Partitioned Data Sets Research Articles

Comparative chloroplast genome and species discrimination analyses in Aralia section Aralia (Araliaceae)

Aralia section Aralia is a subset of Araliaceae; it constitutes a small lineage that has been utilized in East Asian traditional medicines for centuries. Identifying species within this section can be challenging due to the complex interspecific morphological variations and the prevalence of hybridization. Accurate species identification is crucial for the application of these medicinal plants. In this study, we evaluated the effectiveness of 11 distinct barcode systems for distinguishing species of Aralia sect. Aralia. Comparative analyses of the chloroplast genomes suggested that the plastome length, gene content and order, as well as the expansion/contraction of the inverted repeat regions, are highly conserved. Species discrimination analysis revealed that complete plastomes and their partitioned data sets possessed considerably high discriminatory power (90.91%), which was equivalent to that of standard barcodes. In contrast, nuclear ribosomal DNA sequences demonstrated a marked improvement (90.91%) over nuclear ribosomal internal transcribed spacer (72.73%) in species discrimination. Evidently, nuclear markers did not perform better than plastid markers. However, a combination of barcode from both nuclear and chloroplast genomes can accurately identify all species for Aralia sect. Aralia. We therefore call for developing multilocus markers from different genomes with different inheritance for species discrimination in plant groups.

Confidential Training and Inference using Secure Multi-Party Computation on Vertically Partitioned Dataset

Digitalization across all spheres of life has given rise to issues like data ownership and privacy. Privacy-Preserving Machine Learning (PPML), an active area of research, aims to preserve privacy for machine learning (ML) stakeholders like data owners, ML model owners, and inference users. The Paper, CoTraIn-VPD, proposes private ML inference and training of models for vertically partitioned datasets with Secure Multi-Party Computation (SPMC) and Differential Privacy (DP) techniques. The proposed approach addresses complications linked with the privacy of various ML stakeholders dealing with vertically portioned datasets. This technique is implemented in Python using open-source libraries such as SyMPC (SMPC functions), PyDP (DP aggregations), and CrypTen (secure and private training). The paper uses information privacy measures, including mutual information and KL-Divergence, across different privacy budgets to empirically demonstrate privacy preservation with high ML accuracy and minimal performance cost.

Computation semantics of the functional scientific workflow language Cuneiform

AbstractCuneiform is a minimal functional programming language for large-scale scientific data analysis. Implementing a strict black-box view on external operators and data, it allows the direct embedding of code in a variety of external languages like Python or R, provides data-parallel higher order operators for processing large partitioned data sets, allows conditionals and general recursion, and has a naturally parallelizable evaluation strategy suitable for multi-core servers and distributed execution environments like Hadoop, HTCondor, or distributed Erlang. Cuneiform has been applied in several data-intensive research areas including remote sensing, machine learning, and bioinformatics, all of which critically depend on the flexible assembly of pre-existing tools and libraries written in different languages into complex pipelines. This paper introduces the computation semantics for Cuneiform. It presents Cuneiform's abstract syntax, a simple type system, and the semantics of evaluation. Providing an unambiguous specification of the behavior of Cuneiform eases the implementation of interpreters which we showcase by providing a concise reference implementation in Erlang. The similarity of Cuneiform's syntax to the simply typed lambda calculus puts Cuneiform in perspective and allows a straightforward discussion of its design in the context of functional programming. Moreover, the simple type system allows the deduction of the language's safety up to black-box operators. Last, the formulation of the semantics also permits the verification of compilers to and from other workflow languages.

Quick and effective multiple contingency screening algorithm based on long‐tailed distribution

This study proposes a quick and effective algorithm for screening severe N − 2 contingency, thus avoiding exhausting analysis of all conceivable N − 2 contingencies. First, the index of overload contribution rate (OCR) is developed on the basis of distribution factor, which is discovered to be long-tailed distributed by statistics. Then, by fully exploiting the discovered feature, the so-called technique of combination of partitioned data sets (TCPDS) is designed and described. The core of the contingency screening algorithm is to apply the TCPDS to the OCR. Moreover, an index system is established to evaluate the proposed algorithm. Finally, the proposed algorithm is proved to be effective and efficient by the index system when applied to some practical power systems.

Theoretical analysis of constructing wavelet synopsis on partitioned data sets

Currently, the size of data becomes much larger and the distributed data processing is getting very important to manage the huge size of data. The MapReduce well known as Google's data processing environment is the most popular distributed platform with good scalability and fault tolerance. Many traditional algorithms in the single machine environment are being adopted to the MapReduce platform. In this paper we analyze a novel algorithm to generate wavelet synopses on the distributed MapReduce framework. Wavelet synopsis is one of the most popular dimensionality reduction methods and has been studied in various areas such as query optimization, approximate query answering, feature selection, etc. In the proposed algorithm, the wavelet synopsis can be calculated by a single MapReduce phase, and, by minimizing the amount of data communicated through the network of the distributed MapReduce platform, all computations are processed within almost linear time complexity. We theoretically study the properties of constructing wavelet synopsis on partitioned data sets and the correctness of the proposed algorithm.

Phylogenetics of a recent radiation in the mallards and allies (Aves: Anas): Inferences from a genomic transect and the multispecies coalescent

Reconstructing species trees by incorporating information from many independent gene trees reduces the confounding influence of stochastic lineage sorting. Such analyses are particularly important for taxa that share polymorphisms due to incomplete lineage sorting or introgressive hybridization. We investigated phylogenetic relationships among 14 closely related taxa from the mallard (Anas spp.) complex using the multispecies coalescent and 20 nuclear loci sampled from a genomic transect. We also examined how treating recombining loci and hybridizing species influences results by partitioning the data using various protocols. In general, topologies were similar among the various species trees, with major clades consistently composed of the same taxa. However, relationships among these clades and among taxa within clades changed among partitioned data sets. Posterior support generally decreased when filtering for recombination, whereas excluding mallards (Anas platyrhynchos) increased posterior support for taxa known to hybridize with them. Furthermore, branch lengths decreased substantially for recombination-filtered data. Finally, concordance between nuclear and morphometric topologies conflicted with those in the mitochondrial tree, particularly with regard to the placement of the Hawaiian duck (A. wyvilliana), Philippine duck (A. luzonica), and two spot-billed ducks (A. zonorhyncha and A. poecilorhyncha). These results demonstrate the importance of maximizing sequence length and taxon sampling when inferring taxonomic relationships that are confounded by extensive allele sharing.

A novel method for multivariate data modelling: Piecewise Generalized EMPR

A multivariate data modelling problem consists of a number of nodes with associated function values. Increase in multivariance urges us to use divide-and-conquer algorithms in modelling process of these problems. High dimensional model representation based methods can partition a given multivariate data set into less-variate data sets and have the ability of building a model through these partitioned data sets. Generalized HDMR (GHDMR) is one of these methods and it is known that it works well for dominantly and purely additive natures. Piecewise Generalized HDMR is an alternative method and was developed to increase the efficiency of GHDMR but the performance of the method for modelling multiplicative natures is still not sufficient and acceptable. This work aims to develop a new piecewise method based on enhanced multivariance product representation which works well for representing multiplicative natures.

Privacy preserving perceptron learning in malicious model

Privacy preserving data mining algorithms are proposed to protect the participating parties’ data privacy in data mining processes. So far, most of these algorithms only work in the semi-honest model that assumes all parties follow the algorithms honestly. In this paper, we propose two privacy preserving perceptron learning algorithms in the malicious model, for horizontally and vertically partitioned data sets, respectively. So far as we know, our algorithms are the first perceptron learning algorithms that can protect data privacy in the malicious model.

Resolving and dating the phylogeny of Cornales – Effects of taxon sampling, data partitions, and fossil calibrations

The order Cornales descends from the earliest split in the Asterid clade of flowering plants. Despite a few phylogenetic studies, relationships among families within Cornales remain unclear. In the present study, we increased taxon and character sampling to further resolve the relationships and to date the early diversification events of the order. We conducted phylogenetic analyses of sequence data from 26S rDNA and six chloroplast DNA (cpDNA) regions using parsimony (MP), maximum likelihood (ML), and Bayesian inference (BI) methods with different partition models and different data sets. We employed relaxed, uncorrelated molecular clocks on BEAST to date the phylogeny and examined the effects of different taxon sampling, fossil calibration, and data partitions. Our results from ML and BI analyses of the combined cpDNA sequences and combined cpDNA and 26S rDNA data suggested the monophyly of each family and the following familial relationships ((Cornaceae–Alangiaceae)–(Curtisiaceae–Grubbiaceae))–(((Nyssaceae–Davidiaceae)–Mastixiaceae)–((Hydrostachyaceae–(Hydrangeaceae–Loasaceae))). These relationships were strongly supported by posterior probability and bootstrap values, except for the sister relationship between the N–D–M and H–H–L clades. The 26S rDNA data and some MP trees from cpDNA and total evidence suggested some alternative alignments for Hydrostachyaceae within Cornales, but results of SH tests indicated that these trees were significantly worse explanations of the total data. Phylogenetic dating with simultaneous calibration of multiple nodes suggested that the crown group of Cornales originated around the middle Cretaceous and rapidly radiated into several major clades. The origins of most families dated back to the late Cretaceous except for Curtisiaceae and Grubbiaceae which may have diverged in the very early Tertiary. We found that reducing sampling density within families and analyzing partitioned data sets from coding and noncoding cpDNA, 26S rDNA, and combined data sets produced congruent estimation of divergence times, but reducing the number and changing positions of calibration points resulted in very different estimations.

A Matrix Based Indexing HDMR method for multivariate data modelling

Modelling multivariate data of real life problems from engineering, chemistry, physics, mathematics or other related sciences, in which function values are known only at arbitrarily distributed points of the problem domain, is an important and complicated issue since there exist mathematical and computational complexities in the analytical structure construction process coming from the multivariance. The Plain High Dimensional Model Representation (HDMR) method expresses a multivariate problem in terms of less-variate problems. In this work, a Matrix Based Indexing HDMR method is developed to make the Plain HDMR philosophy employable for the multivariate data partitioning process. This new method will have the ability of dealing with less-variate data sets by partitioning the given data set into univariate, bivariate and trivariate data sets. Interpolating these partitioned data sets will construct an approximate analytical structure as the model of the given multivariate data modelling problem.

A Morphological Cladistic Analysis of <I>Terpsichore</I> (Polypodiaceae)

Terpsichore, a Neotropical genus of about 70 species, was shown in previous molecular phylogenetic studies to be polyphyletic. The present study assesses this conclusion using morphology and also examines the phylogenetic relationships of five informal species groups originally described within Terpsichore. A morphological matrix was constructed for 109 qualitative characters and 129 terminals. Maximum parsimony was used to analyze the complete data set and three different partitions of that data set which excluded characters of either leaf shape or indumentum position or both. Terpsichore is polyphyletic in all analyses, and the characters originally used to define the genus are either plesiomorphic or homoplastic. Whereas the analysis of the complete matrix recovered a clade similar to the original circumscription, analyses of the partitioned data sets yielded results similar to those of molecular phylogenetic studies. In all analyses, two of the five infrageneric groups of Terpsichore are resolved as monophyletic, whereas the others are either paraphyletic or polyphyletic. In all analyses, species from the Terpsichore taxifolia Group are recovered as diphyletic, a result corresponding to that found in the previous molecular phylogenetic studies. By scoring morphological characters for this study, it was found that two vascular bundles are present in many grammitid petioles and are not unique to Luisma, mesophyll composed of long-armed stellate parenchyma evolves at least three times and is not unique to Enterosora, and perforated dictyosteles are common in grammitids and not restricted to Zygophlebia.

Phylogeny and Biogeography of Dolichoderine Ants: Effects of Data Partitioning and Relict Taxa on Historical Inference

Ants (Hymenoptera: Formicidae) are conspicuous organisms in most terrestrial ecosystems, often attaining high levels of abundance and diversity. In this study, we investigate the evolutionary history of a major clade of ants, the subfamily Dolichoderinae, whose species frequently achieve ecological dominance in ant communities. This group has also produced some of the world's most successful invasive ants. We use an extensive molecular data set ( approximately 9 kb of sequence data from 10 nuclear genes, covering 48 dolichoderine species and 6 outgroup taxa) to infer the phylogenetic relationships, divergence dates, and biogeographic history of these ants. We evaluate the effects of data partitioning and outgroup composition on phylogenetic inference by estimating relationships under a series of increasingly partitioned data sets and by running analyses both with and without Aneuretus simoni, a rare and localized species that is the nearest living relative of Dolichoderinae. We also examine the effects of excluding 2 data partitions with significant base composition heterogeneity. Our results reveal 4 well-supported and mutually exclusive clades of dolichoderines, corresponding to 4 newly defined tribes: Bothriomyrmecini (B), Dolichoderini (D), Leptomyrmecini (L), and Tapinomini (T). All Bayesian and likelihood analyses yield the same unrooted (ingroup-only) topology, ((D,L),(B,T)), with the outgroups attaching either on the Dolichoderini branch or on the Tapinomini branch. Placement of the root is highly sensitive to choice of model partition and to inclusion/exclusion of Aneuretus. Bayes' factors strongly favor the more partitioned models, and in these Tapinomini is recovered as sister to the remaining dolichoderines, but only if Aneuretus is included. Exclusion of Aneuretus precludes recovery of this topology in all but the most highly partitioned Bayesian analyses and then only with nonsignificant support, underscoring the importance of relict, taxonomically isolated taxa for phylogenetic inference. Removal of 2 partitions with heterogeneous base composition also markedly increases support for placement of the root on the Tapinomini branch. Our divergence date estimates and biogeographic analyses indicate that crown-group dolichoderines arose about 65 million years ago (Ma), although this was preceded by a substantial period (30 million years) of stem group evolution. The 4 extant tribes are estimated to have crown-group origins in the late Paleocene or Eocene (40-60 Ma). Tapinomini and Bothriomyrmecini originated in the Paleotropics and subsequently dispersed to other biogeographic regions. Crown-group Leptomyrmecini arose and diversified in the Neotropics, but they also gave rise to one clade that colonized Australia about 30 Ma and subsequently experienced a massive radiation on that continent. This event occurred later than the diversification of dolichoderines in the northern hemisphere, so that by the time dolichoderines came to dominate the Australian fauna they had already declined in abundance in the Holarctic region.

Using plastid genome-scale data to resolve enigmatic relationships among basal angiosperms

Although great progress has been made in clarifying deep-level angiosperm relationships, several early nodes in the angiosperm branch of the Tree of Life have proved difficult to resolve. Perhaps the last great question remaining in basal angiosperm phylogeny involves the branching order among the five major clades of mesangiosperms (Ceratophyllum, Chloranthaceae, eudicots, magnoliids, and monocots). Previous analyses have found no consistent support for relationships among these clades. In an effort to resolve these relationships, we performed phylogenetic analyses of 61 plastid genes ( approximately 42,000 bp) for 45 taxa, including members of all major basal angiosperm lineages. We also report the complete plastid genome sequence of Ceratophyllum demersum. Parsimony analyses of combined and partitioned data sets varied in the placement of several taxa, particularly Ceratophyllum, whereas maximum-likelihood (ML) trees were more topologically stable. Total evidence ML analyses recovered a clade of Chloranthaceae + magnoliids as sister to a well supported clade of monocots + (Ceratophyllum + eudicots). ML bootstrap and Bayesian support values for these relationships were generally high, although approximately unbiased topology tests could not reject several alternative topologies. The extremely short branches separating these five lineages imply a rapid diversification estimated to have occurred between 143.8 +/- 4.8 and 140.3 +/- 4.8 Mya.

Circumscription and phylogeny of Apiaceae subfamily Saniculoideae based on chloroplast DNA sequences

An estimate of phylogenetic relationships within Apiaceae subfamily Saniculoideae was inferred using data from the chloroplast DNA trnQ- trnK 5′-exon region to clarify the circumscription of the subfamily and to assess the monophyly of its constituent genera. Ninety-one accessions representing 14 genera and 82 species of Apiaceae were examined, including the genera Steganotaenia, Polemanniopsis, and Lichtensteinia which have been traditionally treated in subfamily Apioideae but determined in recent studies to be more closely related to or included within subfamily Saniculoideae. The trnQ- trnK 5′-exon region includes two intergenic spacers heretofore underutilized in molecular systematic studies and the rps16 intron. Analyses of these loci permitted an assessment of the relative utility of these noncoding regions (including the use of indel characters) for phylogenetic study at different hierarchical levels. The use of indels in phylogenetic analyses of both combined and partitioned data sets improves resolution of relationships, increases bootstrap support values, and decreases levels of overall homoplasy. Intergeneric relationships derived from maximum parsimony, Bayesian, and maximum likelihood analyses, as well as from maximum parsimony analysis of indel data alone, are fully resolved and consistent with one another and generally very well supported. We confirm the expansion of subfamily Saniculoideae to include Steganotaenia and Polemanniopsis (as the new tribe Steganotaenieae C.I. Calviño and S.R. Downie) but not Lichtensteinia. Sister group to tribe Steganotaenieae is tribe Saniculeae, redefined to include the genera Actinolema, Alepidea, Arctopus, Astrantia, Eryngium, Petagnaea, and Sanicula. With the synonymization of Hacquetia into Sanicula, all genera are monophyletic. Eryngium is divided into “Old World” and “New World” subclades and within Astrantia sections Astrantia and Astrantiella are monophyletic.

Developing parallel sequential minimal optimization for fast training support vector machine

A parallel version of sequential minimal optimization (SMO) is developed in this paper for fast training support vector machine (SVM). Up to now, SMO is one popular algorithm for training SVM, but it still requires a large amount of computation time for solving large size problems. The parallel SMO is developed based on message passing interface (MPI). Unlike the sequential SMO which handle all the training data points using one CPU processor, the parallel SMO first partitions the entire training data set into smaller subsets and then simultaneously runs multiple CPU processors to deal with each of the partitioned data sets. Experiments show that there is great speedup on the adult data set, the MNIST data set and IDEVAL data set when many processors are used. There are also satisfactory results on the Web data set. This work is very useful for the research where multiple CPU processors machine is available.

Parallel Sequential Minimal Optimization for the Training of Support Vector Machines

Sequential minimal optimization (SMO) is one popular algorithm for training support vector machine (SVM), but it still requires a large amount of computation time for solving large size problems. This paper proposes one parallel implementation of SMO for training SVM. The parallel SMO is developed using message passing interface (MPI). Specifically, the parallel SMO first partitions the entire training data set into smaller subsets and then simultaneously runs multiple CPU processors to deal with each of the partitioned data sets. Experiments show that there is great speedup on the adult data set and the Mixing National Institute of Standard and Technology (MNIST) data set when many processors are used. There are also satisfactory results on the Web data set.

Cancer gene search with data-mining and genetic algorithms

Cancer leads to approximately 25% of all mortalities, making it the second leading cause of death in the United States. Early and accurate detection of cancer is critical to the well being of patients. Analysis of gene expression data leads to cancer identification and classification, which will facilitate proper treatment selection and drug development. Gene expression data sets for ovarian, prostate, and lung cancer were analyzed in this research. An integrated gene-search algorithm for genetic expression data analysis was proposed. This integrated algorithm involves a genetic algorithm and correlation-based heuristics for data preprocessing (on partitioned data sets) and data mining (decision tree and support vector machines algorithms) for making predictions. Knowledge derived by the proposed algorithm has high classification accuracy with the ability to identify the most significant genes. Bagging and stacking algorithms were applied to further enhance the classification accuracy. The results were compared with that reported in the literature. Mapping of genotype information to the phenotype parameters will ultimately reduce the cost and complexity of cancer detection and classification.

지수 평활법을 적용한 시간 연관 규칙

전자상거래가 활성화됨에 따라 고객 개인의 관심에 부합하는 개인화된 정보나 상품 서비스를 제공하기 위하여 시간에 따라 분할하여 연산하는 시간 연관 규칙이 최근에 등장하고 있다. 본 논문은 일반적으로 정의된 연관 규칙에 대해 시간의 변화를 고려하기 위하여 최신 데이터에 가중치를 높여 주는 지수 평활법을 적용한 연관 규칙을 정의하고 이로 탐사하는 알고리즘을 제안한다. 시뮬레이션과 적용사례를 통하여 시간에 따라 분할하여 지수 평활법을 적용한 시간 연관 규칙이 기존의 것보다 실행시간은 다소 많지만 시간을 고려한 정확한 탐색률을 갖으므로 전자 상점 현장 응용에 효과적임을 확인하였다. As electronic commerce progresses, the temporal association rule is developed from partitioned data sets by time to offer personalized services for customer's interest. In this paper, we proposed a temporal association rule with exponential smoothing method that is giving higher weights to recent data than past data. Through simulation and case study, we confirmed that it is more precise than existing temporal association rules but consumes running time.

Polyphyly of the spring-parsleys (Cymopterus): molecular and morphological evidence suggests complex relationships among the perennial endemic genera of western North American Apiaceae

Cladistic analyses of DNA sequences from the nuclear rDNA internal transcribed spacer region and cpDNA rps16 intron and, for a subset of taxa, the cpDNA trnF-trnL-trnT locus were carried out to evaluate the monophyly of Cymopterus and to ascertain its phylogenetic placement among the other perennial genera of Apiaceae (Umbelliferae) subfamily Apioideae endemic to western North America. To elucidate patterns in the evolution of specific fruit characters and to evaluate their utility in circumscribing genera unambiguously, additional evidence was procured from cross-sections of mature fruits and the results of cladistic analysis of 25 morphological characters. Analyses of the partitioned data sets resulted in weakly supported and largely unresolved phylogenetic hypotheses, possibly due to the rapid radiation of the group, whereas the combined analysis of all molecular evidence resulted in a well-resolved phylogeny with higher bootstrap support. The traditionally used fruit characters of wing shape and composition and orientation of mericarp compression are highly variable. The results of these analyses reveal that Cymopterus and Lomatium, the two largest genera of western North American Apiaceae, are polyphyletic, and that their species are inextricably linked with those of other endemic perennial genera of the region (such as, Aletes, Musineon, Oreoxis, Pseudocymopterus, Pteryxia, and Tauschia), many of which are also not monophyletic. Prior emphasis on characters of the fruit in all systems of classification of the group has led to highly artificial assemblages of species. A complete reassessment of generic limits of all western endemic Apiaceae is required, as is further systematic study of this intractable group.Key words: Apiaceae, Cymopterus, phylogeny, ITS, rps16 intron, morphology.

Exploring Component Stability Using Life-Stage Concordance In Sabethine Mosquitoes (Diptera: Culicidae).

- Morphological characters from sabethine mosquitoes were coded from larvae, pupae and adults, and life-stage partitions were evaluated to determine the contribution of each to the topology of a combined cladogram. Initial tests failed to find congruence between characters partitioned by life stage. However, when components from the combined analysis were tested using reduced taxon sets, a high degree of concordance between partitions was observed. A procedure for assessing individual life-stage contribution is employed, in which exhaustive searches are used to explore all possible arrangements for each of the selected components. Seven of the 10 components examined were able to recover the combined topology with a reduced taxon set. Congruent arrangements of taxa were typically observed for two or more life stages, although partitioned data were less resolved and frequently included aberrant topologies (those not supported by other partitioned or combined reduced taxon tree sets). In addition, none of the partitioned data sets gave robust results for all tests, suggesting that studies which emphasize character data from single life stages may support misleading arrangements of taxa. One component on the combined cladogram was not supported by any of the life-stage partitions when analysed separately. These results are complementary to "total evidence" approach, and demonstrate that partitions of data are useful for examining suits of characters which may cause some components of the "total.