Genome dosage alteration caused by chromosome pyramiding and shuffling effects on karyotypic heterogeneity, reproductive diversity, and phenotypic variation in Zea-Tripsacum allopolyploids.

KEY MESSAGE: We developed an array of Zea-Tripsacum tri-hybrid allopolyploids with multiple ploidies. We unveiled that changes in genome dosage due to the chromosomes pyramiding and shuffling of three species effects karyotypic heterogeneity, reproductive diversity, and phenotypic variation in Zea-Tripsacum allopolyploids. Polyploidy, or whole genome duplication, has played a major role in evolution and speciation. The genomic consequences of polyploidy have been extensively studied in many plants; however, the extent of chromosomal variation, genome dosage, phenotypic diversity, and heterosis in allopolyploids derived from multiple species remains largely unknown. To address this question, we synthesized an allohexaploid involving Zea mays, Tripsacum dactyloides, and Z. perennis by chromosomal pyramiding. Subsequently, an allooctoploid and an allopentaploid were obtained by hybridization of the allohexaploid with Z. perennis. Moreover, we constructed three populations with different ploidy by chromosomal shuffling (allopentaploid × Z. perennis, allohexaploid × Z. perennis, and allooctoploid × Z. perennis). We have observed 3 types of sexual reproductive modes and 2 types of asexual reproduction modes in the tri-species hybrids, including 2n gamete fusion (2n + n), haploid gamete fusion (n + n), polyspermy fertilization (n + n + n) or 2n gamete fusion (n + 2n), haploid gametophyte apomixis, and asexual reproduction. The tri-hybrids library presents extremely rich karyotype heterogeneity. Chromosomal compensation appears to exist between maize and Z. perennis. A rise in the ploidy of the trihybrids was linked to a higher frequency of chromosomal translocation. Variation in the degree of phenotypic diversity observed in different segregating populations suggested that genome dosage effects phenotypic manifestation. These findings not only broaden our understanding of the mechanisms of polyploid formation and reproductive diversity but also provide a novel insight into genome pyramiding and shuffling driven genome dosage effects and phenotypic diversity.

Predicting Label Distribution From Tie-Allowed Multi-Label Ranking.

Label distribution offers more information about label polysemy than logical label. There are presently two approaches to obtaining label distributions: LDL (label distribution learning) and LE (label enhancement). In LDL, experts must annotate training instances with label distributions, and a predictive function is trained on this training set to obtain label distributions. In LE, experts must annotate instances with logical labels, and label distributions are recovered from them. However, LDL is limited by expensive annotations, and LE has no performance guarantee. Therefore, we investigate how to predict label distribution from TMLR (tie-allowed multi-label ranking) which is a compromise on annotation cost but has good performance guarantees. On the one hand, we theoretically dissect the relationship between TMLR and label distribution. We define EAE (expected approximation error) to quantify the quality of an annotation, provide EAE bounds for TMLR, and derive the optimal range of label distributions corresponding to a given TMLR annotation. On the other hand, we propose a framework for predicting label distribution from TMLR via conditional Dirichlet mixtures. This framework blends the procedures of recovering and learning label distributions end-to-end and allows us to effortlessly encode our knowledge by a semi-adaptive scoring function. Extensive experiments validate our proposal.

Joint Projection Learning and Tensor Decomposition-Based Incomplete Multiview Clustering.

Incomplete multiview clustering (IMVC) has received increasing attention since it is often that some views of samples are incomplete in reality. Most existing methods learn similarity subgraphs from original incomplete multiview data and seek complete graphs by exploring the incomplete subgraphs of each view for spectral clustering. However, the graphs constructed on the original high-dimensional data may be suboptimal due to feature redundancy and noise. Besides, previous methods generally ignored the graph noise caused by the interclass and intraclass structure variation during the transformation of incomplete graphs and complete graphs. To address these problems, we propose a novel joint projection learning and tensor decomposition (JPLTD)-based method for IMVC. Specifically, to alleviate the influence of redundant features and noise in high-dimensional data, JPLTD introduces an orthogonal projection matrix to project the high-dimensional features into a lower-dimensional space for compact feature learning. Meanwhile, based on the lower-dimensional space, the similarity graphs corresponding to instances of different views are learned, and JPLTD stacks these graphs into a third-order low-rank tensor to explore the high-order correlations across different views. We further consider the graph noise of projected data caused by missing samples and use a tensor-decomposition-based graph filter for robust clustering. JPLTD decomposes the original tensor into an intrinsic tensor and a sparse tensor. The intrinsic tensor models the true data similarities. An effective optimization algorithm is adopted to solve the JPLTD model. Comprehensive experiments on several benchmark datasets demonstrate that JPLTD outperforms the state-of-the-art methods. The code of JPLTD is available at https://github.com/weilvNJU/JPLTD.

Three-Stage Semisupervised Cross-Modal Hashing With Pairwise Relations Exploitation.

Hashing methods have sparked a great revolution in cross-modal retrieval due to the low cost of storage and computation. Benefiting from the sufficient semantic information of labeled data, supervised hashing methods have shown better performance compared with unsupervised ones. Nevertheless, it is expensive and labor intensive to annotate the training samples, which restricts the feasibility of supervised methods in real applications. To deal with this limitation, a novel semisupervised hashing method, i.e., three-stage semisupervised hashing (TS3H) is proposed in this article, where both labeled and unlabeled data are seamlessly handled. Different from other semisupervised approaches that learn the pseudolabels, hash codes, and hash functions simultaneously, the new approach is decomposed into three stages as the name implies, in which all of the stages are conducted individually to make the optimization cost-effective and precise. Specifically, the classifiers of different modalities are learned via the provided supervised information to predict the labels of unlabeled data at first. Then, hash code learning is achieved with a simple but efficient scheme by unifying the provided and the newly predicted labels. To capture the discriminative information and preserve the semantic similarities, we leverage pairwise relations to supervise both classifier learning and hash code learning. Finally, the modality-specific hash functions are obtained by transforming the training samples to the generated hash codes. The new approach is compared with the state-of-the-art shallow and deep cross-modal hashing (DCMH) methods on several widely used benchmark databases, and the experiment results verify its efficiency and superiority.

Multispecies polyploidization, chromosome shuffling, and genome extraction in Zea/Tripsacum hybrids.

By hybridization and special sexual reproduction, we sequentially aggregated Zea mays, Zea perennis, and Tripsacum dactyloides in an allohexaploid, backcrossed it with maize, derived self-fertile allotetraploids of maize and Z. perennis by natural genome extraction, extended their first six selfed generations, and finally constructed amphitetraploid maize using nascent allotetraploids as a genetic bridge. Transgenerational chromosome inheritance, subgenome stability, chromosome pairings and rearrangements, and their impacts on an organism's fitness were investigated by fertility phenotyping and molecular cytogenetic techniques genomic in situ hybridization (GISH) and fluorescence in situ hybridization (FISH). Results showed that diversified sexual reproductive methods produced highly differentiated progenies (2n = 35-84) with varying proportions of subgenomic chromosomes, of which one individual (2n = 54, MMMPT) overcame self-incompatibility barriers and produced a self-fertile nascent near-allotetraploid by preferentially eliminating Tripsacum chromosomes. Nascent near-allotetraploid progenies showed persistent chromosome changes, intergenomic translocations, and rDNA variations for at least up to the first six selfed generations; however, the mean chromosome number preferably maintained at the near-tetraploid level (2n = 40) with full integrity of 45S rDNA pairs, and a trend of decreasing variations by advancing generations with an average of 25.53, 14.14, and 0.37 for maize, Z. perennis, and T. dactyloides chromosomes, respectively. The mechanisms for three genome stabilities and karyotype evolution for formatting new polyploid species were discussed.

Optimization-Based Three-Way Decisions With Interval-Valued Intuitionistic Fuzzy Information.

Due to the effectiveness and advantages of interval-valued intuitionistic fuzzy sets (IVIFSs) in evaluating uncertainty and risk, we introduce IVIFSs into loss functions of decision-theoretic rough sets (DTRSs) and propose an optimization-based approach to interval-valued intuitionistic fuzzy three-way decisions. First, based on the classical DTRSs and two previous optimization models, we construct a new concise linear programming model for simultaneously determining the threshold pair. Our model is mathematically equivalent to the DTRSs and the previous models under the Karush-Kuhn-Tucker (KKT) condition. Second, we extend the constructed model via the IVIFSs of loss functions and we discuss the relations between these loss functions based on a similarity measure function-based ranking method and a multiple score function-based ranking method for IVIFSs. Third, we develop our extended models via two ranking methods and we prove the existence and uniqueness of the optimal solution of the model. The optimization-based method, along with its algorithm for three-way decisions, is designed in an interval-valued intuitionistic fuzzy environment. Compared to the latest existing methods, our method has three advantages (see Advantages 1-3). Finally, an illustrative example is considered, and the advantages of our approach are demonstrated by this example.

Enhanced Group Sparse Regularized Nonconvex Regression for Face Recognition.

Regression analysis based methods have shown strong robustness and achieved great success in face recognition. In these methods, convex l1-norm and nuclear norm are usually utilized to approximate the l0-norm and rank function. However, such convex relaxations may introduce a bias and lead to a suboptimal solution. In this paper, we propose a novel Enhanced Group Sparse regularized Nonconvex Regression (EGSNR) method for robust face recognition. An upper bounded nonconvex function is introduced to replace l1-norm for sparsity, which alleviates the bias problem and adverse effects caused by outliers. To capture the characteristics of complex errors, we propose a mixed model by combining γ-norm and matrix γ-norm induced from the nonconvex function. Furthermore, an l2,γ-norm based regularizer is designed to directly seek the interclass sparsity or group sparsity instead of traditional l2,1-norm. The locality of data, i.e., the distance between the query sample and multi-subspaces, is also taken into consideration. This enhanced group sparse regularizer enables EGSNR to learn more discriminative representation coefficients. Comprehensive experiments on several popular face datasets demonstrate that the proposed EGSNR outperforms the state-of-the-art regression based methods for robust face recognition.

Locality-Constrained Discriminative Matrix Regression for Robust Face Identification.

Regression-based methods have been widely applied in face identification, which attempts to approximately represent a query sample as a linear combination of all training samples. Recently, a matrix regression model based on nuclear norm has been proposed and shown strong robustness to structural noises. However, it may ignore two important issues: the label information and local relationship of data. In this article, a novel robust representation method called locality-constrained discriminative matrix regression (LDMR) is proposed, which takes label information and locality structure into account. Instead of focusing on the representation coefficients, LDMR directly imposes constraints on representation components by fully considering the label information, which has a closer connection to identification process. The locality structure characterized by subspace distances is used to learn class weights, and the correct class is forced to make more contribution to representation. Furthermore, the class weights are also incorporated into a competitive constraint on the representation components, which reduces the pairwise correlations between different classes and enhances the competitive relationships among all classes. An iterative optimization algorithm is presented to solve LDMR. Experiments on several benchmark data sets demonstrate that LDMR outperforms some state-of-the-art regression-based methods.

Low-Rank Tensor Regularized Views Recovery for Incomplete Multiview Clustering.

In real applications, it is often that the collected multiview data contain missing views. Most existing incomplete multiview clustering (IMVC) methods cannot fully utilize the underlying information of missing data or sufficiently explore the consistent and complementary characteristics. In this article, we propose a novel Low-rAnk Tensor regularized viEws Recovery (LATER) method for IMVC, which jointly reconstructs and utilizes the missing views and learns multilevel graphs for comprehensive similarity discovery in a unified model. The missing views are recovered from a common latent representation, and the recovered views conversely improve the learning of shared patterns. Based on the shared subspace representations and recovered complete multiview data, the multilevel graphs are learned by self-representation to fully exploit the consistent and complementary information among views. Besides, a tensor nuclear norm regularizer is introduced to pursue the global low-rank property and explore the interview correlations. An alternating direction minimization algorithm is presented to optimize the proposed model. Moreover, a new initialization method is proposed to promote the effectiveness of our method for latent representation learning and missing data recovery. Extensive experiments demonstrate that our method outperforms the state-of-the-art approaches.

Tripsazea, a Novel Trihybrid of Zea mays, Tripsacum dactyloides, and Zeaperennis.

A trispecific hybrid, MTP (hereafter called tripsazea), was developed from intergeneric crosses involving tetraploid Zea mays (2n = 4x = 40, genome: MMMM), tetraploid Tripsacum dactyloides (2n = 4x = 72, TTTT), and tetraploid Zperennis (2n = 4x = 40, PPPP). On crossing maize-Tripsacum (2n = 4x = 56, MMTT) with Zperennis, 37 progenies with varying chromosome numbers (36-74) were obtained, and a special one (i.e., tripsazea) possessing 2n = 74 chromosomes was generated. Tripsazea is perennial and expresses phenotypic characteristics affected by its progenitor parent. Flow cytometry analysis of tripsazea and its parents showed that tripsazea underwent DNA sequence elimination during allohexaploidization. Of all the chromosomes in diakinesis I, 18.42% participated in heterogenetic pairing, including 16.43% between the M- and P-genomes, 1.59% between the M- and T-genomes, and 0.39% in T- and P-genome pairing. Tripsazea is male sterile and partly female fertile. In comparison with previously synthesized trihybrids containing maize, Tripsacum and teosinte, tripsazea has a higher chromosome number, higher seed setting rate, and vegetative propagation ability of stand and stem. However, few trihybrids possess these valuable traits at the same time. The potential of tripsazea is discussed with respect to the deployment of the genetic bridge for maize improvement and forage breeding.

Allopolyploidization facilitates gene flow and speciation among corn, Zea perennis and Tripsacum dactyloides.

MAIN CONCLUSION: Tripsacum dactyloides is closely related to Zea mays since Zea perennis and the MTP tri- species hybrid have four possible reproductive modes. Eastern gamagrass (Tripsacum dactyloides L.) and tetraploid perennial teosinte (Zea perennis) are well known to possess genes conferring resistance against biotic and abiotic stresses as well as adaptation to flood and aluminum toxic soils. However, plant breeders have been hampered to utilize these and other beneficial traits for maize improvement due to sterility in their hybrids. By crossing a tetraploid maize-inbred line × T. dactyloides, a female fertile hybrid was produced that was crossed with Z. perennis to yield a tri-genomic female fertile hybrid, which was backcrossed with diploid maize to produce BC1 and BC2. The tri-genomic hybrid provided a new way to transfer genetic material from both species into maize by utilizing conventional plant breeding methods. On the basis of cytogenetic observations using multi-color genomic in situ hybridization, the progenies were classified into four groups, in which chromosomes could be scaled both up and down with ease to produce material for varying breeding and genetic purposes via apomixis or sexual reproduction. In the present study, pathways were found to recover maize and to obtain specific translocations as well as a speedy recovery of the T. dactyloides-maize addition line in a second backcross generation. However, phenotypes of the recovered maize were in most cases far from maize as a result of genetic load from T. dactyloides and Z. perennis, and could not be directly used as a maize-inbred line but could serve as an intermediate material for maize improvement. A series of hybrids was produced (having varying chromosome number, constitution, and translocations) with agronomic traits from all three parental species. The present study provides an application of overcoming the initial interspecific barriers among these species. Moreover, T. dactyloides is closely related to Z. mays L. ssp. mays since Z. perennis and the MTP tri- species hybrid have four possible reproductive modes.

Self-Paced Prioritized Curriculum Learning With Coverage Penalty in Deep Reinforcement Learning.

In this paper, a new training paradigm is proposed for deep reinforcement learning using self-paced prioritized curriculum learning with coverage penalty. The proposed deep curriculum reinforcement learning (DCRL) takes the most advantage of experience replay by adaptively selecting appropriate transitions from replay memory based on the complexity of each transition. The criteria of complexity in DCRL consist of self-paced priority as well as coverage penalty. The self-paced priority reflects the relationship between the temporal-difference error and the difficulty of the current curriculum for sample efficiency. The coverage penalty is taken into account for sample diversity. With comparison to deep Q network (DQN) and prioritized experience replay (PER) methods, the DCRL algorithm is evaluated on Atari 2600 games, and the experimental results show that DCRL outperforms DQN and PER on most of these games. More results further show that the proposed curriculum training paradigm of DCRL is also applicable and effective for other memory-based deep reinforcement learning approaches, such as double DQN and dueling network. All the experimental results demonstrate that DCRL can achieve improved training efficiency and robustness for deep reinforcement learning.

Effects of indole-3-butytric acid on lead and zinc accumulations in Pseudostellaria maximowicziana.

Plant hormones can improve the phytoremediation capabilities of heavy metal hyperaccumulator plants. In this study, different doses of indole-3-butytric acid (IBA) were sprayed on the leaves of the lead (Pb) and zinc (Zn) accumulator plant Pseudostellaria maximowicziana, which was planted in Pb-Zn contaminated soil, and the effects of IBA on Pb and Zn accumulation levels in P. maximowicziana were studied. Spraying 25- and 50-mg/L IBA doses increased the stem, leaf and shoot biomasses of P. maximowicziana compared with the control, while 75- and 100-mg/L IBA doses decreased them. The 50-mg/L IBA dose increased the P. maximowicziana contents of chlorophyll a, total chlorophyll and carotenoid of compared with the control, and other doses had no significant effects or decreased these values. Spraying IBA reduced the superoxide dismutase activity of P. maximowicziana compared with the control, but improved the peroxidase and catalase activities. The 50-, 75-, and 100-mg/L IBA doses increased the Pb and Zn contents in P. maximowicziana compared with the control and also increased the amounts of Pb and Zn extracted by P. maximowicziana. Thus, 50 mg/L of IBA could promote the growth and the Pb and Zn phytoremediation capabilities of P. maximowicziana.