Structural Similarity Research Articles

COMPUTATIONAL SCREENING OF POTENT ANTI-INFLAMMATORY COMPOUNDS FOR HUMAN MITOGEN-ACTIVATED PROTEIN KINASE: A COMPREHENSIVE AND COMBINED IN SILICO APPROACH

Objective: Inflammatory diseases have a serious impact on one’s life and represent a diverse group of ailments stemming from various causes and presenting in various forms. p38α of the mitogen-activated protein kinase family plays a crucial role in regulating inflammation, where the activation of this kinase initiates a cascade of events resulting in the production of proinflammatory mediators and cellular stress responses. In this context, attempts were made to identify potent small-molecule inhibitors of p38α and assess their binding affinity through molecular docking studies. Methods: From comprehensive reviews of several published reports, a few compounds, such as P38, P39, VPC00628, and N17, have shown substantial inhibitory activity toward p38α at various concentrations. Hence, these four compounds were chosen as lead compounds, and small-molecule libraries were constructed on the basis of their structural similarity. Next, virtual screening docking was performed to investigate the inhibitory potency of the four libraries toward the p38α isoforms (DFG-out and DFG-in), providing insights into their potential mechanisms of action. Results: In addition, a comprehensive analysis of physicochemical and pharmacokinetic properties was also performed for the identified hits from each library. Our findings have shown that, compared with those of the p38α DFG-in motif, the binding energies of the p38α DFG-out motif are greater. Conclusion: Furthermore, a few compounds from each library presented binding energies higher than those of their respective lead compounds, confirming their potential as novel therapeutic agents against inflammation.

Multi-Level Thresholding Color Image Segmentation Using Modified Gray Wolf Optimizer

The success of image segmentation is mainly dependent on the optimal choice of thresholds. Compared to bi-level thresholding, multi-level thresholding is a more time-consuming process, so this paper utilizes the gray wolf optimizer (GWO) algorithm to address this issue and enhance accuracy. To acquire the optimal thresholds at different levels, we modify the GWO (MGWO) in terms of leader selection, position update, and mutation. We also use the Otsu method and Kapur entropy as objective functions. The performance of MGWO is compared with other color image segmentation algorithms on ten images from the BSD500 dataset in terms of objective values, variance, signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and feature similarity index (FSIM). Experimental and non-parametric statistical analyses demonstrate that MGWO performs excellently in the multi-level thresholding segmentation of color images.

Efficient labeling for fine-tuning chest X-ray bone-suppression networks for pediatric patients.

Pneumonia, a major infectious cause of morbidity and mortality among children worldwide, is typically diagnosed using low-dose pediatric chest X-ray [CXR (chest radiography)]. In pediatric CXR images, bone occlusion leads to a risk of missed diagnosis. Deep learning-based bone-suppression networks relying on training data have enabled considerable progress to be achieved in bone suppression in adult CXR images; however, these networks have poor generalizability to pediatric CXR images because of the lack of labeled pediatric CXR images (i.e., bone images vs. soft-tissue images). Dual-energy subtraction imaging approaches are capable of producing labeled adult CXR images; however, their application is limited because they require specialized equipment, and they are infrequently employed in pediatric settings. Traditional image processing-based models can be used to label pediatric CXR images, but they are semiautomatic and have suboptimal performance. We developed an efficient labeling approach for fine-tuning pediatric CXR bone-suppression networks capable of automatically suppressing bone structures in CXR images for pediatric patients without the need for specialized equipment and technologist training. Three steps were employed to label pediatric CXR images and fine-tune pediatric bone-suppression networks: distance transform-based bone-edge detection, traditional image processing-based bone suppression, and fully automated pediatric bone suppression. In distance transform-based bone-edge detection, bone edges were automatically detected by predicting bone-edge distance-transform images, which were then used as inputs in traditional image processing. In this processing, pediatric CXR images were labeled by obtaining bone images through a series of traditional image processing techniques. Finally, the pediatric bone-suppression network was fine-tuned using the labeled pediatric CXR images. This network was initially pretrained on a public adult dataset comprising 240 adult CXR images (A240) and then fine-tuned and validated on 40 pediatric CXR images (P260_40labeled) from our customized dataset (named P260) through five-fold cross-validation; finally, the network was tested on 220 pediatric CXR images (P260_220unlabeled dataset). The distance transform-based bone-edge detection network achieved a mean boundary distance of 1.029. Moreover, the traditional image processing-based bone-suppression model obtained bone images exhibiting a relative Weber contrast of 93.0%. Finally, the fully automated pediatric bone-suppression network achieved a relative mean absolute error of 3.38%, a peak signal-to-noise ratio of 35.5dB, a structural similarity index measure of 98.1%, and a bone-suppression ratio of 90.1% on P260_40labeled. The proposed fully automated pediatric bone-suppression network, together with the proposed distance transform-based bone-edge detection network, can automatically acquire bone and soft-tissue images solely from CXR images for pediatric patients and has the potential to help diagnose pneumonia in children.

Computational engineering of water-soluble human potassium ion channels through QTY transformation.

Transmembrane potassium ion channels are crucial for ion transport, metabolism, and signaling, and serve as promising targets for anti-cancer therapies. However, their hydrophobic transmembrane nature requires detergents, posing a major bottleneck for experimental handling. In this paper, we present a structural bioinformatics study of six experimentally determined and twelve modeled potassium channel structures, in which hydrophobic amino acids (L, I/V, and F) were systematically replaced with neutral hydrophilic ones (Q, T, and Y), making the proteins more water-soluble. QTY (computationally predicted) and native (experimental and repredicted) variants show remarkable structural similarity (RMSD: ~0.50 Å - ~2.14 Å) despite significant sequence differences. QTY variants, both rigid and refined with MD simulations, maintain comparable to native variants stability, solvent-accessible surface area (SASA), and ionic, aromatic, and van der Waals interactions but differ in the grand average of hydropathy (GRAVY), solubility, and hydrophobic contacts. Overall, our study presents a computational approach for designing hydrophilic potassium ion channels while maintaining the native global structure that could potentially simplify their practical use by eliminating the need for detergents.

An integrated approach for studying exposure, metabolism, and disposition of traditional Chinese medicine using PATBS and MDRB tools: a case study of semen Armeniacae Amarum.

Deciphering the in vivo processes of traditional Chinese medicine (TCM) is crucial for identifying new pharmacodynamic substances and new drugs. Due to the complexity and diversity of components, investigating the exposure, metabolism, and disposition remains a major challenge in TCM research. In recent years, a number of non-targeted smart mass-spectrometry (MS) techniques, such as precise-and-thorough background-subtraction (PATBS) and metabolomics, have realized the intelligent identification of in vivo components of TCM. However, the metabolites characterization still largely relies on manual identification in combination with online databases. We developed a scoring approach based on the structural similarity and minimal mass defect variations between metabolites and prototypes. The current method integrates three dimensions of mass spectral data including m/z, mass defect of MS1 and MS2, and the similarity of MS2 fragments, which was sequentially analyzed by a R-based mass dataset relevance bridging (MDRB) data post-processing technique. The MDRB technology constructed a component relationship network for TCM, significantly improving metabolite identification efficiency and facilitating the mapping of translational metabolic pathways. By combining MDRB with PATBS through this non-targeted identification technology, we developed a comprehensive strategy for identification, characterization and bridging analysis of TCM metabolite in vivo. As a proof of concept, we adopted the proposed strategy to investigate the process of exposure, metabolism, and disposition of Semen Armeniacae Amarum (CKXR) in mice. The currently proposed analytical approach is universally applicable and demonstrates its effectiveness in analyzing complex components of TCMs in vitro and in vivo. Furthermore, it enables the correlation of in vitro and in vivo data, providing insights into the metabolic transformationsamong components sharing the same parent nucleus structure. Finally, the developed MDRB platform is publicly available for ( https://github.com/933ZhangDD/MDRB ) for accelerating TCM research for the scientific community.

Effect of soy protein isolate supplementation on the estradiol biosynthesis pathway in a female rat model

Background Soy protein isolate (SPI) has gained popularity as an alternative to animal proteins. Soy consumption may have a link to rising serum estrogen in humans. This concern is attributed to phytoestrogens, specifically isoflavones present in soy protein, acting as estrogen substitutes and modulators. Objective The objective of this research was to assess the influence of SPI supplementation on estradiol hormone levels and the expression of genes associated with estrogen synthesis in female rats. Materials and methods Female Wistar rats (n=18) were evenly divided into three groups: group 1 (normal control) received oral administration of a saline vehicle. Group 2 (low dose) received 450 mg/kg body weight of SPI for 30 days. Group 3 (high dose) received 900 mg/kg body weight of SPI for 30 days. All administrations were conducted intragastrically. Results and conclusion The results of our study indicate that there was a significant increase in the levels of estradiol hormone in groups receiving low and high doses of SPI when compared with the control group. There was an upregulation in the messenger ribonucleic acid expression of the Cyp19 gene in low-dose and high-dose groups. Moreover, the low-dose group showed upregulation in the expression of the HSD3B gene. These genes are involved in the biosynthesis pathway of estrogen hormone in females. Therefore, the use of SPI should be cautious because it contains phytoestrogens (isoflavones), which have structural similarities to endogenous female estrogen and may cause hormonal disturbance in females.

Innovative anti-proliferative effect of the antiviral favipiravir against MCF-7 breast cancer cells using green nanoemulsion and eco-friendly assessment tools.

Telomerase enzyme prevents telomere shortening during division, having human telomerase reverse transcriptase (hTERT) as its catalytic subunit. Favipiravir (FAV), an RNA-dependent RNA polymerases inhibitor, shared structural similarity with hTERT and thus assumed to have cytotoxic effect on cancer cells, in addition to its prophylactic effect to immunocompromised cancer patients. Nanoemulsion (NE) is a potential tumor cells targeting delivery system, thereby enhancing therapeutic efficacy at the intended site, mitigating systemic toxicity, and overcoming multidrug resistance. The objective of this study is to develop a green FAV nanoemulsion (FNE) that is environmentally friendly and safe for patients, while aiming to enhance its cytotoxic effects. The study also highlights the environmental sustainability of the developed RP-HPLC method and assesses its greenness impact. The FNE formulation underwent thermodynamic stability testing and invitro characterization. Greenness was assessed using advanced selected tools like the Analytical Eco-Scale (AES), Analytical Greenness Metric for Sample Preparation (AGREEprep), and green analytical procedure index (GAPI). The cytotoxic potential of FNE was screened against MCF-7 breast cancer and Vero normal cell lines using SRB assay. Stable and ecofriendly FNE was formulated having a particle size (PS) of 25.29 ± 0.57nm and a zeta potential of -6.79 ± 5.52 mV. The cytotoxic effect of FNE on MCF-7 cells was more potent than FAV with lower IC50 while FNE showed non-toxic effect on VERO normal cell line. Therefore, the FAV nanoemulsion formulation showed targeted cytotoxicity on MCF-7 cells while being non-toxic on normal Vero cells.

The role of surface and structural similarities in the retrieval of realistic perceptual events.

This study investigated whether structural similarities (i.e. abstract frames, e.g. once bitten twice shy) can prevail over surface similarities (i.e. contexts, e.g. restaurant) in driving the retrieval of realistic events involving dynamic, multimodal and perceptually crowded data. After watching an initial set of video clips, participants had to indicate whether a new video clip, that shared surface similarities with an initial event and structural similarities with another one, elicited a retrieval. The results of Experiment 1A showed that retrieval was more likely to be elicited by structural rather than by surface similarities. Experiment 1B confirmed that the surface similarities manipulated in this study were strong enough to elicit substantial retrievals when the competing structural match was neutralized. The pattern of results obtained in Experiment 1A remained unchanged when the number of unrelated video clips within the initial set was increased. The findings suggest that structurally based retrievals still prevail when familiar structures underlie realistic perceptual events. They open new perspectives regarding the settings that promote structurally based retrievals in educational contexts where unfamiliar principles are introduced.

Feedback projection synchronization of discrete chaotic systems and its application to speech encryption

Abstract Chaotic systems are widely used in secure communication due to their sensitivity to initial values, unpredictability, and complex motion trajectories. In this paper, we study the encryption method of chaotic synchronization and introduce a scaling factor based on traditional feedback control synchronization to achieve more accurate projection synchronization. The effectiveness and robustness of the method in chaotic systems are verified through theoretical proofs and numerical simulations. A chaotic masked speech encryption system utilizing bit similarity is designed; the structural similarity index (SSIM) of the decrypted signal with the original signal is as high as 0.992866, while the SSIM value of the encrypted signal with the original signal is only 0.000030, proving the efficiency and security of the encryption process. Additionally, we analyzed the data transmission process of the encryption system. The fusion of the control signal and the encryption sequence into one transmission sequence in the channel not only saves hardware and software design resources but also reduces inter-channel interference and conflict, improving the reliability and stability of the transmission. Experimental results show that the system performs well in terms of data transmission security and anti-interference capability.

Image enhancement with art design: a visual feature approach with a CNN-transformer fusion model

Graphic design, as a product of the burgeoning new media era, has seen its users’ requirements for images continuously evolve. However, external factors such as light and noise often cause graphic design images to become distorted during acquisition. To enhance the definition of these images, this paper introduces a novel image enhancement model based on visual features. Initially, a histogram equalization (HE) algorithm is applied to enhance the graphic design images. Subsequently, image feature extraction is performed using a dual-flow network comprising convolutional neural network (CNN) and Transformer architectures. The CNN employs a residual dense block (RDB) to embed spatial local structure information with varying receptive fields. An improved attention mechanism module, attention feature fusion (AFF), is then introduced to integrate the image features extracted from the dual-flow network. Finally, through image perception quality guided adversarial learning, the model adjusts the initial enhanced image’s color and recovers more details. Experimental results demonstrate that the proposed algorithm model achieves enhancement effects exceeding 90% on two large image datasets, which represents a 5%–10% improvement over other models. Furthermore, the algorithm exhibits superior performance in terms of peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) image quality evaluation metrics. Our findings indicate that the fusion model significantly enhances image quality, thereby advancing the field of graphic design and showcasing its potential in cultural and creative product design.

Two groups and three classes of the conserved structural organization of nucleophile and non-canonical ElbowFlankOxy networks in different superfamily proteins

The nucleophile elbow is a well-known structural motif, which exists in proteins with catalytic triads and contains a catalytic nucleophile and the first node of an oxyanion hole. Here, we show that structural similarities of proteins with the nucleophile elbow extend beyond simple nucleophile elbow motifs. The motifs are incorporated into larger conserved structural organizations, the ElbowFlankOxy networks, incorporating motifs and flanking residues and networks of conserved interactions. A detailed structural analysis shows two major types of ElbowFlankOxy networks, depending on the formation of the oxyanion hole. Additionally, the ElbowFlankOxy networks show three classes: Class 1-2-3, 3-1-2, and 2-3-1, defined by the order in which the catalytic nucleophile and key interacting residues are located in the amino acid sequence, giving rise to six ElbowFlankOxy network variations. This makes it possible to properly position homologous non-catalytic, non-standard, and unusual catalytic triad active sites of proteins with the nucleophile elbow within the fold classification.

Dual-encoder architecture for metal artifact reduction for kV-cone-beam CT images in head and neck cancer radiotherapy.

During a radiotherapy(RT) course, geometrical variations of target volumes, organs at risk, weight changes (loss/gain), tumor regression and/or progression can significantly affect the treatment outcome. Adaptive RT has become the effective methods along with technical advancements in imaging modalities including cone-beam computed tomography (CBCT). Planning CT (pCT) can be modified via deformable image registration (DIR), which is applied to the pair of pCT and CBCT. However, the artifact existed in both pCT and CBCT is a vulnerable factor in DIR. The dose calculation on CBCT is also suggested. Missing information due to the artifacts hinders the accurate dose calculation on CBCT. In this study, we aim to develop a deep learning-based metal artifact reduction(MAR) model to reduce the metal artifacts in CBCT for head and neck cancer RT. To train the proposed MAR model, we synthesized the kV-CBCT images including metallic implants, with and without metal artifacts (simulated image data pairs) through sinogram image handling process. We propose the deep learning architecture which focuses on both artifact removal and reconstruction of anatomic structure using a dual-encoder architecture. We designed four single-encoder models and three dual-encoder models based on UNet (for an artifact removal) and FusionNet (for a tissue restoration). Each single-encoder model contains either UNet or FusionNet, while the dual-encoder models have both UNet and FusionNet architectures. In the dual-encoder models, we implemented different feature fusion methods, including simple addition, spatial attention, and spatial/channel wise attention. Among the models, a dual-encoder model with spatial/channel wise attention showed the highest scores in terms of peak signal-to-noise ratio, mean squared error, structural similarity index, and Pearson correlation coefficient. CBCT images from 34 head and neck cancer patients were used to test the developed models. The dual-encoder model with spatial/channel wise attention showed the best results in terms of artifact index. By using the proposed model to CBCT, one can achieve more accurate synthetic pCT for head and neck patients as well as better tissue recognition and structure delineation for CBCT image itself.

A large-scale screening identifies receptor-like kinases with common features in kinase domains that are potentially related to disease resistance in planta

IntroductionThe plant genome encodes a plethora of proteins with structural similarity to animal receptor protein kinases, collectively known as receptor-like protein kinases (RLKs), which predominantly localize to the plasma membrane where they activate their kinase domains to convey extracellular signals to the interior of the cell, playing crucial roles in various signaling pathways. Despite the large number of members within the RLK family, to date, only a few have been identified as pattern-recognition receptors (PRRs), leaving many potential RLKs that could play roles in plant immunity undiscovered.MethodsIn this study, a recombinant strategy was initially employed to screen the kinase domains of 133 RLKs in the Arabidopsis genome to determine their involvement in the pathogen-triggered immunity (PTI) pathway. Subsequently, 6 potential immune-related recombinant RLKs (rRLKs) were selected for the creation of transgenic materials and underwent functional characterization analysis. Finally, a sequence analysis was conducted on the kinase domains of these 133 RLKs as well as the known immune RLK receptor kinase domains from other species.ResultsIt was found that 24 rRLKs activated the PTI response in Arabidopsis fls2 mutant protoplasts following flg22 treatment. Consistently, when 6 of these rRLKs were individually expressed in fls2 background, they exhibited diverse PTI signal transduction capabilities via different pathways while all retained membrane localization. Intriguingly, sequence analysis revealed multiple conserved amino acid sites within kinase domains of these experimentally identified immune-related RLKs in Arabidopsis. Importantly, these patterns are also preserved in RLKs involved in PTI in other species.DiscussionThis study, on one hand, identifies common features that theoretically can enhance our understanding of immune-related RLKs and facilitate the discovery of novel immune-related RLKs in the future. On the other hand, it provides experimental evidence for the use of recombinant technique to develop diverse rRLKs for molecular breeding, thereby conferring high resistance to plants without compromising their normal growth and development.

Pullenvalenes E-H: Triterpenyl-Aminoglycosides from an Australian Soil-Derived Fungus, Clonostachys sp.

Chemical profiling of soil-derived microbes collected under the auspices of the Australian citizen science initiative Soils for Science detected two fungi, Clonostachys sp. S4S-07771A07 and Coccidiodes sp. S4S-14879B01, capable of producing pullenvalenes, a rare class of triterpene glycoside. Cultivation profiling followed by scaled up cultivation and fractionation of the former yielded the known pullenvalenes A-D (1-4) and the new analogues E-H (5-8), with structures secured by detailed spectroscopic analysis and biogenetic considerations. This study reveals that the pullenvalenes 1-8 are produced by several genera of fungi (Clonostachys, Coccidiodes and Talaromyces) recovered from different geographic locations and substrates. We also draw attention to structural and biosynthetic similarities with the known Red Sea sponge metabolites neviotines A-D (9-12) and abudinols A-B (13-14), prompting speculation that the latter may be products of sponge-associated fungi.

Maturation of Tertiary Lymphoid Structures.

Tertiary lymphoid structures (TLS) are organized collections of B and T lymphocytes that arise in nonlymphoid tissue in response to chronic, unresolved inflammation. TLS have structural and functional similarities to germinal centers found in lymph nodes and are believed to support the establishment of lymph node-like adaptive immune responses at local sites of inflammation. However, understanding of the underlying biology of these structures remains limited, particularly the different stages of TLS life cycle and the signals governing the initiation, maturation, and termination of TLS. Here, we review current understanding of the maturation of TLS and the signals and cell types involved in various stages of development with particular emphasis on recent studies of TLS in cancer, where evidence suggests that TLS may play an important role in supporting antitumor immune responses in solid tumors.

Film image processing and production based on high-performance calculation

In film and television production, efficient and precise image processing is vital for achieving realistic visual effects. Therefore, exploring and applying advanced image processing technologies has become an essential method for elevating the production quality of film and television projects. This work investigates the application of artificial intelligence (AI) technology in the processing and production of animated images in film and television scenarios. By comparing the performance of standard Generative Adversarial Network (GAN), DenseNet, and CycleGAN models under different noise conditions, it is found that CycleGAN performs the best in image denoising and detail restoration. Experimental results demonstrate that CycleGAN achieves a Peak Signal-to-noise Ratio (PSNR) of 30.1dB and a Structural Similarity Index Measure (SSIM) of 0.88 under Gaussian noise conditions. Moreover, CycleGAN achieves a PSNR of 29.5dB and an SSIM of 0.85 under salt-and-pepper noise conditions. It outperforms the other models in both conditions. Additionally, CycleGAN’s mean absolute error is significantly lower than that of the other models. This work demonstrates that CycleGAN can more effectively handle complex noise and generate high-quality images under unsupervised learning conditions. These findings provide new directions for future image processing research and offer important references for model selection in practical applications. This work not only offers new perspectives on the development of animation image processing technology but also establishes a theoretical foundation for applying advanced AI techniques in film and television production. Through comparative analysis of various deep learning models, this work highlights the superior performance of CycleGAN under complex noise conditions. This advancement not only drives progress in image processing technology but also provides effective solutions for efficient production and quality enhancement of future film and television works.

Melt Spinnability Comparison of Mechanically and Chemically Recycled Polyamide 6 for Plastic Waste Reuse

With the increasing volume of synthetic fiber waste, interest in plastic reuse technologies has grown. To address this issue, physical and chemical recycling techniques for polyamide, a major component of textile waste, have been developed. This study investigates the remelting and reforming properties of four types of pristine and recycled polyamide 6, focusing on how the microstructural arrangement of recycled polyamides affects polymer fiber formation. DSC and FT-IR were used to determine the thermal properties and chemical composition of the reformed thin films. Differences in the elongation behavior of molten fibers during the spinning process were also observed, and the morphology of the resulting fibers was examined via SEM. Birefringence analysis revealed that the uniformity of the molecular structure greatly influenced differences in the re-fiberization process, suggesting that chemically recycled polyamide is the most suitable material for re-fiberization with its high structural similarity to pristine polyamide.

MCRANet: MTSL-based connectivity region attention network for PD-L1 status segmentation in H&E stained images

The quantitative analysis of Programmed death-ligand 1 (PD-L1) via Immunohistochemical (IHC) plays a crucial role in guiding immunotherapy. However, IHC faces challenges, including high costs, time consumption and result variability. Conversely, Hematoxylin-Eosin (H&E) staining offers cost-effectiveness, speed, and stable results. Nonetheless, H&E staining, which solely visualizes cellular morphological features, lacks clinical applicability in detecting biomarker expressions like PD-L1. Substituting H&E staining for IHC in determining PD-L1 status is a clinically significant and challenging task. Motivated by above observations, we propose a Multi-Task supervised learning (MTSL)-based connectivity region attention network (MCRANet) for PD-L1 status segmentation in H&E stained images. To reduce interference from non-tumor areas, the MTSL-based region attention is proposed to enhances the network's capability to distinguish between tumor and non-tumor regions. Consequently, this augmentation further improves the network's segmentation efficacy for PD-L1 positive and negative regions. Furthermore, the PD-L1 expression regions demonstrate interconnection throughout the tissue section. Leveraging this topological prior knowledge, we integrate a connectivity modeling module (CM module) within the MTSL-based region attention module (MRA module) to enhance the precision of MTSL-based region attention localization. This integration further improves the structural similarity between the segmentation results and the ground truth. Extensive visual and quantitative results demonstrate that our supervised-learning-guided MRA module produces more interpretable attention and the introduced CM module provides accurate positional attention to the MRA module. Compared to other state-of-the-art networks, MCRANet exhibits superior segmentation performance with a dice similarity coefficient (DSC) of 79.6 % on the lung squamous cell carcinoma (LUSC) PD-L1 status dataset.

Bimodal PET/MRI generative reconstruction based on VAE architectures.

•Objective:In this study, we explore positron emission tomography(PET)/magnetic resonance imaging (MRI) joint reconstruction within a deeplearning (DL) framework, introducing a novel synergistic method.
•Approach:We propose a new approach based on a variational autoencoder (VAE)constraint combined with the alternating direction method of multipliers (ADMM)optimization technique. We compare several VAE architectures, including jointVAE, mixture of experts (MoE) and product of experts (PoE), to determine theoptimal latent representation for the two modalities. We trained then evaluatedthe architectures on a brain PET/MRI dataset.
•Main results:We showed that our approach takes advantage of each modalitysharing information to each other, which results in improved peak signal-to-noiseratio (PSNR) and structural similarity (SSIM) as compared with traditionalreconstruction methods, particularly for short acquisition times. We find that theone particular architecture, MMJSD, is the most effective for our methodology.
•Significance:The proposed method outperforms classical approaches especiallyin noisy and undersampled conditions by making use of the two modalities together to compensate for the missing information.

MFIB 2.0: a major update of the database of protein complexes formed by mutual folding of the constituting protein chains.

While the majority of proteins with available structures are able to fold independently and mediate interactions only after acquiring their folded state, a subset of the known protein complexes contains protein chains that are intrinsically disordered in isolation. The Mutual Folding Induced by Binding (MFIB) database collects and classifies protein complexes, wherein all constituent protein chains would be unstable/disordered in isolation but fold into a well-defined 3D complex structure upon binding. This phenomenon is often termed as cooperative folding and binding or mutual synergistic folding (MSF). Here we present a major update to the database: we collected and annotated hundreds of new protein complexes fulfilling the criteria of MSF, leading to an almost six-fold increase in the size of the database. Many novel features have also been introduced, such as clustering of the complexes based on structural similarity and domain types, assigning different evidence levels to each entry and adding the evidence coverage label that allowed us to include complexes of multi(sub)domain monomers with partial MSF. The MFIB 2.0 database is available at https://mfib.pbrg.hu.