Component Discriminant Analysis Research Articles

Image Retrieval via Canonical Correlation Analysis and Binary Hypothesis Testing

Canonical Correlation Analysis (CCA) is a classic multivariate statistical technique, which can be used to find a projection pair that maximally captures the correlation between two sets of random variables. The present paper introduces a CCA-based approach for image retrieval. It capitalizes on feature maps induced by two images under comparison through a pre-trained Convolutional Neural Network (CNN) and leverages basis vectors identified through CCA, together with an element-wise selection method based on a Chernoff-information-related criterion, to produce compact transformed image features; a binary hypothesis test regarding the joint distribution of transformed feature pair is then employed to measure the similarity between two images. The proposed approach is benchmarked against two alternative statistical methods, Linear Discriminant Analysis (LDA) and Principal Component Analysis with whitening (PCAw). Our CCA-based approach is shown to achieve highly competitive retrieval performances on standard datasets, which include, among others, Oxford5k and Paris6k.

Inferring the biogeography of New Zealand’s only endemic holoparasitic plant, the threatened Dactylanthus taylorii (Mystropetalaceae)

ABSTRACT Despite a rich history of botanical research, little is known about the distribution of genetic variation within New Zealand plant species. With a native vascular flora that consists of ∼80% endemics, nearly one-half of which are threatened, there is a dire need for a better understanding of intraspecific diversity for individual plant species and of biogeographic patterns for the wider New Zealand flora. To these ends, we have investigated population genetic structure in the endemic root holoparasite, Dactylanthus taylorii (pua o te Rēinga, Mystropetalaceae). A total of 235 individuals from 31 locations across the North Island of New Zealand (Te Ika-a-Māui, Aotearoa) were genotyped for a suite of nuclear microsatellite loci. Genotype data were analysed at the individual and population levels to assess genetic diversity within and among populations and to investigate the geographic distribution of genetic variation within the species. Population structure was assessed using Bayesian clustering, discriminant analysis of principal components, and a Neighbor-Net analysis. A hierarchical pattern of genetic diversity was revealed, with evidence for three broad geographic regional groupings (West, East, and Central Volcanic Plateau (CVP)). Upon further analysis, each of the three geographic groups further subdivide hierarchically into two or three genetic subgroups. Genetic diversity was found to be relatively high in this endangered species but varies considerably across populations. The distribution of genetic groups appears to have been influenced by past volcanic activity as the CVP group is largely coincident with a distinct tephra isopach from the 1.8 ka Taupō eruption and the West and East groups may have diverged due to recurring volcanic activity in the Central Volcanic Zone. The genetic structure resolved here provides important information for the management of D. taylorii specifically and for our understanding of New Zealand biogeography generally.

Elemental Chemometrics as Tools to Depict Stalked Barnacle (Pollicipes pollicipes) Harvest Locations and Food Safety.

The stalked barnacle Pollicipes pollicipes is an abundant species on the very exposed rocky shore habitats of the Spanish and Portuguese coasts, constituting also an important economical resource, as a seafood item with high commercial value. Twenty-four elements were measured by untargeted total reflection X-ray fluorescence spectroscopy (TXRF) in the edible peduncle of stalked barnacles sampled in six sites along the Portuguese western coast, comprising a total of 90 individuals. The elemental profile of 90 individuals originated from several geographical sites (N = 15 per site), were analysed using several chemometric multivariate approaches (variable in importance partial least square discriminant analysis (VIP-PLS-DA), stepwise linear discriminant analysis (S-LDA), linear discriminant analysis (LDA), random forests (RF) and canonical analysis of principal components (CAP)), to evaluate the ability of each approach to trace the geographical origin of the animals collected. As a suspension feeder, this species introduces a high degree of background noise, leading to a comparatively lower classification of the chemometric approaches based on the complete elemental profile of the peduncle (canonical analysis of principal components and linear discriminant analysis). The application of variable selection approaches such as the VIP-PLS-DA and S-LDA significantly increased the classification accuracy (77.8% and 84.4%, respectively) of the samples according to their harvesting area, while reducing the number of elements needed for this classification, and thus the background noise. Moreover, the selected elements are similar to those selected by other random and non-random approaches, reinforcing the reliability of this selection. This untargeted analytical procedure also allowed to depict the degree of risk, in terms of human consumption of these animals, highlighting the geographical areas where these delicacies presented lower values for critical elements compared to the standard thresholds for human consumption.

Low genetic variation of foliar traits among Prosopis chilensis (Leguminosae) provenances.

Prosopis chilensis (Molina) Stuntz (Leguminosae) is a valuable native species in Argentina that has been proposed to be used in reforestation, afforestation and restoration programmes. Natural provenances show important differentiation in height, shape, spine size, fruits and foliar traits throughout their distribution in the semiarid Monte ecoregion. The goal of this work was to characterize the genetic basis of the leaf variation in P. chilensis aiming to contribute to the improvement management program. We analyzed morphological variation and estimate narrow sense heritability for ten quantitative traits from a provenance-progeny trial founded from open pollinated families. We assessed the variance components by a generalized linear mixed model. Differences among provenances were quantified through univariate QST statistics and multivariate discriminant analysis of principal components. Finally, univariate and multivariate neutrality test were conducted to unveil the evolutionary forces that shape the variation. Univariate and multivariate analysis showed low genetic variation in foliar traits among provenances grown in the common garden. Consistently, the QST estimates for each trait were low. Both, the univariate (QST-FST comparison) and the multivariate neutrality test suggest that the leaf variation among provenances may be shaped by genetic drift rather than selective forces. Heritability estimates were significant only for leaflet apex and leaflet apex/leaflet area. Since genetic variation for most foliar traits among provenances estimated under controlled environmental conditions were very low or absent, the variation described in the wild would be explained merely by plastic response to varying environments. These results are discussed in terms of adaptive strategies and the use of different provenances as seed sources within the framework of the improvement program. It is expected that P. chilensis seeds or seedlings from trees selected under economical criteria will be able to develop in different areas thanks to the phenotypic plasticity of leaf traits.

Further Evidence of Population Admixture in the Serbian Honey Bee Population.

Simple SummaryThe western honey bee is one of the most ecologically and economically important pollinator species. Due to human interference, it faces serious challenges, not only in number decline and habitat loss, but also in natural subspecies diversity and distribution. The conservation of genetic diversity and perseverance of locally adapted populations and subspecies becomes a crucial task in the face of rapid environmental changes. In order to further assess present genetic variability in Serbian honey bee populations, we analyzed 14 microsatellite loci and then compared nine of them with previously published data. Our results suggest that Serbia now harbors a distinct, relatively homogenous honey bee population, although some local differences are still preserved.Socioeconomic interests and beekeeper preferences have often taken precedence over the conservation of locally native honey bee subspecies, leading to the predominance of admixture populations in human-dominated areas. To assess the genetic diversity of contemporary managed Serbian honey bee colonies, we used 14 microsatellite loci and analyzed 237 worker bees from 46 apiaries in eight localities of northern and southern Serbia. Furthermore, we compared data for nine microsatellite loci with 338 individuals from Italy, Hungary, Poland, and Spain. The standard parameters of genetic diversity in Serbian honey bee populations were in line with other analyses, although somewhat smaller. STRUCTURE analysis showed the existence of two equally distributed genetic clusters and Analysis of molecular variances could not confirm the presence of a geographically discrete population but showed local differences. Discriminant analysis of principal components showed overlapping of worker bees from different parts of Serbia. Clear genetic differentiation can be observed when comparing all populations between geographical regions and their corresponding subspecies. The absence of the A. m. macedonica subspecies from its historical distribution range in southern Serbia as well as the lack of distinctive geographical groups suggest that selective breeding, queen import, and migratory beekeeping practices strongly influence the genetic structure and diversity of honey bees, leading to the genetic uniformization and creation of the admixture population.

Genetic Diversity and Population Structure of Doum Palm (Hyphaene compressa) Using Genotyping by Sequencing.

Doum palm (Hyphaene compressa) is a perennial economic plant primarily growing in Kenya’s Arid and Semi-Arid Lands (ASALs). It is heavily relied upon for food, animal feed, construction materials and medicine, making it an ideal plant for resource sustainability. However, the limited information on its genetic resources has hindered its breeding and conservation studies. This study used the genotyping by sequencing approach to identify Single Nucleotide Polymorphisms. These SNPs were further used to assess the genetic diversity and population structure of 96 H. compressa accessions from Coastal, Northern and Eastern ASAL regions of Kenya using two approaches; reference-based and de novo-based assemblies. STRUCTURE analysis grouped the sampled accessions into two genetic clusters (Cluster 1 and Cluster 2). Cluster 1 included accessions from the Northern region, whereas Cluster 2 included all accessions from Eastern and Coastal regions. Accessions from Kwale (Coastal) had mixed ancestry from both Cluster 1 and Cluster 2. These STRUCTURE findings were further supported by principal components analysis, discriminant analysis of principal components and phylogenetic analysis. Analysis of molecular variance indicated greater genetic variation within populations (92.7%) than among populations (7.3%). An overall FST of 0.074 was observed, signifying moderate genetic differentiation among populations. The results of this study will provide information useful in breeding, marker-assisted selection and conservation management of H. compressa.

Phylogeography of a canopy-forming kelp, Eisenia bicyclis (Laminariales, Phaeophyceae), based on a genome-wide sequencing analysis.

Analyses of phylogeographic patterns and genetic diversity provide fundamental information for the management and conservation of species. However, little is published about these patterns in Japanese kelp species. In this study, we conducted phylogeographic analyses of a canopy-forming kelp, Eisenia bicyclis, based on genome-wide SNPs identified by ddRAD-seq. We obtained 1,299 SNPs for 76 samples from nine localities across the distribution. STRUCTURE, NeighborNet, and discriminant analysis of principal components consistently showed high genetic differentiation among the Eastern Pacific, Central Pacific, and Sea of Japan coastal regions. Relatively strong gene flow was detected only within populations in the Eastern Pacific and in the Sea of Japan. Genetic diversity and genetic uniqueness were high in the Central Pacific and low in the Sea of Japan. These results suggest that there were at least three independent refugia corresponding to the three regions during the Last Glacial Maximum (LGM). Furthermore, relatively larger populations in the Central Pacific and smaller populations in the Sea of Japan have been maintained in the demographic history from before the LGM to the present. These phylogeographic histories were supported by an Approximate Bayesian Computation analysis. From a conservation genetics perspective, the loss of southern populations in the Central Pacific would greatly reduce the total genetic diversity of the species. Southern populations in the Sea of Japan, which have relatively low genetic diversity, may be highly vulnerable to environmental change, such as heat waves and increased feeding. Therefore, careful monitoring and conservation are needed in the two regions.

Environmental differences explain subtle yet detectable genetic structure in a widespread pollinator

BackgroundThe environment is a strong driver of genetic structure in many natural populations, yet often neglected in population genetic studies. This may be a particular problem in vagile species, where subtle structure cannot be explained by limitations to dispersal. Consequently, these species might falsely be considered quasi-panmictic and hence potentially mismanaged. A species this might apply to, is the buff-tailed bumble bee (Bombus terrestris), an economically important and widespread pollinator, which is considered to be quasi-panmictic at mainland continental scales. Here we aimed to (i) quantify genetic structure in 21+ populations of the buff-tailed bumble bee, sampled throughout two Eastern European countries, and (ii) analyse the degree to which structure is explained by environmental differences, habitat permeability and geographic distance. Using 12 microsatellite loci, we characterised populations of this species with Fst analyses, complemented by discriminant analysis of principal components and Bayesian clustering approaches. We then applied generalized dissimilarity modelling to simultaneously assess the informativeness of geographic distance, habitat permeability and environmental differences among populations in explaining divergence.ResultsGenetic structure of the buff-tailed bumble bee quantified by means of Fst was subtle and not detected by Bayesian clustering. Discriminant analysis of principal components suggested insignificant but still noticeable structure that slightly exceeded estimates obtained through Fst analyses. As expected, geographic distance and habitat permeability were not informative in explaining the spatial pattern of genetic divergence. Yet, environmental variables related to temperature, vegetation and topography were highly informative, explaining between 33 and 39% of the genetic variation observed.ConclusionsIn contrast to previous studies reporting quasi-panmixia in continental populations of this species, we demonstrated the presence of subtle population structure related to environmental heterogeneity. Environmental data proved to be highly useful in unravelling the drivers of genetic structure in this vagile and opportunistic species. We highlight the potential of including these data to obtain a better understanding of population structure and the processes driving it in species considered to be quasi-panmictic.

The Genetic Diversity of Enset (Ensete ventricosum) Landraces Used in Traditional Medicine Is Similar to the Diversity Found in Non-medicinal Landraces.

Enset (Ensete ventricosum) is a multipurpose crop extensively cultivated in southern and southwestern Ethiopia for human food, animal feed, and fiber. It has immense contributions to the food security and rural livelihoods of 20 million people. Several distinct enset landraces are cultivated for their uses in traditional medicine. These landraces are vulnerable to various human-related activities and environmental constraints. The genetic diversity among the landraces is not verified to plan conservation strategy. Moreover, it is currently unknown whether medicinal landraces are genetically differentiated from other landraces. Here, we characterize the genetic diversity of medicinal enset landraces to support effective conservation and utilization of their diversity. We evaluated the genetic diversity of 51 enset landraces, of which 38 have reported medicinal value. A total of 38 alleles across the 15 simple sequence repeat (SSR) loci and a moderate level of genetic diversity (He = 0.47) were detected. Analysis of molecular variation (AMOVA) revealed that only 2.4% of the total genetic variation was contributed by variation among the medicinal and non-medicinal groups of landraces, with an FST of 0.024. A neighbor-joining tree showed four separate clusters with no correlation to the use-values of the landraces. Except for two, all “medicinal” landraces with distinct vernacular names were found to be genetically different, showing that vernacular names are a good indicator of genetic distinctiveness in these specific groups of landraces. The discriminant analysis of the principal components also confirmed the absence of distinct clustering between the two groups. We found that enset landraces were clustered irrespective of their use-value, showing no evidence for genetic differentiation between the enset grown for ‘medicinal’ uses and non-medicinal landraces. This suggests that enset medicinal properties may be restricted to a more limited number of genotypes, might have resulted from the interaction of genotype with the environment or management practice, or partly misreported. The study provides baseline information that promotes further investigations in exploiting the medicinal value of these specific landraces.

Integrating morphological and genetic data at different climate and geographic scales in an endangered freshwater musselUnio durieui(Bivalvia: Unionidae) endemic to northern Tunisia

ABSTRACTThe freshwater mussel Unio durieui is a rare bivalve endemic to North African watersheds. Although this is an important species from an ecological and conservation perspective, information on its genetic diversity and population structure is lacking. Here, we combine geometric morphometric and molecular genetic analyses to characterize and evaluate the status of populations of this endangered species from four Tunisian watershed basins (Cap Bon, Ichkeul, northwest and Mejerda River basins). By using a multiparameter approach, we sought to test the hypothesis that populations from these geographic regions could be considered as distinct management units. Shell outline shapes were assessed using elliptic Fourier analysis followed by linear discriminant analysis of principal components. Multivariate analysis applied on normalized Fourier coefficients showed highly significant differences in shell shape among the populations. Population genetic analysis based on inter-simple sequence repeats revealed low levels of diversity and high differentiation between the populations studied. Two groups of populations are clearly evident, thus separating samples from the Northwest from those from the other watersheds. Although there is no evidence for isolation by distance patterns, genetic and phenotypic diversity appeared significantly correlated with climatic variables. These findings suggest isolation by environment and the nominal species U. durieui may constitute two biological species. Conservation programme actions should consider this substructure in establishing effective recovery strategies for this endangered species.

High-throughput single nucleotide polymorphism genotyping reveals population structure and genetic diversity of tall fescue (

Single nucleotide polymorphisms (SNPs) are the most abundant form of genetic variation in eukaryote genomes and may be useful for selection of genetically distant parents for crossing in breeding programs. In this study, genome-wide SNPs of tall fescue (Festuca arundinacea Schreb.) were genotyped by using double-digest restriction-site-associated DNA sequencing to determine the genetic variation and differentiation among and within forage, turf and hybrid populations. After filtering, 16 036 SNPs were used to investigate genetic diversity and for structure analysis. SNP markers clearly differentiated the populations from each other. However, the hybrid population had a higher genetic similarity with the turf population than with the forage population. The results were confirmed by genetic differentiation (fixation index, Fst) and gene flow (Nm) statistics, so that low Fst and high Nm were observed between turf and hybrid groups, indicating less genetic distance and a high similarity between them. Based on the results of the weighted pair group method with arithmetic mean clustering, discriminant analysis of principal components and analysis of molecular variance, greater genetic differentiations were found among diverse turf, forage and hybrid populations, especially between turf and forage ones. In conclusion, numerous informative SNPs, natural allelic diversity-led domestication patterns, basic genetic variation statistics (e.g. Fst and heterozygosity) and population structures have multidimensional applicability for tall fescue genomics-assisted breeding.

Discrimination of Pineapples Varieties based on Volatile Organic Compounds using Chromatographic and Chemometrics Analysis

This study analyzed the volatile organic compounds (VOCs) of four different pineapple varieties (MD2, Sarawak, Josephine and Morris). Head Space-Solid Phase Micro Extraction (HS-SPME) was used to extract the VOCs of those pineapples using Gas Chromatography-Mass Spectrometry (GC-MS). Analysis of pineapples extracts identified many types of VOCS including esters, alcohols, terpenes, aldehydes, ketones, saturated hydrocarbons and phenols. Thirtyfive selected VOCs were used for varietal discrimination using chemometrics techniques namely principal components analysis (PCA), cluster analysis (CA) and discriminant analysis (DA). The application of PCA and CA was able to group the pineapple samples according to their varieties based on their respective VOCs. Standard mode DA had discriminated the samples with 98.75% correctness using all 35 VOCs. The identification of significant compounds found in the studied samples was successfully confirmed (p<0.05) by PCA. The total variance of PC1 and PC2 is 63.34% exhibiting that the ester compounds show strong and moderate loading values. One of the unique identifiers is the presence of β-Myrcene with a loading value 0.93 found in PC1 where it is described as pleasant terpenes odor and balsamic descriptor. GCMS and chemometric techniques provided meaningful chemically quantitative data.

Evaluating the legacy of multiple introductions of American martens on spatiotemporal patterns of genetic diversity

AbstractSpecies reintroductions are successful when established populations maintain both demographic stability and genetic diversity. Such a result may be obtained by ensuring both structural habitat connectivity and genetic connectivity among reintroduced and remnant populations. Nevertheless, prezygotic barriers such as assortative mating can prevent the flow of genetic material between populations, even when migration between populations is high. Limited gene flow may be particularly relevant for reintroductions that were sourced either from captive-bred populations or from disparate locations in the wild. American martens (Martes americana) have been reintroduced repeatedly in the Upper Midwestern United States in an effort to establish self-sustaining populations. We quantified levels of genetic diversity within and spatial genetic variance among four marten populations during two time periods separated by 10 years. Spatially informed and naïve discriminant analysis of principal components were used to assign individuals to populations. Results indicate that heterozygosity declined and inbreeding coefficients increased between the two collection periods, while genetic structure among populations also increased. Data are consistent with assortative mating contributing to reapportioning of genetic variation. Population assignment tests show that migration among populations is apparent, but admixture (based on cluster membership probabilities) is low and declined over time. Specifically, martens may be successfully dispersing between populations but a lack of admixture indicates a lack of reproductive contributions to genetic diversity by migrants. Because marten reintroductions in this region are well-documented and well-monitored, lessons can be derived from results to inform future reintroductions. We encourage a careful balance of supplementing genetic diversity via augmentation while avoiding translocation of animals from disparate populations that may result in reproductive isolation of migrants. In combination with the maintenance of a functionally connected landscape, this strategy would maximize the likelihood of a successful reintroduction in terms both of demography and genetics.

Phenotyping and Genotyping Analyses Reveal the Spread of Puccinia striiformis f. sp. tritici Aeciospores From Susceptible Barberry to Wheat in Qinghai of China.

Puccinia striiformis f. sp. tritici Eriks., the cause of wheat yellow or stripe rust on wheat, undergoes sexual reproduction on barberry, but it is unclear if barberry plays any role in stripe rust epidemics under natural conditions. P. striiformis f. sp. tritici was isolated from its alternate host barberry (Berberis spp.) and primary host wheat in the vicinity of barberry by inoculation of aeciospores and urediniospores on Mingxian 169 cultivar in Qinghai province of China in 2018. The P. striiformis f. sp. tritici isolates from barberry and wheat were characterized to virulence patterns by inoculation on 24 differentials bearing Yr gene under control conditions and analyzed using 12 polymorphic simple sequence repeat (SSR) markers. The occurrence frequency of P. striiformis f. sp. tritici on barberry was 1.87% by inoculation aecia, collected from barberry on Mingxian 169 of wheat. A close virulence relationship was presented between P. striiformis f. sp. tritici isolates from both barberry and wheat based on virulence simple matching coefficient and principal coordinates analysis (PCoA). Additionally, the same genetic ancestry, based on structure analysis by STRUCTURE program and genetic relationship analyses using discriminant analysis of principal components and PCoA, was shared between P. striiformis f. sp. tritici isolates from barberry and those from wheat. Together, all the results indicated that the role of barberry in providing aeciospores as an inoculum source causing wheat stripe rust epidemic in Qinghai in spring is of considerable importance.

Genotyping by sequencing suggests overwintering of Peronospora destructor in southwestern Québec, Canada.

Several Peronospora species are carried by wind over short and long distances, from warmer climates where they survive on living plants to cooler climates. In eastern Canada, this annual flow of sporangia was thought to be the main source of Peronospora destructor responsible for onion downy mildew. However, the results of a recent study showed that the increasing frequency of onion downy mildew epidemics in eastern Canada is associated with warmer autumns, milder winters, and previous year disease severity, suggesting overwintering of the inoculum in an area where the pathogen is not known to be endogenous. In this study, genotyping by sequencing was used to investigate the population structure of P. destructor at the landscape scale. The study focused on a particular region of southwestern Québec—Les Jardins de Napierville—to determine if the populations were clonal and regionally differentiated. The data were characterized by a high level of linkage disequilibrium, characteristic of clonal organisms. Consequently, the null hypothesis of random mating was rejected when tested on predefined or nonpredefined populations, indicating that linkage disequilibrium was not a function of population structure and suggesting a mixed reproduction mode. Discriminant analysis of principal components performed with predefined population assignment allowed grouping P. destructor isolates by geographical regions, while analysis of molecular variance confirmed that this genetic differentiation was significant at the regional level. Without using a priori population assignment, isolates were clustered into four genetic clusters. These results represent a baseline estimate of the genetic diversity and population structure of P. destructor.

Exhaled Metabolite Patterns to Identify Recent Asthma Exacerbations.

Asthma is a chronic respiratory disease that can lead to exacerbations, defined as acute episodes of worsening respiratory symptoms and lung function. Predicting the occurrence of these exacerbations is an important goal in asthma management. The measurement of exhaled breath by electronic nose (eNose) may allow for the monitoring of clinically unstable asthma and exacerbations. However, data on its ability to perform this is lacking. We aimed to evaluate whether eNose could identify patients that recently had asthma exacerbations. We performed a cross-sectional study, measuring exhaled breath using the SpiroNose in adults with a physician-reported diagnosis of asthma. Patients were randomly divided into a training (n = 252) and validation (n = 109) set. For the analysis of eNose signals, principal component (PC) and linear discriminant analysis (LDA) were performed. LDA, based on PC1-4, reliably discriminated between patients who had a recent exacerbation from those who had not (training receiver operating characteristic (ROC)–area under the curve (AUC) = 0.76,95% CI 0.69–0.82), (validation AUC = 0.76, 95% CI 0.64–0.87). Our study showed that, exhaled breath analysis using eNose could accurately identify asthma patients who recently had an exacerbation, and could indicate that asthma exacerbations have a specific exhaled breath pattern detectable by eNose.

Integrating neural and ocular attention reorienting signals in virtual reality

Objective. Reorienting is central to how humans direct attention to different stimuli in their environment. Previous studies typically employ well-controlled paradigms with limited eye and head movements to study the neural and physiological processes underlying attention reorienting. Here, we aim to better understand the relationship between gaze and attention reorienting using a naturalistic virtual reality (VR)-based target detection paradigm. Approach. Subjects were navigated through a city and instructed to count the number of targets that appeared on the street. Subjects performed the task in a fixed condition with no head movement and in a free condition where head movements were allowed. Electroencephalography (EEG), gaze and pupil data were collected. To investigate how neural and physiological reorienting signals are distributed across different gaze events, we used hierarchical discriminant component analysis (HDCA) to identify EEG and pupil-based discriminating components. Mixed-effects general linear models (GLM) were used to determine the correlation between these discriminating components and the different gaze events time. HDCA was also used to combine EEG, pupil and dwell time signals to classify reorienting events. Main results. In both EEG and pupil, dwell time contributes most significantly to the reorienting signals. However, when dwell times were orthogonalized against other gaze events, the distributions of the reorienting signals were different across the two modalities, with EEG reorienting signals leading that of the pupil reorienting signals. We also found that the hybrid classifier that integrates EEG, pupil and dwell time features detects the reorienting signals in both the fixed (AUC = 0.79) and the free (AUC = 0.77) condition. Significance. We show that the neural and ocular reorienting signals are distributed differently across gaze events when a subject is immersed in VR, but nevertheless can be captured and integrated to classify target vs. distractor objects to which the human subject orients.

Reproductive barriers in cassava: Factors and implications for genetic improvement.

Cassava breeding is hampered by high flower abortion rates that prevent efficient recombination among promising clones. To better understand the factors causing flower abortion and propose strategies to overcome them, we 1) analyzed the reproductive barriers to intraspecific crossing, 2) evaluated pollen-pistil interactions to maximize hand pollination efficiency, and 3) identified the population structure of elite parental clones. From 2016 to 2018, the abortion and fertilization rates of 5,748 hand crossings involving 91 parents and 157 progenies were estimated. We used 16,300 single nucleotide polymorphism markers to study the parents' population structure via discriminant analysis of principal components, and three clusters were identified. To test for male and female effects, we used a mixed model in which the environment (month and year) was fixed, while female and male (nested to female) were random effects. Regardless of the population structure, significant parental effects were identified for abortion and fertilization rates, suggesting the existence of reproductive barriers among certain cassava clones. Matching ability between cassava parents was significant for pollen grains that adhered to the stigma surface, germinated pollen grains, and the number of fertilized ovules. Non-additive genetic effects were important to the inheritance of these traits. Pollen viability and pollen-pistil interactions in cross- and self-pollination were also investigated to characterize pollen-stigma compatibility. Various events related to pollen tube growth dynamics indicated fertilization abnormalities. These abnormalities included the reticulated deposition of callose in the pollen tube, pollen tube growth cessation in a specific region of the stylet, and low pollen grain germination rate. Generally, pollen viability and stigma receptivity varied depending on the clone and flowering stage and were lost during flowering. This study provides novel insights into cassava reproduction that can assist in practical crossing and maximize the recombination of contrasting clones.

Evaluation of Various DR Techniques in Massive Patient Datasets using HDFS

The objective of comparing various dimensionality techniques is to reduce feature sets in order to group attributes effectively with less computational processing time and utilization of memory. The various reduction algorithms can decrease the dimensionality of dataset consisting of a huge number of interrelated variables, while retaining the dissimilarity present in the dataset as much as possible. In this paper we use, Standard Deviation, Variance, Principal Component Analysis, Linear Discriminant Analysis, Factor Analysis, Positive Region, Information Entropy and Independent Component Analysis reduction algorithms using Hadoop Distributed File System for massive patient datasets to achieve lossless data reduction and to acquire required knowledge. The experimental results demonstrate that the ICA technique can efficiently operate on massive datasets eliminates irrelevant data without loss of accuracy, reduces storage space for the data and also the computation time compared to other techniques.

Supervised dimensionality reduction technology of generalized discriminant component analysis and its kernelization forms

Supervised subspace projection technology is a major method for dimensionality reduction in pattern recognition. At present, most supervised subspace projection algorithms are derived from the multi-dimensional extended version of Fisher linear discriminant analysis (FDA), also known as Multi-dimensional Fisher discriminant analysis (MD-FDA). However, MD-FDA needs to be improved further because the projection vectors in the noise-subspace cannot be sorted and the ill-condition of the within-class scatter matrix may cause severe numerical instabilities. Generalized discriminant component analysis (GDCA), the generalization of MD-FDA, together with its kernelization forms are proposed and correspondingly rigorous mathematical proofs are detailed in this paper. By virtue of 5 validation data sets derived from UCI Machine Learning Repository and our laboratory, the theoretical validity and technical advantages of GDCA as well as its kernelization forms are verified, and the effectiveness of the newly proposed method is demonstrated in comparison with 36 kinds of state-of-the-art dimensionality reduction algorithms.