InfoGCN++: Learning Representation by Predicting the Future for Online Skeleton-based Action Recognition.

Skeleton-based action recognition has made significant advancements recently, with models like InfoGCN showcasing remarkable accuracy. However, these models exhibit a key limitation: they necessitate complete action observation prior to classification, which constrains their applicability in realtime situations such as surveillance and robotic systems. To overcome this barrier, we introduce InfoGCN++, an innovative extension of InfoGCN, explicitly developed for online skeletonbased action recognition. InfoGCN++ augments the abilities of the original InfoGCN model by allowing real-time categorization of action types, independent of the observation sequence's length. It transcends conventional approaches by learning from current and anticipated future movements, thereby creating a more thorough representation of the entire sequence. Our approach to prediction is managed as an extrapolation issue, grounded on observed actions. To enable this, InfoGCN++ incorporates Neural Ordinary Differential Equations, a concept that lets it effectively model the continuous evolution of hidden states. Following rigorous evaluations on three skeleton-based action recognition benchmarks, InfoGCN++ demonstrates exceptional performance in online action recognition. It consistently equals or exceeds existing techniques, highlighting its significant potential to reshape the landscape of real-time action recognition applications. Consequently, this work represents a major leap forward from InfoGCN, pushing the limits of what's possible in online, skeleton-based action recognition.

Bt-Modified Transgenic Rice May Shift the Composition and Diversity of Rhizosphere Microbiota.

Plants significantly shape root-associated microbiota, making rhizosphere microbes useful environmental indicator organisms for safety assessment. Here, we report the pyrosequencing of the bacterial 16S ribosomal RNA in rhizosphere soil samples collected from transgenic cry1Ab/cry1Ac Bt rice Huahui No. 1 (GM crop) and its parental counterpart, Minghui63. We identified a total of 2579 quantifiable bacterial operational taxonomic units (OTUs). Many treatment-enriched microbial OTUs were identified, including 14 NonGM-enriched OTUs and 10 GM-enriched OTUs. OTUs belonging to the phyla Proteobacteria, Actinobacteria, Acidobacteria, Firmicutes, Nitrospirae, Chlorobi and GN04 were identified as statistically different in abundance between GM and the other two treatments. Compared with the different impacts of different rice varieties on microbiota, the impact of rice planting on microbiota is more obvious. Furthermore, Huahui No. 1 transgenic Bt rice had a greater impact on the rhizosphere bacterial communities than Minghui63. Early developmental stages of the transgenic Bt rice had a significant impact on many Bacillaceae communities. Soil chemical properties were not significantly altered by the presence of transgenic Bt rice. The peak concentration level of Bt protein products was detected during the seedling stage of transgenic Bt rice, which may be an intriguing factor for bacterial diversity variations. Based on these findings, we conclude that transgenic Bt rice has a significant impact on root-associated bacteria. This information may be leveraged in future environmental safety assessments of transgenic Bt rice varieties.

Reducing Seed Shattering in Weedy Rice by Editing SH4 and qSH1 Genes: Implications in Environmental Biosafety and Weed Control through Transgene Mitigation.

Mitigating the function of acquired transgenes in crop wild/weedy relatives can provide an ideal strategy to reduce the possible undesired environmental impacts of pollen-mediated transgene flow from genetically engineered (GE) crops. To explore a transgene mitigation system in rice, we edited the seed-shattering genes, SH4 and qSH1, using a weedy rice line ("C9") that originally had strong seed shattering. We also analyzed seed size-related traits, the total genomic transcriptomic data, and RT-qPCR expression of the SH4 or qSH1 gene-edited and SH4/qSH1 gene-edited weedy rice lines. Substantially reduced seed shattering was observed in all gene-edited weedy rice lines. The single gene-edited weedy rice lines, either the SH4 or qSH1 gene, did not show a consistent reduction in their seed size-related traits. In addition, reduced seed shattering was closely linked with the weakness and absence of abscission layers and reduced abscisic acid (ABA). Additionally, the genes closely associated with ABA biosynthesis and signaling transduction, as well as cell-wall hydrolysis, were downregulated in all gene-edited weedy rice lines. These findings facilitate our deep insights into the underlying mechanisms of reduced seed shattering in plants in the rice genus Oryza. In addition, such a mitigating technology also has practical applications for reducing the potential adverse environmental impacts caused by transgene flow and for managing the infestation of weedy rice by acquiring the mitigator from GE rice cultivars through natural gene flow.

PVNet: Pixel-Wise Voting Network for 6DoF Object Pose Estimation.

This paper addresses the problem of instance-level 6DoF object pose estimation from a single RGB image. Many recent works have shown that a two-stage approach, which first detects keypoints and then solves a Perspective-n-Point (PnP) problem for pose estimation, achieves remarkable performance. However, most of these methods only localize a set of sparse keypoints by regressing their image coordinates or heatmaps, which are sensitive to occlusion and truncation. Instead, we introduce a Pixel-wise Voting Network (PVNet) to regress pixel-wise vectors pointing to the keypoints and use these vectors to vote for keypoint locations. This creates a flexible representation for localizing occluded or truncated keypoints. Another important feature of this representation is that it provides uncertainties of keypoint locations that can be further leveraged by the PnP solver. Experiments show that the proposed approach outperforms the state of the art on the LINEMOD, Occluded LINEMOD, YCB-Video, and Tless datasets, while being efficient for real-time pose estimation. We further create a Truncated LINEMOD dataset to validate the robustness of our approach against truncation. The code is available at https://github.com/zju3dv/pvnet.

Fast and Robust Multi-Person 3D Pose Estimation and Tracking From Multiple Views.

This paper addresses the problem of reconstructing 3D poses of multiple people from a few calibrated camera views. The main challenge of this problem is to find the cross-view correspondences among noisy and incomplete 2D pose predictions. Most previous methods address this challenge by directly reasoning in 3D using a pictorial structure model, which is inefficient due to the huge state space. We propose a fast and robust approach to solve this problem. Our key idea is to use a multi-way matching algorithm to cluster the detected 2D poses in all views. Each resulting cluster encodes 2D poses of the same person across different views and consistent correspondences across the keypoints, from which the 3D pose of each person can be effectively inferred. The proposed convex optimization based multi-way matching algorithm is efficient and robust against missing and false detections, without knowing the number of people in the scene. Moreover, we propose to combine geometric and appearance cues for cross-view matching. Finally, an efficient tracking method is proposed to track the detected 3D poses across the multi-view video. The proposed approach achieves the state-of-the-art performance on the Campus and Shelf datasets, while being efficient for real-time applications.

Performance improvement for a 2D convolutional neural network by using SSC encoding on protein-protein interaction tasks.

BACKGROUND: The interactions of proteins are determined by their sequences and affect the regulation of the cell cycle, signal transduction and metabolism, which is of extraordinary significance to modern proteomics research. Despite advances in experimental technology, it is still expensive, laborious, and time-consuming to determine protein-protein interactions (PPIs), and there is a strong demand for effective bioinformatics approaches to identify potential PPIs. Considering the large amount of PPI data, a high-performance processor can be utilized to enhance the capability of the deep learning method and directly predict protein sequences. RESULTS: We propose the Sequence-Statistics-Content protein sequence encoding format (SSC) based on information extraction from the original sequence for further performance improvement of the convolutional neural network. The original protein sequences are encoded in the three-channel format by introducing statistical information (the second channel) and bigram encoding information (the third channel), which can increase the unique sequence features to enhance the performance of the deep learning model. On predicting protein-protein interaction tasks, the results using the 2D convolutional neural network (2D CNN) with the SSC encoding method are better than those of the 1D CNN with one hot encoding. The independent validation of new interactions from the HIPPIE database (version 2.1 published on July 18, 2017) and the validation of directly predicted results by applying a molecular docking tool indicate the effectiveness of the proposed protein encoding improvement in the CNN model. CONCLUSION: The proposed protein sequence encoding method is efficient at improving the capability of the CNN model on protein sequence-related tasks and may also be effective at enhancing the capability of other machine learning or deep learning methods. Prediction accuracy and molecular docking validation showed considerable improvement compared to the existing hot encoding method, indicating that the SSC encoding method may be useful for analyzing protein sequence-related tasks. The source code of the proposed methods is freely available for academic research at https://github.com/wangy496/SSC-format/ .

Seq-SymRF: a random forest model predicts potential miRNA-disease associations based on information of sequences and clinical symptoms.

Increasing evidence indicates that miRNAs play a vital role in biological processes and are closely related to various human diseases. Research on miRNA-disease associations is helpful not only for disease prevention, diagnosis and treatment, but also for new drug identification and lead compound discovery. A novel sequence- and symptom-based random forest algorithm model (Seq-SymRF) was developed to identify potential associations between miRNA and disease. Features derived from sequence information and clinical symptoms were utilized to characterize miRNA and disease, respectively. Moreover, the clustering method by calculating the Euclidean distance was adopted to construct reliable negative samples. Based on the fivefold cross-validation, Seq-SymRF achieved the accuracy of 98.00%, specificity of 99.43%, sensitivity of 96.58%, precision of 99.40% and Matthews correlation coefficient of 0.9604, respectively. The areas under the receiver operating characteristic curve and precision recall curve were 0.9967 and 0.9975, respectively. Additionally, case studies were implemented with leukemia, breast neoplasms and hsa-mir-21. Most of the top-25 predicted disease-related miRNAs (19/25 for leukemia; 20/25 for breast neoplasms) and 15 of top-25 predicted miRNA-related diseases were verified by literature and dbDEMC database. It is anticipated that Seq-SymRF could be regarded as a powerful high-throughput virtual screening tool for drug research and development. All source codes can be downloaded from https://github.com/LeeKamlong/Seq-SymRF .

Sandwich immunoassay coupled with isothermal exponential amplification reaction: An ultrasensitive approach for determination of tumor marker MUC1.

An ultrasensitive strategy based on sandwich immunoassay coupled with isothermal exponential amplification reaction (IMEXPAR) is proposed for the determination of tumor protein Mucin 1 (MUC1). An immuno-PCR plate was prepared from modification of the primary MUC1-antibody (Ab1) onto the inner-well of the PCR plate. A biotinylated secondary MUC1-antibody tagged with the biotinylated EXPAR primer (P-Ab2) was prepared through biotin-streptavidin reaction. In the presence of target MUC1, sandwich-type combinations were specifically formed in the immuno-PCR plate. With further addition of amplification template, polymerase and nicking enzyme, EXPAR was specifically triggered, producing numerous primer replica in minutes, and greatly enhanced fluorescence of SYBR Green I. The proposed strategy has a good linear relationship with the logarithm of the MUC1 concentration ranging from 3 pM to 3 nM with a limit of detection of 1.63 pM (S/N = 3), which is two orders of magnitude lower than those of other methods. Owing to the specificity of immuno-reaction and EXPAR, the selectivity of the strategy is favorable, even if for the homologous protein. The proposed strategy was further applied for the MUC1 determination in human serum, and a satisfactory recovery range of 98.7%-105.3% was obtained. The strategy can be facilely extended to the ultrasensitive determination of various proteins.

Tensor Maps for Synchronizing Heterogeneous Shape Collections.

Establishing high-quality correspondence maps between geometric shapes has been shown to be the fundamental problem in managing geometric shape collections. Prior work has focused on computing efficient maps between pairs of shapes, and has shown a quantifiable benefit of joint map synchronization, where a collection of shapes are used to improve (denoise) the pairwise maps for consistency and correctness. However, these existing map synchronization techniques place very strong assumptions on the input shapes collection such as all the input shapes fall into the same category and/or the majority of the input pairwise maps are correct. In this paper, we present a multiple map synchronization approach that takes a heterogeneous shape collection as input and simultaneously outputs consistent dense pairwise shape maps. We achieve our goal by using a novel tensor-based representation for map synchronization, which is efficient and robust than all prior matrix-based representations. We demonstrate the usefulness of this approach across a wide range of geometric shape datasets and the applications in shape clustering and shape co-segmentation.

Influence of R and S enantiomers of 1-octen-3-ol on gene expression of Penicillium chrysogenum.

Inhibition of spore germination offers an attractive and effective target for controlling fungal species involved in food spoilage. Mushroom alcohol (1-octen-3-ol) functions as a natural self-inhibitor of spore germination for many fungi and, therefore, provides a useful tool for probing the molecular events controlling the early stages of fungal growth. In Penicillium spp., the R and S enantiomers of 1-octen-3-ol delayed spore germination and sporulation in four species of Penicillium involved in soils of fruit and grains, but to different degrees. Because of its well-annotated genome, we used Penicillium chrysogenum to perform a comprehensive comparative transcriptomic analysis of cultures treated with the two enantiomers. Altogether, about 80% of the high-quality reads could be mapped to 11,396 genes in the reference genome. The top three active pathways were metabolic (978 transcripts), biosynthesis of secondary metabolites (420 transcripts), and microbial metabolism in diverse environments (318 transcripts). When compared to the control, treatment with (R)-(-)-1-octen-3-ol affected the transcription levels of 91 genes, while (S)-(+)-1-octen-3-ol affected only 41 genes. Most of the affected transcripts were annotated and predicted to be involved in transport, establishment of localization, and transmembrane transport. Alternative splicing and SNPs' analyses indicated that, compared to the control, the R enantiomer had greater effects on the gene expression pattern of Penicillium chrysogenum than the S enantiomer. A qRT-PCR analysis of 28 randomly selected differentially expressed genes confirmed the transcriptome data. The transcriptomic data have been deposited in NCBI SRA under the accession number SRX1065226.

Identification of Drug-Disease Associations Using Information of Molecular Structures and Clinical Symptoms via Deep Convolutional Neural Network.

Identifying drug-disease associations is helpful for not only predicting new drug indications and recognizing lead compounds, but also preventing, diagnosing, treating diseases. Traditional experimental methods are time consuming, laborious and expensive. Therefore, it is urgent to develop computational method for predicting potential drug-disease associations on a large scale. Herein, a novel method was proposed to identify drug-disease associations based on the deep learning technique. Molecular structure and clinical symptom information were used to characterize drugs and diseases. Then, a novel two-dimensional matrix was constructed and mapped to a gray-scale image for representing drug-disease association. Finally, deep convolution neural network was introduced to build model for identifying potential drug-disease associations. The performance of current method was evaluated based on the training set and test set, and accuracies of 89.90 and 86.51% were obtained. Prediction ability for recognizing new drug indications, lead compounds and true drug-disease associations was also investigated and verified by performing various experiments. Additionally, 3,620,516 potential drug-disease associations were identified and some of them were further validated through docking modeling. It is anticipated that the proposed method may be a powerful large scale virtual screening tool for drug research and development. The source code of MATLAB is freely available on request from the authors.

Reverse transcription loop-mediated isothermal amplification to rapidly detect Rice ragged stunt virus.

Rice ragged stunt virus (RRSV) is a very important virus that infects rice and causes serious yield losses in Asian countries and other major rice planting areas. Thus, it is urgent to establish an efficient and practical approach for identification and diagnosis in the field. Our results indicated that reverse transcription loop-mediated isothermal amplification (RT-LAMP) reactions are more efficient and sensitive than RT-PCR for RRSV detection. The optimal LAMP conditions were as follows: 0.4-1.2 µM internal primers, 0.2-0.25 µM external primers, 0.8 µM loop primers, and incubation at 62 °C or 63 °C for 30 min. Furthermore, the RT-LAMP primers specifically targeted RRSV virus and resulted in typical waterfall-like bands by gel electrophoresis and sigmoidal amplification curves. The primers could not be used to amplify other common plant viruses including Papaya ringspot virus (PRSV), Rice yellow stunt virus (RYSV), Sorghum mosaic virus (SrMV), Cactus virus X (CVX), Melon yellow spot virus (MYSV) and Southern rice black-streaked dwarf virus (SRBSDV). Ten-fold serial dilutions of RRSV cDNA indicated that RT-LAMP is much faster and at least ten times more sensitive than RT-PCR in detecting the virus. The waterfall-like product bands could be observed within one hour. In the field study, about 77% samples were identified as RRSV. RT-LAMP has many benefits over RT-PCR such as low cost and high accuracy, sensitivity, and specificity. This technology meets the requirements for rapid diagnosis of plant virus diseases in the field to best guide management practices for growers.

Genome-wide comparative analysis of papain-like cysteine protease family genes in castor bean and physic nut.

Papain-like cysteine proteases (PLCPs) are a class of proteolytic enzymes involved in many plant processes. Compared with the extensive research in Arabidopsis thaliana, little is known in castor bean (Ricinus communis) and physic nut (Jatropha curcas), two Euphorbiaceous plants without any recent whole-genome duplication. In this study, a total of 26 or 23 PLCP genes were identified from the genomes of castor bean and physic nut respectively, which can be divided into nine subfamilies based on the phylogenetic analysis: RD21, CEP, XCP, XBCP3, THI, SAG12, RD19, ALP and CTB. Although most of them harbor orthologs in Arabidopsis, several members in subfamilies RD21, CEP, XBCP3 and SAG12 form new groups or subgroups as observed in other species, suggesting specific gene loss occurred in Arabidopsis. Recent gene duplicates were also identified in these two species, but they are limited to the SAG12 subfamily and were all derived from local duplication. Expression profiling revealed diverse patterns of different family members over various tissues. Furthermore, the evolution characteristics of PLCP genes were also compared and discussed. Our findings provide a useful reference to characterize PLCP genes and investigate the family evolution in Euphorbiaceae and species beyond.

SMAC: Simultaneous Mapping and Clustering Using Spectral Decompositions.

We introduce a principled approach for simultaneous mapping and clustering (SMAC) for establishing consistent maps across heterogeneous object collections (e.g., 2D images or 3D shapes). Our approach takes as input a heterogeneous object collection and a set of maps computed between some pairs of objects, and outputs a homogeneous object clustering together with a new set of maps possessing optimal intra- and inter-cluster consistency. Our approach is based on the spectral decomposition of a data matrix storing all pairwise maps in its blocks. We additionally provide tight theoretical guarantees for the accuracy of SMAC under established noise models. We also demonstrate the usefulness of our approach on synthetic and real datasets.

Publisher Correction: Use of RNAi technology to develop a PRSV-resistant transgenic papaya.

A correction to this article has been published and is linked from the HTML version of this paper. The error has been fixed in the paper.

Use of RNAi technology to develop a PRSV-resistant transgenic papaya.

Papaya ringspot virus (PRSV) seriously limits papaya (Carica papaya L.) production in tropical and subtropical areas throughout the world. Coat protein (CP)- transgenic papaya lines resistant to PRSV isolates in the sequence-homology-dependent manner have been developed in the U.S.A. and Taiwan. A previous investigation revealed that genetic divergence among Hainan isolates of PRSV has allowed the virus to overcome the CP-mediated transgenic resistance. In this study, we designed a comprehensive RNAi strategy targeting the conserved domain of the PRSV CP gene to develop a broader-spectrum transgenic resistance to the Hainan PRSV isolates. We used an optimized particle-bombardment transformation system to produce RNAi-CP-transgenic papaya lines. Southern blot analysis and Droplet Digital PCR revealed that line 474 contained a single transgene insert. Challenging this line with different viruses (PRSV I, II and III subgroup) under greenhouse conditions validated the transgenic resistance of line 474 to the Hainan isolates. Northern blot analysis detected the siRNAs products in virus-free transgenic papaya tissue culture seedlings. The siRNAs also accumulated in PRSV infected transgenic papaya lines. Our results indicated that this transgenic papaya line has a useful application against PRSV in the major growing area of Hainan, China.

Greedy Direction Method of Multiplier for MAP Inference of Large Output Domain.

Maximum-a-Posteriori (MAP) inference lies at the heart of Graphical Models and Structured Prediction. Despite the intractability of exact MAP inference, approximate methods based on LP relaxations have exhibited superior performance across a wide range of applications. Yet for problems involving large output domains (i.e., the state space for each variable is large), standard LP relaxations can easily give rise to a large number of variables and constraints which are beyond the limit of existing optimization algorithms. In this paper, we introduce an effective MAP inference method for problems with large output domains. The method builds upon alternating minimization of an Augmented Lagrangian that exploits the sparsity of messages through greedy optimization techniques. A key feature of our greedy approach is to introduce variables in an on-demand manner with a pre-built data structure over local factors. This results in a single-loop algorithm of sublinear cost per iteration and O(log(1/ε))-type iteration complexity to achieve ε sub-optimality. In addition, we introduce a variant of GDMM for binary MAP inference problems with a large number of factors. Empirically, the proposed algorithms demonstrate orders of magnitude speedup over state-of-the-art MAP inference techniques on MAP inference problems including Segmentation, Protein Folding, Graph Matching, and Multilabel prediction with pairwise interaction.

Genome Sequencing and Analysis of the Postharvest Fungus Penicillium expansum R21.

Blue mold is the vernacular name of a common postharvest disease of stored apples, pears, and quince that is caused by several common species of Penicillium This study reports the draft genome sequence of Penicillium expansum strain R21, which was isolated from a red delicious apple in 2011 in Pennsylvania.

Translation Synchronization via Truncated Least Squares.

In this paper, we introduce a robust algorithm, TranSync, for the 1D translation synchronization problem, in which the aim is to recover the global coordinates of a set of nodes from noisy measurements of relative coordinates along an observation graph. The basic idea of TranSync is to apply truncated least squares, where the solution at each step is used to gradually prune out noisy measurements. We analyze TranSync under both deterministic and randomized noisy models, demonstrating its robustness and stability. Experimental results on synthetic and real datasets show that TranSync is superior to state-of-the-art convex formulations in terms of both efficiency and accuracy.

Genome Sequencing and Analysis of the Filamentous Fungus Penicillium sclerotiorum 113, Isolated after Hurricane Sandy.

Penicillium sclerotiorum is a distinctive species within the genus Penicillium that usually produces vivid orange to red colonies, sometimes with colorful sclerotia. Here, we report the first draft genome sequence of P. sclerotiorum strain 113, isolated in 2013 in the aftermath of Hurricane Sandy from a flooded home in New Jersey.