiments have been done on both synthetic and real datasets. The experiments show that the proposed algorithms are feasible, correct and valid.

Localized incorporation of outer membrane components in the pathogen Brucella abortus.

The zoonotic pathogen Brucella abortus is part of the Rhizobiales, which are alpha-proteobacteria displaying unipolar growth. Here, we show that this bacterium exhibits heterogeneity in its outer membrane composition, with clusters of rough lipopolysaccharide co-localizing with the essential outer membrane porin Omp2b, which is proposed to allow facilitated diffusion of solutes through the porin. We also show that the major outer membrane protein Omp25 and peptidoglycan are incorporated at the new pole and the division site, the expected growth sites. Interestingly, lipopolysaccharide is also inserted at the same growth sites. The absence of long-range diffusion of main components of the outer membrane could explain the apparent immobility of the Omp2b clusters, as well as unipolar and mid-cell localizations of newly incorporated outer membrane proteins and lipopolysaccharide. Unipolar growth and limited mobility of surface structures also suggest that new surface variants could arise in a few generations without the need of diluting pre-existing surface antigens.

Spatial Colocation Pattern Mining with the Maximum Membership Threshold.

Spatial Colocation Pattern Mining with the Maximum Membership Threshold.

Neonatal isolation modulates glucocorticoid-receptor function and synaptic plasticity of hippocampal and amygdala neurons in a rat model of single prolonged stress.

Early life and stressful experiences affect hippocampal and amygdala structure and function. They also increase the incidence of mental and nervous system disorders in adults. However, prospective studies have yet to show if early-life experiences affect the risk/severity of post-traumatic stress disorder (PTSD). We applied neonatal isolation (NI) alone, single prolonged stress (SPS) alone and NI + SPS to rats. We evaluated anxiety-like behavior and spatial memory of behavior using open field, elevated plus maze, and Morris water maze tests. Then, we measured expression of glucocorticoid receptors (GRs) and synaptic-related proteins by immunofluorescence, immunohistochemistry and western blotting in the hippocampus and amygdala. NI + SPS exacerbated the increased anxiety levels and impaired spatial memory induced by NI alone or SPS alone. NI alone or SPS alone induced varying degrees of change in expression of GRs and synaptic proteins (synapsin I and postsynaptic density protein-95) in the hippocampus and amygdala. There were opposite changes in GR expression in the hippocampal dentate gyrus and basolateral amygdala. The degree of such change was exacerbated considerably by NI + SPS. In addition, neuroligin (NLG)-1 and NLG-2 were distributed in postsynaptic sites of excitatory and inhibitory synapses, respectively. NI, SPS, and NI + SPS altered the patterns of NLG-1 and NLG-2 colocalization as well as their intensity. NI + SPS strengthened the increased ratio of NLG-1/NLG-2 in the hippocampus, but decreased this ratio in the amygdala. NI and SPS together induced greater degrees of change in anxiety and spatial memory, as well as GR and synaptic protein levels, in the hippocampus and amygdala than the changes induced by NI alone or SPS alone.

A visualization approach for discovering colocation patterns

ABSTRACTColocation mining is one of the major spatial data mining tasks. When discovering colocation patterns, spatial statistics or data mining approaches are commonly used. Colocation mining results are typically presented in a textual form and do not provide any spatial information; thus, the results lack an intuitive approach to obtain cognition of colocation rules. Here, we propose a visualization approach to discover colocation patterns for two independent point distributions and generate visual results. This approach makes use of the ability of human color perception. For two geographic features, our approach first generates density surfaces of the input features and then visualizes the density surfaces using a red or green light with different intensities. Then, based on the law of additive color mixing, our approach mixes the colors of the two density surfaces to generate a colocation rule map. The visualization approach can also provide local details of colocation and be used for local colocation analysis. Users can detect colocation patterns and their distribution from the colocation rule maps. We use both synthetic data and real data to test the performance of our approach.

Autophagy in Myelodysplastic Syndromes: The Role of HIF-1a/REDD1 Molecular Pathway

Background: Hypoxia is a prominent feature of the BM microenvironment, influencing both normal and malignant hematopoiesis. HIF-1α, which is a key regulator of hypoxia responses by mediating the transition to glycolytic metabolism, serves as a cell cycle checkpoint of HSC quiescence and function. It has been proposed that differential HIF-1α protein expression between hypoxic endosteal and less hypoxic vascular niche finely regulates normal hematopoiesis by promoting both quiescence and survival of HSCs, as well as proliferation and differentiation of HPCs. DNA damage response 1 gene (REDD1) is a direct transcriptional target of HIF-1α linking hypoxia to energy regulation and autophagy. Recent evidence suggests that metabolism and autophagy are developmentally programmed and essential for effective hematopoiesis.Aims: To study the implication of HIF-1α/REDD1/autophagy/metabolism axis in differentiation/maturation of hematopoietic BM cells of MDS patients.Methods: BM aspiration and biopsy samples were collected from 15 untreated MDS patients from all subtypes except MDS-RARS and 7 age-matched controls with non-malignant hematologic disorder. Demographic, clinical, laboratory and karyotypic parameters were recorded. BM biopsies were immunohistochemicallly stained by fluorescent-labeled 2-nitroimidazole to assess hypoxic areas in BM. CD34 and myeloid lineage cells were isolated using magnetic beads and ficoll double-layer protocol, respectively. BM cell populations were determined by FACS analysis using standard gating strategies. HIF-1α and REDD1 gene and protein expression was evaluated by qRT-PCR and FACS analysis, respectively. Autophagy was determined by immunofluorescence for LAMP-1/LC3B and immunoblotting for LC3B/p62 (SQSTM1), whereas mitophagy by immunofluorescence for LC3B/TOMM20. Mitochondrial membrane potential (ΔΨ) and mitochondrial mass were analyzed by FACS analysis using mitotrackers. Metabolomic analysis of myeloid lineage cells was performed by liquid chromatography mass spectrometry (LC-MSn). Raw data files were processed using several chemo-informatics tools.Results:We found a preferential strong accumulation of 2-nitroimidazole in intrasinusoidal regions of MDS BM, indicating that hypoxia is a fundamental feature of BM in MDS. We demonstrated a statistically significant REDD1 gene overexpression and an increased intracellular protein co-expression of HIF-1α and REDD1 protein levels in both CD34 and myeloid cells from MDS compared to controls, as determined by RT-qPCR and FACS analysis, respectively. Higher REDD1 protein expression was shown in patients with high grade dysplasia as assessed by the Ogata classification system. Moreover, both CD34 and myeloid cells from MDS demonstrated increased LC3B puncta compared to controls with concurrent staining for CD34 and MPO. The quantitative evaluation of LC3B by Western blot revealed high level of expression of LC3B-II in the MDS myeloid cells compared to controls indicating increased autophagic activity. The observed p62/SQSTM1 degradation along with the colocalization pattern of LC3B/LAMP-1 suggest increased autophagic flux. Metabolomic analysis of MDS myeloid lineage cells compared to controls revealed excessive glycolysis, defective oxidative phosphorylation and increased reductive carboxylation glutaminolysis associated with elevated level of intracellular 2-hydroxyglutarate, all indicative of HIF-1α driven metabolism. The co-localization between TOMM20 marker and autophagosomes in MDS myeloid cells was compatible with increased mitophagy whereas, MDS myeloid cells, were characterized by a reduction of mitochondrial mass and membrane potential in comparison to controls, as determined by FACS analysis.Conclusion:Our results provide evidence for the first time of the hypoxia-driven HIF-1α/REDD1/autophagy axis in the pathophysiology of MDS. Our study suggests that this deregulated pathway is responsible for the production of 2-hydroxyglutarate, an oncometabolite, which is implicated in dysregulated epigenetic homeostasis. All the above may lead to the dysregulated metabolism and differentiation potential of the myeloid cells, thus unraveling a new pathogenetic mechanism for the MDS development. DisclosuresNo relevant conflicts of interest to declare.

Relationships of endocrine cells to each other and to other cell types in the human gastric fundus and corpus.

Gastric endocrine cell hormones contribute to the control of the stomach and to signalling to the brain. In other gut regions, enteroendocrine cells (EECs) exhibit extensive patterns of colocalisation of hormones. In the current study, we characterise EECs in the human gastric fundus and corpus. We utilise immunohistochemistry to investigate EECs with antibodies to ghrelin, serotonin (5-HT), somatostatin, peptide YY (PYY), glucagon-like peptide 1, calbindin, gastrin and pancreastatin, the latter as a marker of enterochromaffin-like (ECL) cells. EECs were mainly located in regions of the gastric glands populated by parietal cells. Gastrin cells were absent and PYY cells were very rare. Except for about 25% of 5-HT cells being a subpopulation of ECL cells marked by pancreastatin, colocalisation of hormones in gastric EECs was infrequent. Ghrelin cells were distributed throughout the fundus and corpus; most were basally located in the glands, often very close to parietal cells and were closed cells i.e., not in contact with the lumen. A small proportion had long processes located close to the base of the mucosal epithelium. The 5-HT cells were of at least three types: small, round, closed cells; cells with multiple, often very long, processes; and a subgroup of ECL cells. Processes were in contact with their surrounding cells, including parietal cells. Mast cells had very weak or no 5-HT immunoreactivity. Somatostatin cells were a closed type with long processes. In conclusion, four major chemically defined EEC types occurred in the human oxyntic mucosa. Within each group were cells with distinct morphologies and relationships to other mucosal cells.

Soluble transforming growth factor beta-1 enhances murine mast cell release of Interleukin 6 in IgE-independent and Interleukin 13 in IgE-dependent settings in vitro.

IntroductionFor immune cells transforming growth factor beta-1 (TGF-β1) can enhance or repress effector functions. Here, we characterize the effects of TGF-β1 on IgE-mediated and IL-33-mediated activation of primary murine mast cells derived from hematopoietic stem cells (bone marrow derived mast cells; BMMC). We also investigated potential interactions between TGF-β1 and stem cell factor (SCF). We conclude TGF-β1 plays a selectively stimulatory role for mast cell cultures in vitro.MethodsBMMCs from C57BL/6 mice were differentiated with IL-3 and then treated with TGF-β1. BMMCs were exposed to TGF-β1, primed with IgE, activated with antigen, and then IL-6 and IL-13 cytokine release was quantified using ELISA. Additionally, the effects of TGF-β1 on both IgE and IL-33-mediated short term activation were observed via flow cytometric analysis of both surface LAMP-1 expression and intracellular IL-6. Receptor colocalization was visualized using fluorescence confocal microscopy and individual receptor expression levels were also quantified.ResultsResting IL-6 production increased with TGF-β1 but significance was lost following BMMC activation via IgE receptor (FcεRI) crosslinking. This was similar to a comparison effect due to SCF treatment alone, which also enhanced resting levels of IL-6. TGF-β1 treatment enhanced release of IL-13 only with FcεRI-IgE-mediated activation. TGF-β1 suppressed mobilization of IL-6 with short-term BMMC activation when stimulated with IL-33. Lastly, colocalization patterns of the SCF receptor (CD117) and FcεRI with IgE crosslinking were unaffected by TGF-β1 treatment, but individual expression levels for FcεRI, CD117, and TGFβRII were all reduced following either IgE activation or TGF-β1 treatment; this reduction was partially recovered in BMMCs that were both activated by IgE and treated with TGF-β1.DiscussionThese data reveal a novel positive effect of soluble TGF-β1 on mast cell activation in vitro, suggesting mast cells may be activated through a non-canonical pathway by TGF-β1. Understanding this interaction will provide insight into the potential role of mast cells in settings where TGF-β1 is produced in an aberrant manner, such as in and around high grade tumors.

Integrative Strategy for Finding Co-Location Patterns between Adverse Birth Outcomes and Industrial Air Pollution

Integrative Strategy for Finding Co-Location Patterns between Adverse Birth Outcomes and Industrial Air Pollution

Parallel approach to incremental co-location pattern mining

Co-location pattern mining is an important area of spatial data mining. In many real word applications, new data is continuously arriving to the system and is stored in spatial databases. As co-location discovery is computationally demanding task, it is crucial to maintain co-location patterns for such dynamic databases without recalculating them from scratch. In this paper we present the first GPU-parallelized solution to this problem. Our contribution is threefold: 1) we present a modified version of EUCOLOC algorithm using the iCPI-tree method called iCPI-EUCOLOC, 2) we modify state-of-the-art MGPUCPM algorithm to implement iCPI-EUCOLOC algorithm on GPUs and 3) we present experimental results showing large performance improvements over the original MGPUCPM algorithm. Our solution allows to reduce user waiting times and is economically beneficial due to the reduced overall computation time.