Tree Transformation Research Articles

A holistic matrix norm-based alternative solution method for Markov reward games

In this study, we focus on examining single-agent stochastic games, especially Markov reward games represented in the form of a decision tree. We propose an alternative solution method based on the matrix norms for these games. In contrast to the existing methods such as value iteration, policy iteration, and dynamic programming, which are state-and-action-based approaches, the proposed matrix norm-based method considers the relevant stages and their actions as a whole and solves it holistically for each stage without computing the effects of each action on each state's reward individually. The new method involves a distinct transformation of the decision tree into a payoff matrix for each stage and the utilization of the matrix norm of the obtained payoff matrix. Additionally, the concept of the moving matrix is integrated into the proposed method to incorporate the impacts of all actions on the stage simultaneously, rendering the method holistic. Moreover, we present an explanatory algorithm for the implementation of the method and also provide a comprehensive solution diagram explaining the method figuratively. As a result, we offer a new and alternative perspective for solving the games with the help of the proposed method due to the simplicity of utilization of the matrix norms in addition to the existing methods. For clarification of the matrix norm-based method, we demonstrate the figurative application of the method on a benchmark Markov reward game with 2-stages and 2-actions and a comprehensive implementation of the method on a game consisting of 3-stages and 3-actions.

Tolerance of Triploid Hybrids of White Poplar ‘Beilinxiongzhu 1’ to Genetic Transformation Screening Agents In Vitro

Genetic transformation of forest trees is essential for validating gene functions and breeding new varieties through molecular means. Appropriate selective pressure is critical for creating an effective screening system. ‘Beilinxiongzhu 1’ sensitivity testing showed that the critical tolerance concentrations for hygromycin (Hyg), kanamycin (Kan), and glyphosate (PPT) in leaf explants were 2.0 mg/L, 20 mg/L, and 1.0 mg/L, respectively. Among the physiological indicators, soluble sugar content, soluble protein content, and endogenous hormone levels were identified as key markers of the effects of the different antibiotic treatments. Transcriptome analysis showed that Hyg treatment resulted in a large number of differentially expressed genes (DEGs) involved in leaf cell wall synthesis and glucose metabolism. Under Kan treatment, the DEGs were associated with pathways such as ribosome biosynthesis and histone packaging in eukaryotes. Under PPT treatment, significant DEGs were related to ABC transporters. DEGs common to all three antibiotics were involved in glutathione metabolism pathways. A weighted gene co-expression network analysis identified TRXH2, H3.2, H2B, GST, U71K1, and CHS as key genes in response to antibiotic stress. By elucidating the physiological and molecular mechanisms by which different antibiotics affect leaf sprouting, our study serves as a reference for research into the genetic transformation of poplar leaves.

A fast strategy-solving method for adversarial team games utilizing warm starting

Adversarial team games (ATGs) have garnered significant attention in recent years, leading to the emergence of various solutions such as linear programming algorithms, multi-agent reinforcement learning, and game tree transformation. The ATGs involve large-scale game trees, resulting in higher time costs for computation. In this paper, we focus on expediting the solution of team-maxmin equilibrium with correlation (TMECor), which can be considered the equilibrium that maximizes the team’s payoff. To address this, we propose a transformation with seed strategies (TSS). TSS leverages reinforcement learning to calculate player strategies. We initialize the strategies of all players, referred to as seed strategies, and incorporate them into the multi-agent game tree during the transformation process. These seed strategies serve as the starting strategies for counterfactual regret minimization (CFR). CFR initializes the strategies and cumulative regret of all players based on the seed strategy. By warm starting the whole process, our method accelerates the solving of TMECor. We conducted nine experiments using Kuhn poker and Leduc Hold’em poker. The results demonstrated that TSS improved the solving speed of TMECor.

Effects of Aviation Mutagenesis on Soil Chemical Indexes, Enzyme Activities, and Metabolites of Dahongpao (Camellia sinensis) Tea Trees.

Aviation mutagenesis is a breeding method for the rapid selection of superior plant varieties. In this study, rhizosphere soil chemical indexes, soil enzyme activities, and soil metabolites were measured in Dahongpao tea trees with aviation mutagenesis (TM) and without aviation mutagenesis (CK). The main soil metabolites distinguishing TM and CK and their relationships with soil chemical indexes and soil enzyme activities were analyzed and obtained. The results showed that there was no significant change in the rhizosphere soils' pH of TM tea trees compared to CK (p = 0.91), while all other chemical indexes of TM were significantly higher than CK (p < 0.05). In addition, the activities of enzymes related to soil nutrient cycling such as urease, protease, sucrase, acid phosphatase and cellulase, and enzymes related to soil antioxidants such as superoxide dismutase, catalase, peroxidase, and polyphenol oxidase were significantly increased (p < 0.05) in the rhizosphere soils of TM tea trees compared to CK. Soil metabolite analysis showed that the main soil metabolites distinguishing CK from TM were carbohydrates, nitrogen compounds, and amines. Of these, carbohydrates and nitrogen compounds were significantly positively correlated with soil chemical indexes and soil enzymes, whereas amine was significantly negatively correlated with soil chemical indexes such as organic matter, total nitrogen, total potassium, available nitrogen, available phosphorus; amine showed significant negative correlation with soil enzymes such as catalase, peroxidase, polyphenol oxidase, and urease. It can be seen that aviation mutagenesis is conducive to improving the ability of tea tree rhizosphere aggregation and transformation of soil nutrients, increasing the total amount of soil nutrients and the content of available nutrients, which is more conducive to promoting the uptake of nutrients by the tea tree, and thus promoting the growth of the tea tree.

Growth and irreducibility in path-incompressible trees

We study randomness-preserving transformations of path-incompressible trees, namely trees of finite randomness deficiency. We characterize their branching density, and show: (a) sparse perfect path-incompressible trees can be effectively densified, almost surely; (b) there exists a path-incompressible tree with infinitely many paths which does not compute any perfect path-incompressible tree with computable oracle-use.

Wavelet Tree Transformer: Multihead Attention With Frequency-Selective Representation and Interaction for Remote Sensing Object Detection

Wavelet Tree Transformer: Multihead Attention With Frequency-Selective Representation and Interaction for Remote Sensing Object Detection

Improving Code Summarization with Tree Transformer Enhanced by Position-related Syntax Complement

Improving Code Summarization with Tree Transformer Enhanced by Position-related Syntax Complement

G‐MATT: Single‐step retrosynthesis prediction using molecular grammar tree transformer

AbstractVarious template‐based and template‐free approaches have been proposed for single‐step retrosynthesis prediction in recent years. While these approaches demonstrate strong performance from a data‐driven metrics standpoint, many model architectures do not incorporate underlying chemistry principles. Here, we propose a novel chemistry‐aware retrosynthesis prediction framework that combines powerful data‐driven models with prior domain knowledge. We present a tree‐to‐sequence transformer architecture that utilizes hierarchical SMILES grammar‐based trees, incorporating crucial chemistry information that is often overlooked by SMILES text‐based representations, such as local structures and functional groups. The proposed framework, grammar‐based molecular attention tree transformer (G‐MATT), achieves significant performance improvements compared to baseline retrosynthesis models. G‐MATT achieves a promising top‐1 accuracy of 51% (top‐10 accuracy of 79.1%), an invalid rate of 1.5%, and a bioactive similarity rate of 74.8% on the USPTO‐50K dataset. Additional analyses of G‐MATT attention maps demonstrate the ability to retain chemistry knowledge without relying on excessively complex model architectures.

Current Achievements and Future Challenges of Genotype-Dependent Somatic Embryogenesis Techniques in Hevea brasiliensis

The rubber tree (Hevea brasiliensis) is the most important commercial plant for producing natural rubber. Immature seed inner integument and anther-derived somatic embryogenesis techniques play a crucial role in the in vitro large-scale propagation and genetic transformation of the rubber tree. However, somatic embryogenesis is highly genotype-dependent, that is, only a limited number of H. brasiliensis genotypes, such as CATAS73397, CATAS917, and PB260, can be efficiently induced by somatic embryogenesis and used for large-scale propagation or transformation. The genotype dependence of the somatic embryogenesis technique is a conundrum for the application of Hevea biotechnology in most commercially important cultivars, such as Reken628 and CATAS879. Previous studies have shown that several somatic embryogenesis regulators can overcome genotype dependence and enhance the transformation and regeneration efficiency of recalcitrant plants and cultivars. In this review, we first describe the relevant successful applications of somatic embryogenesis technology in seedling production and genetic modification of H. brasiliensis. Second, we discuss the genotype dependence of somatic embryogenesis as the major challenge currently. Third, we summarize the recent significant advances in the understanding of the molecular mechanisms underlying somatic embryogenesis in other plants. Finally, we suggest a roadmap for using somatic embryogenesis regulatory genes to facilitate genotype-dependent somatic embryogenesis technology in H. brasiliensis.

Cry1Ba1-mediated toxicity of transgenic Bergera koenigii and Citrus sinensis to the Asian citrus psyllid Diaphorina citri.

The Asian citrus psyllid, Diaphorina citri, vectors the bacterial causative agent of citrus greening disease, which has severely impacted citrus production on a global scale. As the current repeated application of chemical insecticides is unsustainable for management of this insect and subsequent protection of groves, we investigated the potential use of the bacteria-derived pesticidal protein, Cry1Ba1, when delivered via transgenic citrus plants. Having demonstrated transformation of the Indian curry leaf tree, Bergera koenigii, for Cry1Ba1 expression for use as a trap plant, we produced transgenic plants of Duncan grapefruit, Citrus paridisi, Valencia sweet orange, Citrus sinensis, and Carrizo citrange, C. sinensis x Poncirus trifoliata, for expression of Cry1Ba1. The presence of the cry1ba1 gene, and cry1ba1 transcription were confirmed. Western blot detection of Cry1Ba1 was confirmed in most cases. When compared to those from wild-type plants, leaf discs from transgenic Duncan and Valencia expressing Cry1Ba1 exhibited a "delayed senescence" phenotype, similar to observations made for transgenic B. koenigii. In bioassays, significant reductions in the survival of adult psyllids were noted on transgenic B. koenigii and Valencia sweet orange plants expressing Cry1Ba1, but not on transgenic Duncan grapefruit or Carrizo citrange. In contrast to psyllids fed on wild type plants, the gut epithelium of psyllids fed on transgenic plants was damaged, consistent with the mode of action of Cry1Ba1. These results indicate that the transgenic expression of a bacterial pesticidal protein in B. koenigii and Valencia sweet orange offers a viable option for management of D. citri, that may contribute to solutions that counter citrus greening disease.

Multilevel hybrid accurate handcrafted model for myocardial infarction classification using ECG signals.

Myocardial infarction (MI) is detected using electrocardiography (ECG) signals. Machine learning (ML) models have been used for automated MI detection on ECG signals. Deep learning models generally yield high classification performance but are computationally intensive. We have developed a novel multilevel hybrid feature extraction-based classification model with low time complexity for MI classification. The study dataset comprising 12-lead ECGs belonging to one healthy and 10 MI classes were downloaded from a public ECG signal databank. The model architecture comprised multilevel hybrid feature extraction, iterative feature selection, classification, and iterative majority voting (IMV). In the hybrid handcrafted feature (HHF) generation phase, both textural and statistical feature extraction functions were used to extract features from ECG beats but only at a low level. A new pooling-based multilevel decomposition model was presented to enable them to create features at a high level. This model used average and maximum pooling to create decomposed signals. Using these pooling functions, an unbalanced tree was obtained. Therefore, this model was named multilevel unbalanced pooling tree transformation (MUPTT). On the feature extraction side, two extractors (functions) were used to generate both statistical and textural features. To generate statistical features, 20 commonly used moments were used. A new, improved symmetric binary pattern function was proposed to generate textural features. Both feature extractors were applied to the original MI signal and the decomposed signals generated by the MUPTT. The most valuable features from among the extracted feature vectors were selected using iterative neighborhood component analysis (INCA). In the classification phase, a one-dimensional nearest neighbor classifier with ten-fold cross-validation was used to obtain lead-wise results. The computed lead-wise results derived from all 12 leads of the same beat were input to the IMV algorithm to generate ten voted results. The most representative was chosen using a greedy technique to calculate the overall classification performance of the model. The HHF-MUPTT-based ECG beat classification model attained excellent performance, with the best lead-wise accuracy of 99.85% observed in Lead III and 99.94% classification accuracy using the IMV algorithm. The results confirmed the high MI classification ability of the presented computationally lightweight HHF-MUPTT-based model.

Soil metagenomic analysis on changes of functional genes and microorganisms involved in nitrogen-cycle processes of acidified tea soils.

Nitrogen (N) is the first essential nutrient for tea growth. However, the effect of soil acidification on soil N cycle and N forms in tea plantation are unclear. In this study, the nitrogen contents, soil enzyme activity and N mineralization rate in acidified soil of tea plantation were measured. Moreover, the effects of soil acidification on N cycling functional genes and functional microorganisms were explored by soil metagenomics. The results showed that the NH4 +-N, available N and net N mineralization rate in the acidified tea soil decreased significantly, while the NO3 --N content increased significantly. The activities of sucrase, protease, catalase and polyphenol oxidase in the acidified tea soil decreased significantly. The abundance of genes related to ammonification, dissimilatory N reduction, nitrification and denitrification pathway in the acidified tea soil increased significantly, but the abundance of functional genes related to glutamate synthesis and assimilatory N reduction pathway were opposite. In addition, the abundance of Proteobacteria, Actinobacteria, Chloroflexi, Nitrospirae, Actinomadura, Nitrospira etc. microorganisms related to nitrification, denitrification and pathogenic effect increased significantly in the acidified tea soil. The correlation results showed that soil pH and N forms were correlated with soil enzyme activity, N cycling function genes and microbial changes. In conclusion, soil acidification results in significant changes in enzyme activity, gene abundance and microorganism involved in various N cycle processes in acidified tea soil, which leads to imbalance of soil N form ratio and is not conducive to N transformation and absorption of tea trees.

Dependency multi-weight-view graphs for event detection with label co-occurrence

Event detection is an extremely important step for extracting events, and its goal is to detect the occurrences of events and categorize them. Prior works use either a fine-grained approach or a coarse-grained approach to detect events. However, both approaches encounter two problems: (1) Enriched feature information obtained from external tools probably introduces unnecessary errors. (2) Ignoring the phenomenon of event co-occurrence in a sentence degrades the performance of event detection. To address these issues, a novel framework called Dependency Multi-Weight-View Graph (DMWVG) is proposed to detect events without trigger tagging and external feature engineering. Specifically, DMWVG leverages the tree transformer to automatically generate syntactic relations with different weights to mitigate the errors generated by external syntactic parsing tools. In addition, an event correlation matrix is constructed with the help of the event label co-occurrence to deliver essential information on triggers and to promote the performance of event detection. Finally, the experimental results indicate that our approach is competitive with existing SOTA methods on four datasets (75.77% F1-score on ACE 2005, 82.11% F1-score on MLEE, 75.61% F1-score on BioNLP 09, and 72.20% F1-score on BioNLP 11).

Growth and Irreducibility in Path-Incompressible Trees

We study effective randomness-preserving transformations of path-incompressible trees. There exists a path-incompressible tree with infinitely many paths, which does not compute any perfect pathwise-random tree. Sparse perfect path-incompressible trees can be effectively densified, almost surely. We characterize the density of branching that pathwise-random trees can have.

Natural Language Processing and Computational Linguistics

Natural Language Processing and Computational Linguistics

Advances and Perspectives of Transgenic Technology and Biotechnological Application in Forest Trees.

Transgenic technology is increasingly used in forest-tree breeding to overcome the disadvantages of traditional breeding methods, such as a long breeding cycle, complex cultivation environment, and complicated procedures. By introducing exogenous DNA, genes tightly related or contributed to ideal traits—including insect, disease, and herbicide resistance—were transferred into diverse forest trees, and genetically modified (GM) trees including poplars were cultivated. It is beneficial to develop new varieties of GM trees of high quality and promote the genetic improvement of forests. However, the low transformation efficiency has hampered the cultivation of GM trees and the identification of the molecular genetic mechanism in forest trees compared to annual herbaceous plants such as Oryza sativa. In this study, we reviewed advances in transgenic technology of forest trees, including the principles, advantages and disadvantages of diverse genetic transformation methods, and their application for trait improvement. The review provides insight into the establishment and improvement of genetic transformation systems for forest tree species. Challenges and perspectives pertaining to the genetic transformation of forest trees are also discussed.

Use of a novel R2R3-MYB transcriptional activator of anthocyanin biosynthesis as visual selection marker for rubber tree (Hevea brasiliensis) transformation

Hevea brasiliensis is presently the only commercial source of natural rubber. The use of anthocyanin production as a visual and non-destructive selection marker may be a viable strategy for increasing the transformation efficiency of this species. In this study, high performance liquid chromatography (HPLC) analysis confirmed that the cyanidin-type anthocyanin is a major determinant of red color in Hevea tissues. It was also found that the anthocyanin production in Hevea somatic embryos could be activated by methyl jasmonate (MeJA) treatment. By sequence alignment and phylogenetic analysis, a novel Hevea R2R3-MYB transcription factor (TF) was isolated and designated here as Hevea brasiliensis Anthocyanin 2 (HbAN2). The expressions of HbAN2 correlate well with the anthocyanin levels in Hevea tissues. The regulatory roles of HbAN2 in anthocyanin biosynthesis were functionally characterized in the native and heterologous systems. In tobacco suspension culture cells, the expression of HbAN2 alone or in combination with two HbbHLH genes activated the promoter of NtDFR, a key gene responsible for anthocyanin biosynthesis. In tobacco plants and Hevea embryos, overexpression of HbAN2 upregulated the late stage anthocyanin pathway genes, resulting in increased anthocyanin accumulation in the transformed parts of the explants, which were therefore easily distinguishable from the non transformed parts. Although transgenic Hevea shoots failed to regenerate into whole plants, these results suggest that HbAN2 is a regulator of anthocyanin biosynthesis and has the potential to be used as visual selection marker for Hevea transformation.

Coffea arabica L. Resistant to Coffee Berry Borer (Hypothenemus hampei) Mediated by Expression of the Bacillus thuringiensis Cry10Aa Protein.

Coffea spp. are tropical plants used for brewing beverages from roasted and grounded seeds, the favorite drink in the world. It is the most important commercial crop plant and the second most valuable international commodity after oil. Global coffee trade relies on two Coffea species: C. arabica L. (arabica coffee) comprising 60% and C. canephora (robusta) comprising the remaining 40%. Arabica coffee has lower productivity and better market price than robusta. Arabica coffee is threatened by disease (i.e., coffee leaf rust), pests [i.e., Hypothenemus hampei or coffee berry borer (CBB) and nematodes], and susceptibility to climate change (i.e., drought and aluminum toxicity). Plant biotechnology by means of tissue culture inducing somatic embryogenesis (SE) process, genetic transformation, and genome editing are tools that can help to solve, at least partially, these problems. This work is the continuation of a protocol developed for stable genetic transformation and successful plant regeneration of arabica coffee trees expressing the Bacillus thuringiensis (Bt) toxin Cry10Aa to induce CBB resistance. A highly SE line with a high rate of cell division and conversion to plants with 8-month plant regeneration period was produced. To validate this capability, gene expression analysis of master regulators of SE, such as BABY BOOM (BBM), FUS3, and LEC1, embryo development, such as EMB2757, and cell cycle progression, such as ETG1 and MCM4, were analyzed during induction and propagation of non-competent and highly competent embryogenic lines. The particle bombardment technique was used to generate stable transgenic lines after 3 months under selection using hygromycin as selectable marker, and 1 month in plant regeneration. Transgenic trees developed fruits after 2 years and demonstrated expression of the Bt toxin ranging from 3.25 to 13.88 μg/g fresh tissue. Bioassays with transgenic fruits on CBB first instar larvae and adults induced mortalities between 85 and 100% after 10 days. In addition, transgenic fruits showed a seed damage lower than 9% compared to 100% of control fruits and adult mortality. This is the first report on stable transformation and expression of the Cry10Aa protein in coffee plants with the potential to control CBB.

On the average Steiner 3-eccentricity of trees

The Steinerk-eccentricity of a vertex v of a graph G is the maximum Steiner distance over all k-subsets of V(G) which contain v. In this paper Steiner 3-eccentricity is studied on trees. Some general properties of the Steiner 3-eccentricity of trees are given. A tree transformation which does not increase the average Steiner 3-eccentricity is given. As its application, several lower and upper bounds for the average Steiner 3-eccentricity of trees are derived.

Encoding Dependency Information inside Tree Transformer

Representing a sentence in a high dimensional space is fundamental for most natural language processing (NLP) tasks at present. These representations depend on the underlying structures upon which they are built. Two scenarios are possible: one is to view the sentence as a sequence of words and another is to consider its inherent grammatical structure. It is possible to equip the first way with some external grammatical knowledge, but to capture a proper syntax would be close to impossible. Therefore, we investigate the second one by extending the design of an existing dependency tree transformer (DT-Transformer). We propose adding a novel edge encoding mechanism to this prior architecture. Experiments show that in sentence encoding, having access to information about the relationships between “head” words and their “dependent” words and how the heads are influenced by the dependent words achieves better sentence representation. Evaluation on the four tasks shows noteworthy results compared to the existing DT-Transformer, standard Transformer, LSTM-based models, and tree- structured LSTMs. Extensive experimentation with representing the edge embeddings as different distributions (mean and standard deviation), encoding the edges in different ways, and an ablation study to find where to place each module in the architecture and which modules to use in the design is also provided.