Auxiliary Domain Research Articles

Structure-based functional analysis of a novel NADPH-producing glyceraldehyde-3-phosphate dehydrogenase from Corynebacterium glutamicum

Corynebacterium glutamicum is an industrial workhorse applied in the production of valuable biochemicals. In the process of bio-based chemical production, improving cofactor recycling and mitigating cofactor imbalance are considered major solutions for enhancing the production yield and efficiency. Although, glyceraldehyde-3-phosphate dehydrogenase (GapDH), a glycolytic enzyme, can be a promising candidate for a sufficient NADPH cofactor supply, however, most microorganisms have only NAD-dependent GapDHs. In this study, we performed functional characterization and structure determination of novel NADPH-producing GapDH from C. glutamicum (CgGapX). Based on the crystal structure of CgGapX in complex with NADP cofactor, the unique structural features of CgGapX for NADP stabilization were elucidated. Also, N-terminal additional region (Auxiliary domain, AD) appears to have an effect on enzyme stabilization. In addition, through structure-guided enzyme engineering, we developed a CgGapX variant that exhibited 4.3-fold higher kcat, and 1.2-fold higher kcat/KM values when compared with wild-type. Furthermore, a bioinformatic analysis of 100 GapX-like enzymes from 97 microorganisms in the KEGG database revealed that the GapX-like enzymes possess a variety of AD, which seem to determine enzyme stability. Our findings are expected to provide valuable information for supplying NADPH cofactor pools in bio-based value-added chemical production.

PDA: Progressive Domain Adaptation for Semantic Segmentation

The unsupervised domain adaptation semantic segmentation task is challenging due to the distribution shift problem between the source and the target domains. In this paper, we provide a novel perspective to address this problem. Currently, in the literature, it is shown that output-level domain adaptation networks (OL-DAN) can generate outputs with smaller distribution shift. Motivated by this phenomenon, a Progressive Domain Adaptation (PDA) framework is proposed, which uses the outputs generated by OL-DAN as the auxiliary domain images to progressively align the distribution between the source and target domains. Unlike existing two-stage input-level domain adaptation methods which use an image translation network as a standalone model to generate the auxiliary domain images, the PDA is an end-to-end framework that contains an OL-DAN and a domain fusion domain adaptation network (DF-DAN). The OL-DAN aims to gradually generate the outputs with smaller distribution shift and more accurate semantic structures as the auxiliary domain images in every iteration optimization. The DF-DAN is proposed to further mine the domain-invariant information from the auxiliary images and then fuse the features learned from the original images and the auxiliary images to obtain a richer representation. Finally, the distribution between the source and target domains is aligned to optimize the OL-DAN and DF-DAN. Experiments demonstrate that the proposed PDA achieves superior performance on three cross-domain semantic segmentation benchmarks.

Distinct RPA functions promote eukaryotic DNA replication initiation and elongation.

Replication protein A (RPA) binds single-stranded DNA (ssDNA) and serves critical functions ineukaryoticDNA replication, the DNA damage response,and DNA repair. During DNA replication, RPA is required for extended origin DNA unwinding and DNA synthesis. To determine the requirements for RPA during these processes, we tested ssDNA-binding proteins (SSBs) from different domains of life in reconstituted Saccharomyces cerevisiae origin unwinding and DNA replication reactions. Interestingly, Escherichia coli SSB, but not T4 bacteriophage Gp32, fully substitutes for RPA in promoting origin DNA unwinding. Using RPA mutants, we demonstrated that specific ssDNA-binding properties of RPA are required for origin unwinding but that its protein-interaction domains are dispensable. In contrast, we found that each of these auxiliary RPA domains have distinct functions at the eukaryotic replication fork. The Rfa1 OB-F domain negatively regulates lagging-strand synthesis, while the Rfa2 winged-helix domain stimulates nascent strand initiation. Together, our findings reveal a requirement for specific modes of ssDNA binding in the transition to extensive origin DNA unwinding and identify RPA domains that differentially impact replication fork function.

Distribution shift alignment in visual domain adaptation

Domain adaptation provides the possibility of utilizing the knowledge gained from an auxiliary domain to accomplish the task in another related domain. In this paper, a common feature representation of these domains is learned through a new proposed idea to reduce inter-domain differences more precisely which results in higher accuracy for unsupervised domain adaptation. To decrease these divergences, the proposed method finds a subspace that reduces data distribution discrepancy between domains. To achieve this goal, minimization of the divergence of the class conditional probability distributions is considered. By considering class conditional distributions, more discriminative information of data is preserved compared to applying marginal distributions. Also, an explicit parametric distribution of the source and target domains is considered to reduce the discrepancy between the two domains data which results in higher accuracy compared to the other relevant domain adaptation methods. Experimental studies on benchmark image classification tasks confirmed our assumptions and indicated the significant improvement given by the proposed method compared to the other state-of-the-art methods.

The Impetus of Artificial Intelligence on Periodontal Diagnosis: A Brief Synopsis.

The current advances in digitized data additions, machine learning and computing framework, leadto the swiftly emerging concept of "Artificial Intelligence" (AI), that are developing into areas that were formerly contemplated for human expertise.AI is a relatively rapid paced mechanics wherein the computer technology is tuned to perform human tasks. An auxiliary domain of AI is machine learning (ML), and Deep learning, a subclass ofML technique comprehends multi-layer mathematical operations. AI-based applications have tremendous potential to improve and systematize patient care thereby alleviating dentists from laborious regular tasks, and facilitate personalized, predictive and preventive dentistry. In the dental clinic, AI can execute a variety of easy tasks with greater accuracy, minimal manpower, and with fewer mistakes over human equivalents. These tasks range from appointment scheduling and coordination to helping with clinical evaluation and therapy. Besides, this could assist in the early diagnosis of dental and maxillofacial abnormalities like periodontal ailments, root caries, bony lesions, and facial malformations in addition to automatically identifying and classifying dental restorations on digital radiographs. This brusque narrative review describes the AI-based systems, their respective applications in periodontal diagnosis, the multifarious studies, possible limitations and the predictable future of AI-based dental diagnostics and treatment planning.

Sequential and Graphical Cross-Domain Recommendations with a Multi-View Hierarchical Transfer Gate

Cross-domain recommender systems could potentially improve the recommendation performance by means of transferring abundant knowledge from the auxiliary domain to the target domain. They could help address some key challenges in recommender systems, such as data sparsity and cold start. However, most existing cross-domain recommendation approaches represent the user preferences based on a single kind of user’s feature or behavior and fail to explore the hidden interaction effects of different kinds of features or behaviors. In this article, we propose the S equential and G raphical Cross -Domain Recommendations with a Multi-View Hierarchical Transfer Gate (SGCross) to transfer user representations from multiple perspectives. The SGCross model constructs a user profile by learning the personal preference from a personal view, the dynamic preference from a temporal view, as well as the collaborative preference from a collaborative view. Specifically, a Multi-view Hierarchical Gate (MHG) is designed to transfer the informative representations of user knowledge on different views from the auxiliary domain separately, aiming to enhance the user representations. Furthermore, a two-stage attentive fusion module is designed to integrate transferred information at two levels: the domain level and the view level. Extensive experiments on the Amazon dataset and the Douban dataset have demonstrated that SGCross effectively improves the accuracy of cross-domain recommendations and outperforms the state-of-the-art baseline models.

Cross-domain video action recognition via adaptive gradual learning

Video-based Unsupervised Domain Adaptation (UDA) methods concentrate on addressing domain shift and improving the robustness of video models. It can be naturally applied to cross-domain action recognition tasks, but the inherent complexity of videos makes this task more challenging. Though several recent attempts have achieved superior performance in the video UDA field, most of them intend to mitigate domain shift directly from the source domain to the target domain. These methods will cause a performance drop when the domain distribution shift is large. To better realize the domain adaptation, this paper proposes an effective gradual learning framework by constructing multiple auxiliary domains to achieve progressive transfer for cross-domain video action recognition. Specifically, a Dynamic CutMix Mechanism (DCM) is introduced to build the auxiliary domains that mitigate the domain gap caused by object distance and background discrepancies. Furthermore, the Gradual Transfer Strategy (GTS) utilizes these auxiliary domains to realize the cross-domain action classification gradually. Extensive experiments validate the effectiveness of our proposed method, and the experimental results can significantly outperform state-of-the-art methods on multiple benchmark cross-domain datasets.

Entrepreneurial borrowing overdue prediction based on stacking model transfer learning

PurposeBy introducing Stacking algorithm to solve the underfitting problem caused by insufficient data in traditional machine learning, this paper provides a new solution to the cold start problem of entrepreneurial borrowing risk control.Design/methodology/approachThe authors introduce semi-supervised learning and integrated learning into the field of migration learning, and innovatively propose the Stacking model migration learning, which can independently train models on entrepreneurial borrowing credit data, and then use the migration strategy itself as the learning object, and use the Stacking algorithm to combine the prediction results of the source domain model and the target domain model.FindingsThe effectiveness of the two migration learning models is evaluated with real data from an entrepreneurial borrowing. The algorithmic performance of the Stacking-based model migration learning is further improved compared to the benchmark model without migration learning techniques, with the model area under curve value rising to 0.8. Comparing the two migration learning models reveals that the model-based migration learning approach performs better. The reason for this is that the sample-based migration learning approach only eliminates the noisy samples that are relatively less similar to the entrepreneurial borrowing data. However, the calculation of similarity and the weighing of similarity are subjective, and there is no unified judgment standard and operation method, so there is no guarantee that the retained traditional credit samples have the same sample distribution and feature structure as the entrepreneurial borrowing data.Practical implicationsFrom a practical standpoint, on the one hand, it provides a new solution to the cold start problem of entrepreneurial borrowing risk control. The small number of labeled high-quality samples cannot support the learning and deployment of big data risk control models, which is the cold start problem of the entrepreneurial borrowing risk control system. By extending the training sample set with auxiliary domain data through suitable migration learning methods, the prediction performance of the model can be improved to a certain extent and more generalized laws can be learned.Originality/valueThis paper introduces the thought method of migration learning to the entrepreneurial borrowing scenario, provides a new solution to the cold start problem of the entrepreneurial borrowing risk control system and verifies the feasibility and effectiveness of the migration learning method applied in the risk control field through empirical data.

REMIT: Reinforced Multi-Interest Transfer for Cross-Domain Recommendation

Cold-start problem is one of the most challenging problems for recommender systems. One promising solution to this problem is cross-domain recommendation (CDR) which leverages rich information from an auxiliary source domain to improve the performance of recommender system in the target domain. In particular, the family of embedding and mapping methods for CDR is very effective, which explicitly learn a mapping function from source embeddings to target embeddings to transfer user’s preferences. Recent works usually transfer an overall source embedding by modeling a common or personalized preference bridge for all users. However, a unified user embedding cannot reflect the user’s multiple interests in auxiliary source domain. In this paper, we propose a novel framework called reinforced multi-interest transfer for CDR (REMIT). Specifically, we first construct a heterogeneous information network and employ different meta-path based aggregations to get user’s multiple interests in source domain, then transform different interest embeddings with different meta-generated personalized bridge functions for each user. To better coordinate the transformed user interest embeddings and the item embedding in target domain, we systematically develop a reinforced method to dynamically assign weights to transformed interests for different training instances and optimize the performance of target model. In addition, the REMIT is a general framework that can be applied upon various base models in target domain. Our extensive experimental results on large real-world datasets demonstrate the superior performance and compatibility of REMIT.

Updated protein domain annotation of the PARP protein family sheds new light on biological function.

AlphaFold2 and related computational tools have greatly aided studies of structural biology through their ability to accurately predict protein structures. In the present work, we explored AF2 structural models of the 17 canonical members of the human PARP protein family and supplemented this analysis with new experiments and an overview of recent published data. PARP proteins are typically involved in the modification of proteins and nucleic acids through mono or poly(ADP-ribosyl)ation, but this function can be modulated by the presence of various auxiliary protein domains. Our analysis provides a comprehensive view of the structured domains and long intrinsically disordered regions within human PARPs, offering a revised basis for understanding the function of these proteins. Among other functional insights, the study provides a model of PARP1 domain dynamics in the DNA-free and DNA-bound states and enhances the connection between ADP-ribosylation and RNA biology and between ADP-ribosylation and ubiquitin-like modifications by predicting putative RNA-binding domains and E2-related RWD domains in certain PARPs. In line with the bioinformatic analysis, we demonstrate for the first time PARP14's RNA-binding capability and RNA ADP-ribosylation activity in vitro. While our insights align with existing experimental data and are probably accurate, they need further validation through experiments.

Attention-Guided Autoencoder for Automated Progression Prediction of Subjective Cognitive Decline With Structural MRI.

Subjective cognitive decline (SCD) is the preclinical stage of Alzheimer's disease (AD) which happens even earlier than mild cognitive impairment (MCI). Progressive SCD will convert to MCI with the potential of further evolving to AD. Therefore, early identification of progressive SCD with neuroimaging techniques (e.g., structural MRI) is of great clinical value for early intervention of AD. However, existing MRI-based machine/deep learning methods usually suffer the small-sample-size problem and lack interpretability. To this end, we propose an interpretable autoencoder model with domain transfer learning (IADT) for progression prediction of SCD. Firstly, the proposed model can leverage MRIs from both the target domain (i.e., SCD) and auxiliary domains (e.g., AD and NC) for progressive SCD identification. Besides, it can automatically locate the disease-related brain regions of interest (defined in brain atlases) through an attention mechanism, which shows good interpretability. In addition, the IADT model is straightforward to train and test with only 5 ∼10 seconds on CPUs and is suitable for medical tasks with small datasets. Extensive experiments on the publicly available ADNI dataset and a private CLAS dataset have demonstrated the effectiveness of the proposed method.

Cross-domain recommendation with Multi-Auxiliary Domains via Consistent and Selective Cluster-Level Knowledge Transfer

Recommender systems are software tools and techniques designed to provide users with useful items suggestions. For most existing recommender systems, data sparsity is a challenging and common problem. The cross-domain recommender systems can mitigate the sparsity problem of data in a target domain by introducing auxiliary domain with relatively dense data. However, there are three challenging issues in the existing cross-domain recommender systems: (1) Most cross-domain recommender systems are designed for a single auxiliary domain, which may lead to insufficient knowledge transfer. However, in practice, there may be multiple auxiliary domains related to the target domain that are available; (2) Due to the distribution divergence between the auxiliary and target domains, inconsistent knowledge may be transferred to the target domain; (3) Because of the diversity among domains, they do not necessarily share exactly the same rating pattern. Moreover, not all cluster-level knowledge of auxiliary domains is beneficial to the target domain, and some cluster-level knowledge may be harmful to the target domain. To deal with the above three issues, we propose a Cross-Domain Recommendation with Multi-Auxiliary Domains via Consistent and Selective Cluster-Level Knowledge Transfer, called MD-CSKT. First, rating matrices from multiple auxiliary domains are decomposed jointly with a single target rating matrix to transfer richer knowledge from multiple auxiliary domains. Secondly, the Maximum Mean Discrepancy (MMD) regularization constraint is incorporated into the joint decomposition to reduce the distribution discrepancy between each auxiliary domain and the target domain, so as to achieve consistent cluster-level knowledge transfer. Thirdly, selective cluster-level knowledge transfer is realized by adaptively selecting of cluster-level rating knowledge from multiple auxiliary domains in the joint matrix factorization. The experimental results on six real-world datasets in three categories demonstrate that the presented MD-CSKT method substantially outperforms nine comparison methods and achieves the improvement of recommendation accuracy in the target domain. Compared with the state-of-the-art comparison methods, the overall performance of the MD-CSKT method is improved by more than 25%.

Hybrid Recommendation Network Model with a Synthesis of Social Matrix Factorization and Link Probability Functions

Recommender systems are becoming an integral part of routine life, as they are extensively used in daily decision-making processes such as online shopping for products or services, job references, matchmaking for marriage purposes, and many others. However, these recommender systems are lacking in producing quality recommendations owing to sparsity issues. Keeping this in mind, the present study introduces a hybrid recommendation model for recommending music artists to users which is hierarchical Bayesian in nature, known as Relational Collaborative Topic Regression with Social Matrix Factorization (RCTR-SMF). This model makes use of a lot of auxiliary domain knowledge and provides seamless integration of Social Matrix Factorization and Link Probability Functions into Collaborative Topic Regression-based recommender systems to attain better prediction accuracy. Here, the main emphasis is on examining the effectiveness of unified information related to social networking and an item-relational network structure in addition to item content and user-item interactions to make predictions for user ratings. RCTR-SMF addresses the sparsity problem by utilizing additional domain knowledge, and it can address the cold-start problem in the case that there is hardly any rating information available. Furthermore, this article exhibits the proposed model performance on a large real-world social media dataset. The proposed model provides a recall of 57% and demonstrates its superiority over other state-of-the-art recommendation algorithms.

Recent advances in agricultural disease image recognition technologies: A review

SummaryIn the world of intelligent agriculture, agricultural disease picture detection plays a critical role. Plant disease identification must be efficient if agricultural production is to be increased sustainably. Anomalies in plants affected by viruses, insects, nutritional deficiencies, or poor weather conditions have traditionally been diagnosed by human experts. Deep learning (DL) and transfer learning, two sophisticated machine learning techniques synonymous with the advancement of artificial intelligence technologies in recent years, have started to be used for the identification of agricultural diseases. However, a variety of significant obstacles exist in the way of widespread use of these approaches. This paper looks at DL and transfer learning in particular, and discusses recent developments in using these advanced technologies to recognize agricultural disease images. Many DL architectures have recently been adopted, along with visualization tools, which are critical for identifying signs and classifying plant diseases. The analysis and evaluation of these two approaches show that transfer learning is the better choice given existing agricultural disease data tools. The paper then looks at the key problems that need to be addressed for research in this area to advance, such as image dataset creation, the collection of big data auxiliary domains, and the optimization of the transfer learning process. The construction of technically feasible agricultural disease image recognition systems requires the creation of image datasets collected under real cultivation conditions. This survey aims to learn more about DL capabilities in plant disease detection to increase device efficiency and accuracy in future research.

Investigating the stability of an explicit ADE-FDTD scheme for modeling graphene: Avoiding erroneous conclusions

In recent years, an explicit auxiliary differential equation finite difference time domain (ADE-FDTD) scheme has been frequently used in graphene simulations. In this respect, a differencing scheme in which the magnetic field and the current density are collocated in time was developed and it has been recently stated that this explicit ADE-FDTD implementation is “conditionally stable with the maximum time-step size bounded by the common Courant–Friedrichs–Lewy (CFL) limit”. In this communication, the stability of this explicit ADE-FDTD scheme is revisited and it is shown that the conventional CFL constraint is not retained, and a time-step less than the CFL limit must be used in all domain grids to insure the stability in the whole system. Although this time-step stringent can be of little effect for some graphene applications, and in order to avoid drawing erroneous conclusions, it is shown that when the given ADE-FDTD scheme is used for simulating graphene structures that include noble and other plasmonic materials the derived time-step stringent can be of significant effect. To retain the CFL stability constraint, alternative formulation based on the Runge–Kutta (RK) time-differencing scheme is also presented in this communication. Contrary to the other existing CFL-stable ADE-FDTD approaches, the presented RK-FDTD method is equally applicable to both magnetized and unmagnetized graphene structures without incurring additional computational cost. Furthermore, the RK-FDTD scheme is extended for modeling the complex-frequency-shifted perfectly matched layer (CFS-PML) mesh truncating technique. Finally, the stability and accuracy of the explicit ADE/RK-FDTD schemes are demonstrated through numerical examples.

Partial Domain Adaptation for Scene Classification From Remote Sensing Imagery

Although domain adaptation approaches have been proposed to tackle cross-regional, multitemporal, and multisensor remote sensing applications since they do not require any human interpretation in the target domain, most current works assume identical label space across the source and the target domains. However, in real-world applications, we often transfer knowledge from a large-scale dataset with rich annotations to a small-scale target dataset with scarcity of labels. In most cases, the label space of the source domain is usually large enough to subsume that of the target domain, which is termed partial domain adaptation. In this article, we propose a new partial domain adaptation algorithm for remote sensing scene classification and our proposed method contains three major parts. First, we employ a progressive auxiliary domain module to alleviate the negative transfer effect caused by outlier classes. Second, we adopt an improved domain adversarial neural network (DANN) with multiweights to better encourage domain confusion. Last but not least, we design an attentive complement entropy regularization to improve the prediction confidence for samples and avoid untransferable samples (such as the samples belonging to outlier classes in the source domain) being mistakenly classified. We collect three common remote sensing datasets to evaluate our proposed method. Our method achieves an average accuracy of 79.36%, which considerably outperforms other state-of-the-art partial domain adaptation methods with an average accuracy improvement of 1.90%–12.45% and attaining a 13.67% gain compared to the straightforward deep learning model (ResNet-50). The experiment results indicate that our approach shows promising prospects for solving more general and practical domain adaptation problems where the label space of the source domain subsumes that of the target domain.

Patterns of abundance, chromosomal localization, and domain organization among c-di-GMP-metabolizing genes revealed by comparative genomics of five alphaproteobacterial orders

BackgroundBis-(3′-5′)-cyclic dimeric guanosine monophosphate (c-di-GMP) is a bacterial second messenger that affects diverse processes in different bacteria, including the cell cycle, motility, and biofilm formation. Its cellular levels are controlled by the opposing activities of two types of enzymes, with synthesis by diguanylate cyclases containing a GGDEF domain and degradation by phosphodiesterases containing either an HD-GYP or an EAL domain. These enzymes are ubiquitous in bacteria with up to 50 encoded in some genomes, the specific functions of which are mostly unknown.ResultsWe used comparative analyses to identify genomic patterns among genes encoding proteins with GGDEF, EAL, and HD-GYP domains in five orders of the class Alphaproteobacteria. GGDEF-containing sequences and GGDEF-EAL hybrids were the most abundant and had the highest diversity of co-occurring auxiliary domains while EAL and HD-GYP containing sequences were less abundant and less diverse with respect to auxiliary domains. There were striking patterns in the chromosomal localizations of the genes found in two of the orders. The Rhodobacterales’ EAL-encoding genes and Rhizobiales’ GGDEF-EAL-encoding genes showed opposing patterns of distribution compared to the GGDEF-encoding genes. In the Rhodobacterales, the GGDEF-encoding genes showed a tri-modal distribution with peaks mid-way between the origin (ori) and terminus (ter) of replication and at ter while the EAL-encoding genes peaked near ori. The patterns were more complex in the Rhizobiales, but the GGDEF-encoding genes were biased for localization near ter.ConclusionsThe observed patterns in the chromosomal localizations of these genes suggest a coupling of synthesis and hydrolysis of c-di-GMP with the cell cycle. Moreover, the higher proportions and diversities of auxiliary domains associated with GGDEF domains and GGDEF-EAL hybrids compared to EAL or HD-GYP domains could indicate that more stimuli affect synthesis compared to hydrolysis of c-di-GMP.

Chromatin Sensing by the Auxiliary Domains of KDM5C Regulates Its Demethylase Activity and Is Disrupted by X-linked Intellectual Disability Mutations

The H3K4me3 chromatin modification, a hallmark of promoters of actively transcribed genes, is dynamically removed by the KDM5 family of histone demethylases. The KDM5 demethylases have a number of accessory domains, two of which, ARID and PHD1, lie between the segments of the catalytic domain. KDM5C, which has a unique role in neural development, harbors a number of mutations adjacent to its accessory domains that cause X-linked intellectual disability (XLID). The roles of these accessory domains remain unknown, limiting an understanding of how XLID mutations affect KDM5C activity. Through in vitro binding and kinetic studies using nucleosomes, we find that while the ARID domain is required for efficient nucleosome demethylation, the PHD1 domain alone has an inhibitory role in KDM5C catalysis. In addition, the unstructured linker region between the ARID and PHD1 domains interacts with PHD1 and is necessary for nucleosome binding. Our data suggests a model in which the PHD1 domain inhibits DNA recognition by KDM5C. This inhibitory effect is relieved by the H3 tail, enabling recognition of flanking DNA on the nucleosome. Importantly, we find that XLID mutations adjacent to the ARID and PHD1 domains break this regulation by enhancing DNA binding, resulting in the loss of specificity of substrate chromatin recognition and rendering demethylase activity lower in the presence of flanking DNA. Our findings suggest a model by which specific XLID mutations could alter chromatin recognition and enable euchromatin-specific dysregulation of demethylation by KDM5C.

The role of auxiliary domains in modulating CHD4 activity suggests mechanistic commonality between enzyme families

CHD4 is an essential, widely conserved ATP-dependent translocase that is also a broad tumour dependency. In common with other SF2-family chromatin remodelling enzymes, it alters chromatin accessibility by repositioning histone octamers. Besides the helicase and adjacent tandem chromodomains and PHD domains, CHD4 features 1000 residues of N- and C-terminal sequence with unknown structure and function. We demonstrate that these regions regulate CHD4 activity through different mechanisms. An N-terminal intrinsically disordered region (IDR) promotes remodelling integrity in a manner that depends on the composition but not sequence of the IDR. The C-terminal region harbours an auto-inhibitory region that contacts the helicase domain. Auto-inhibition is relieved by a previously unrecognized C-terminal SANT-SLIDE domain split by ~150 residues of disordered sequence, most likely by binding of this domain to substrate DNA. Our data shed light on CHD4 regulation and reveal strong mechanistic commonality between CHD family members, as well as with ISWI-family remodellers.

Self-Supervised Auxiliary Domain Alignment for Unsupervised 2D Image-Based 3D Shape Retrieval

Unsupervised 2D image-based 3D shape retrieval aims to match the similar 3D unlabeled shapes when given a 2D labeled sample. Although a lot of methods have made a certain degree of progress, the performance of this task is still restricted due to the lack of target labels resulting in tremendous domain gap. In this paper, we aim to explore the discriminative representation of the unlabeled target 3D shapes and facilitate the procedure of domain adaptation by taking full advantage of multi-view information. To achieve the above goals, we propose an effective self-supervised auxiliary domain alignment (SADA) for unsupervised 2D image-based 3D shape retrieval. SADA mainly contains multi-view guided self-supervised feature learning and two auxiliary domain alignments, including intermediate domain alignment and multi-domain alignment. Firstly, we group multiple views of each 3D shape into two sub-target domains based on the view similarities and regard each other as the constraint to optimize the feature learning in an unsupervised manner. To reduce the difficulty of directly aligning the domain discrepancy, we combine the source labeled samples and target samples (pseudo labels) with the same category to generate an intermediate domain, which translates the source-target alignment into source-intermediate and intermediate-target alignments. Moreover, to explore the inner characteristics of target 3D shapes and provide more clues for better adaptation, multi-domain alignment is proposed to convert the source and single target domain alignment to the source and multiple target domain (one target domain and two sub-target domains) alignments. The adversarial training and semantic alignment are employed to fully excavate the relations between source domain and multiple target domains. Experiments on two challenging datasets show that the proposed method achieves competing performance in the unsupervised 2D image-based 3D shape retrieval task.