Glia-to-Neuron Conversion by CRISPR-CasRx Alleviates Symptoms of Neurological Disease in Mice.

Conversion of glial cells into functional neurons represents a potential therapeutic approach for replenishing neuronal loss associated with neurodegenerative diseases and brain injury. Previous attempts in this area using expression of transcription factors were hindered by the low conversion efficiency and failure of generating desired neuronal types in vivo. Here, we report that downregulation of a single RNA-binding protein, polypyrimidine tract-binding protein 1 (Ptbp1), using in vivo viral delivery of a recently developed RNA-targeting CRISPR system CasRx, resulted in the conversion of Müller glia into retinal ganglion cells (RGCs) with a high efficiency, leading to the alleviation of disease symptoms associated with RGC loss. Furthermore, this approach also induced neurons with dopaminergic features in the striatum and alleviated motor defects in a Parkinson's disease mouse model. Thus, glia-to-neuron conversion by CasRx-mediated Ptbp1 knockdown represents a promising in vivo genetic approach for treating a variety of disorders due to neuronal loss.

Modeling the Zigzag Curves of Transition Metal Ions' Charge Transition Levels in Crystals.

Predicting the defect levels of transition metal (TM) dopants in the band gap of crystals is critical in determining the charge states of TM dopants and explaining their electronic and optical properties. By analyzing the calculated charge transition levels and the crystal-field strengths of all the 3d-TM ions in several insulators, we demonstrate that the variation trend of the 3d-TM dopants in a crystal is a scaling of the variation of 3d-electron binding energies (ionization potential) of the free TM ions corrected by adding the contribution of the 3d-orbital's crystal-field splitting. We therefore develop a model to predict the relative location of TM ions' defect levels in the band gap from the defect level and crystal-field splitting of a reference TM ion in the host of concern. The model is applied to predict the defect levels of the series of TM ions in ß-Ga2O3 and ZnO, which have moderate to small band gaps, making some of the levels fall into the conduction or valence bands. These results show that the model may serve as a quick reference for related material design and optimization.

An Effective Plant Small Secretory Peptide Recognition Model Based on Feature Correction Strategy.

Plant small secretory peptides (SSPs) play an important role in the regulation of biological processes in plants. Accurately predicting SSPs enables efficient exploration of their functions. Traditional experimental verification methods are very reliable and accurate, but they require expensive equipment and a lot of time. The method of machine learning speeds up the prediction process of SSPs, but the instability of feature extraction will also lead to further limitations of this type of method. Therefore, this paper proposes a new feature-correction-based model for SSP recognition in plants, abbreviated as SE-SSP. The model mainly includes the following three advantages: First, the use of transformer encoders can better reveal implicit features. Second, design a feature correction module suitable for sequences, named 2-D SENET, to adaptively adjust the features to obtain a more robust feature representation. Third, stack multiple linear modules to further dig out the deep information on the sample. At the same time, the training based on a contrastive learning strategy can alleviate the problem of sparse samples. We construct experiments on publicly available data sets, and the results verify that our model shows an excellent performance. The proposed model can be used as a convenient and effective SSP prediction tool in the future. Our data and code are publicly available at https://github.com/wrab12/SE-SSP/.

ML-NPI: Predicting Interactions between Noncoding RNA and Protein Based on Meta-Learning in a Large-Scale Dynamic Graph.

Deep learning methods can accurately study noncoding RNA protein interactions (NPI), which is of great significance in gene regulation, human disease, and other fields. However, the computational method for predicting NPI in large-scale dynamic ncRNA protein bipartite graphs is rarely discussed, which is an online modeling and prediction problem. In addition, the results published by researchers on the Web site cannot meet real-time needs due to the large amount of basic data and long update cycles. Therefore, we propose a real-time method based on the dynamic ncRNA-protein bipartite graph learning framework, termed ML-GNN, which can model and predict the NPIs in real time. Our proposed method has the following advantages: first, the meta-learning strategy can alleviate the problem of large prediction errors in sparse neighborhood samples; second, dynamic modeling of newly added data can reduce computational pressure and predict NPIs in real-time. In the experiment, we built a dynamic bipartite graph based on 300000 NPIs from the NPInterv4.0 database. The experimental results indicate that our model achieved excellent performance in multiple experiments. The code for the model is available at https://github.com/taowang11/ML-NPI, and the data can be downloaded freely at http://bigdata.ibp.ac.cn/npinter4.

Mechanistic insights for regulating the site occupancy, valence states and optical transitions of Mn ions in yttrium-aluminum garnets via codoping.

The site-dependent photoluminescence of activators can be regulated by the sintering atmosphere, coexistence conditions, and especially cation codoping, which have been intensively studied for design and optimization of optical functional materials. Here, first-principles calculations are performed to determine the regulation of the site occupancy, valence states and optical transitions of Mn activators via codoping in yttrium aluminum garnets (YAGs), which contain three different cation sites. Without any codopants, Mnoct3+ dominates in defect concentration and photoluminescence, which can hardly be tuned by the sintering atmosphere or coexistence conditions of YAGs with other competing compounds. With the low formation energy of Ca2+, Be2+, Mg2+, and Sr2+ codopants and in an oxidation sintering atmosphere, the Fermi energy is lowered and the concentration and luminescence of Mnoct4+ are enhanced. Na+ and Li+ codopants with relatively high formation energy have little influence on tuning the Fermi energy. Then with the low formation energy of Ti4+, Si4+ codopants and in a reducing sintering atmosphere, the Fermi energy is lifted and the luminescence of Mndod2+ and Mnoct2+ is enhanced as a result of increased concentrations. The proposed first-principles scheme, with general applicability and encouraging predictive power, provides an effective approach for elucidating the effects of codoping impurities on the design and optimization of optical materials.

Recent Advances in Sensor Technology for Healthcare and Biomedical Applications (Volume II).

With remarkable progress being witnessed in recent years in the development of sensors, these advances in sensor technology provide unprecedented opportunities for (1) the early diagnosis and prevention of human diseases by detecting critical biomarkers; (2) health assessments by monitoring and analyzing human physiological signals in healthcare and biomedical applications; and (3) the efficient evaluation of human-health-relevant environmental factors by monitoring and measuring environmental determinants [...].

Multimodal Image Fusion for X-ray Grating Interferometry.

X-ray grating interferometry (XGI) can provide multiple image modalities. It does so by utilizing three different contrast mechanisms-attenuation, refraction (differential phase-shift), and scattering (dark-field)-in a single dataset. Combining all three imaging modalities could create new opportunities for the characterization of material structure features that conventional attenuation-based methods are unable probe. In this study, we proposed an image fusion scheme based on the non-subsampled contourlet transform and spiking cortical model (NSCT-SCM) to combine the tri-contrast images retrieved from XGI. It incorporated three main steps: (i) image denoising based on Wiener filtering, (ii) the NSCT-SCM tri-contrast fusion algorithm, and (iii) image enhancement using contrast-limited adaptive histogram equalization, adaptive sharpening, and gamma correction. The tri-contrast images of the frog toes were used to validate the proposed approach. Moreover, the proposed method was compared with three other image fusion methods by several figures of merit. The experimental evaluation results highlighted the efficiency and robustness of the proposed scheme, with less noise, higher contrast, more information, and better details.

Color-changing fluorescent DNA probe containing solvatochromic Dansyl-nucleoside surrogate for sensing local variation of DNA duplex.

A novel Dansyl-nucleoside surrogate (Dns) based on (±)-trans-4-(hydroxymethyl) piperidin-3-ol was designed and synthesized. The Dns exhibited excellent solvatochromic properties. About 90 nm of red-shift accompanied color change from green to orange could be achieved with an increase of solvent polarity. The Dns was incorporated into oligodeoxynucleotide by phosphoroamidite chemistry. Two kinds of Dns-incorporated fluorescent DNA probes were designed and synthesized for sensing variation of DNA duplexes based on color-changing manner. As a result, the color-changing DNA probe not only can detect complementary oligonucleotide, but also can distinguish mismatch flanked in Dansyl/nucleobase pair by naked eye. Moreover, the change of fluorescence color of sample solutions could be captured by smartphone, and the photographs could be digitalized by image-processing software. Thus, the Dns-incorporated fluorescent DNA probe is expected to open the way to point-of-care assays in the future.

Elucidating the Multisite and Multivalence Nature of Mn Ions in Solids and Predicting Their Optical Transition Properties: A Case Study on a Series of Garnet Hosts.

The abundant site occupancy and optical transitions of multivalence Mn dopants in luminescent materials have attracted much attention. Here, detailed first-principles calculations based on density functional theory have been carried out to clarify the multisite and multivalence nature of Mn ions in solids and predict their optical transition properties by using garnets as prototype systems. The formation energies of dodecahedral, octahedral, and tetrahedral coordinated Mn dopants are evaluated with chemical potential environments, and the preferable site occupancy and valence state of Mn ions in three garnet systems are clarified. The results show that even in a fixed atmosphere, taking Ca3Al2Ge3O12 in air as an example, not only can the preference of Mn ions switch between dodecahedral and octahedral sites, but also can the valence state change from Mn2+ to Mn3+ and Mn4+. Furthermore, for all of the three garnet systems, the calculation results of the energy-level structure and photoluminescence of Mn ions at different sites in the different valence states provide a reliable interpretation of the available spectroscopic data. The proposed first-principles scheme, with general applicability and encouraging predictive power, provides an effective approach for elucidating and characterizing the site occupancy, valence state, and optical transition of Mn activators in insulators, and will greatly benefit the design and optimization of related materials.

The photoluminescence of isolated and paired Bi3+ ions in layered LnOCl crystals: a first-principles study.

Luminescent ns2 centers have shown great potential for applications as phosphors and scintillators. First-principles calculations based on density functional theory are performed to systematically analyze the luminescent centers of isolated and paired Bi3+(6s2) ions in layered LnOCl (Ln = Y, Gd, La) crystals. The spin-orbit coupling and orbital hybridization both show important effects on the luminescence properties. The luminescence of the isolated Bi ion is confirmed as the interconfigurational transition of 3P0,1 → 1S0. For the Bi pair, the adiabatic potential energy surfaces are calculated and the charge transfer excited state is the most stable, which accounts for the visible emission of a large Stokes shift. Furthermore, the electron-hole pair separation, absorption, excitonic state and emission of the material with fully-concentrated Bi3+, BiOCl, are discussed. This study shows that the first-principles calculations can serve as an effective tool for the photoluminescence analysis and engineering of materials activated with isolated, paired and even fully-concentrated ns2 ions.

Synaptic control of spinal GRPR+ neurons by local and long-range inhibitory inputs.

Spinal gastrin-releasing peptide receptor-expressing (GRPR+) neurons play an essential role in itch signal processing. However, the circuit mechanisms underlying the modulation of spinal GRPR+ neurons by direct local and long-range inhibitory inputs remain elusive. Using viral tracing and electrophysiological approaches, we dissected the neural circuits underlying the inhibitory control of spinal GRPR+ neurons. We found that spinal galanin+ GABAergic neurons form inhibitory synapses with GRPR+ neurons in the spinal cord and play an important role in gating the GRPR+ neuron-dependent itch signaling pathway. Spinal GRPR+ neurons also receive inhibitory inputs from local neurons expressing neuronal nitric oxide synthase (nNOS). Moreover, spinal GRPR+ neurons are gated by strong inhibitory inputs from the rostral ventromedial medulla. Thus, both local and long-range inhibitory inputs could play important roles in gating itch processing in the spinal cord by directly modulating the activity of spinal GRPR+ neurons.

Supramolecular Recognition of Three Way Junction DNA by a Cationic Calix[3]carbazole.

DNA three-way junctions (TWJ-DNA) are intermediate structures in DNA replication and/or recombination. They play very important roles in biological processes, but more subtle functions are still unknown due partially to the lack of a fluorescent ligand. In this study, a cationic calix[3]carbazole (2) has been synthesized and its properties of interacting with TWJ-DNA have been evaluated by UV/Vis and fluorescence spectroscopy, circular dichroism (CD), gel electrophoresis, and 1 H NMR studies. The results show that 2 binds to the central hydrophobic cavity of TWJ-DNA. Moreover, it could selectively bind to TWJ-DNA over duplex and quadruplex DNA. Furthermore, 2 possesses the capability of serving as the TWJ-DNA probe as its trap-II excimer emission is turned on by TWJ-DNA.

Identification, duplication, evolution and expression analyses of caleosins in Brassica plants and Arabidopsis subspecies.

Caleosins are a class of Ca(2+) binding proteins that appear to be ubiquitous in plants. Some of the main proteins embedded in the lipid monolayer of lipid droplets, caleosins, play critical roles in the degradation of storage lipids during germination and in lipid trafficking. Some of them have been shown to have histidine-dependent peroxygenase activity, which is believed to participate in stress responses in Arabidopsis. In the model plant Arabidopsis thaliana, caleosins have been examined extensively. However, little is known on a genome-wide scale about these proteins in other members of the Brassicaceae. In this study, 51 caleosins in Brassica plants and Arabidopsis lyrata were investigated and analyzed in silico. Among them, 31 caleosins, including 7 in A. lyrata, 11 in Brassica oleracea and 13 in Brassica napus, are herein identified for the first time. Segmental duplication was the main form of gene expansion. Alignment, motif and phylogenetic analyses showed that Brassica caleosins belong to either the H-family or the L-family with different motif structures and physicochemical properties. Our findings strongly suggest that L-caleosins are evolved from H-caleosins. Predicted phosphorylation sites were differentially conserved in H-caleosin and L-caleosins, respectively. 'RY-repeat' elements and phytohormone-related cis-elements were identified in different caleosins, which suggest diverse physiological functions. Gene structure analysis indicated that most caleosins (38 out of 44) contained six exons and five introns and their intron phases were highly conserved. Structurally integrated caleosins, such as BrCLO3-3 and BrCLO4-2, showed high expression levels and may have important roles. Some caleosins, such as BrCLO2 and BoCLO8-2, lost motifs of the calcium binding domain, proline knot, potential phosphorylation sites and haem-binding sites. Combined with their low expression, it is suggested that these caleosins may have lost function.

A dual functional peptide carrying in vitro selected catalytic and binding activities.

When minimal functional sequences are used, it is possible to integrate multiple functions on a single peptide chain, like a "single stroke drawing". Here a dual functional peptide was designed by combining in vitro selected catalytic and binding activities. For catalytic activity, we performed in vitro selection for a peptide aptamer binding to hemin by using ribosome display and isolated a peptide that had peroxidase activity in the presence of hemin. By combining the selected catalytic peptide with a peptide antigen, which can be recognized by an antibody, an enzyme-antibody conjugate-like peptide was obtained. This study demonstrates a successful strategy to create dual functionalized peptide chains for use in immunoassays.

[Randomized, blind, parallel-controlled and multiple-centre clinical trial on the effectiveness and safety of leuprolide acetate in the treatment of endometriosis].

OBJECTIVE: To evaluate the effectiveness and safety of leuprolide acetate in the treatment of endometriosis. METHODS: From Nov. 2007 to Oct. 2012, the patients who confirmed to be endometriosis were randomly divided into test group of 113 cases and control group of 116 cases. The test drug was the sustained-release agent of leuprolide acetate. The control drug was Enantone. The drugs were used for 3 times in total. After treatment, the ovarian mass volumes measured with type-B ultrasound, the scores of the patient's subjective symptoms during non-menstrual and menstruation days, the pelvic signs during non-menstrual days, the changes of hormones [estradiol (E2), FSH, LH], and adverse events were observed. RESULTS: After the treatment, the rate of changes of ovarian mass volume (among them, at 12 weeks after the first injection, the median was -55.83% in the test group, -68.22% in the control group, P = 0.336), the distinct improvement rate of symptom scores and pelvic signs during non-menstrual days [among them, at 12 weeks after the first injection, the rate of lower abdomen pain was 47.5% (48/101) in the test group, 44.0% (44/100) in the control group, P = 0.881], the hormone (E2, FSH, LH) levels [among them, at 12 weeks after the first injection, the serum level of E2, was (33±38) pmol/L in the test group, (38±40) pmol/L in the control group, P = 0.414; the serum level of FSH, was (5.1±2.8) U/L in the test group, (5.3±2.3) U/L in the control group, P = 0.666; the serum level of LH, was (0.6±0.8) U/L in the test group, (0.6±0.9) U/L in the control group, P = 0.907], had no statistically significant difference between the two groups (all P > 0.05). The distinct improvement rate and improvement rate of symptom (lower abdomen pain, low back pain) scores during menstruation days at 12 weeks after the first injection, the rates of lower abdomen pain were 73.9% (34/46), 15.2% (7/46) respectively in the test group, 72.3% (34/47), 2.1% (1/47) respectively in the control group, had statistically significant difference between the two groups (P = 0.026). There was no serious adverse event occurred in both two groups. The incidence rate of adverse event was 33.6% (38/113) in test group, 23.2% (27/116) in control group, there was no significant difference between the two groups (P = 0.082). CONCLUSION: Leuprolide acetate is effective and safe in the treatment of endometriosis.

Fusion of multi-source relationships and topology to infer lncRNA-protein interactions.

Long non-coding RNAs (lncRNAs) are important factors involved in biological regulatory networks. Accurately predicting lncRNA-protein interactions (LPIs) is vital for clarifying lncRNA's functions and pathogenic mechanisms. Existing deep learning models have yet to yield satisfactory results in LPI prediction. Recently, graph autoencoders (GAEs) have seen rapid development, excelling in tasks like link prediction and node classification. We employed GAE technology for LPI prediction, devising the FMSRT-LPI model based on path masking and degree regression strategies and thereby achieving satisfactory outcomes. This represents the first known integration of path masking and degree regression strategies into the GAE framework for potential LPI inference. The effectiveness of our FMSRT-LPI model primarily relies on four key aspects. First, within the GAE framework, our model integrates multi-source relationships of lncRNAs and proteins with LPN's topological data. Second, the implemented masking strategy efficiently identifies LPN's key paths, reconstructs the network, and reduces the impact of redundant or incorrect data. Third, the integrated degree decoder balances degree and structural information, enhancing node representation. Fourth, the PolyLoss function we introduced is more appropriate for LPI prediction tasks. The results on multiple public datasets further demonstrate our model's potential in LPI prediction.

GAM-MDR: probing miRNA-drug resistance using a graph autoencoder based on random path masking.

MicroRNAs (miRNAs) are found ubiquitously in biological cells and play a pivotal role in regulating the expression of numerous target genes. Therapies centered around miRNAs are emerging as a promising strategy for disease treatment, aiming to intervene in disease progression by modulating abnormal miRNA expressions. The accurate prediction of miRNA-drug resistance (MDR) is crucial for the success of miRNA therapies. Computational models based on deep learning have demonstrated exceptional performance in predicting potential MDRs. However, their effectiveness can be compromised by errors in the data acquisition process, leading to inaccurate node representations. To address this challenge, we introduce the GAM-MDR model, which combines the graph autoencoder (GAE) with random path masking techniques to precisely predict potential MDRs. The reliability and effectiveness of the GAM-MDR model are mainly reflected in two aspects. Firstly, it efficiently extracts the representations of miRNA and drug nodes in the miRNA-drug network. Secondly, our designed random path masking strategy efficiently reconstructs critical paths in the network, thereby reducing the adverse impact of noisy data. To our knowledge, this is the first time that a random path masking strategy has been integrated into a GAE to infer MDRs. Our method was subjected to multiple validations on public datasets and yielded promising results. We are optimistic that our model could offer valuable insights for miRNA therapeutic strategies and deepen the understanding of the regulatory mechanisms of miRNAs. Our data and code are publicly available at GitHub:https://github.com/ZZCrazy00/GAM-MDR.

Assessing the Diagnostic Efficacy of Handgrip Dynamometry and Diaphragmatic Ultrasound in Intensive Care Unit-Acquired Weakness.

Objective: The aim of this study is to examine the diagnostic significance of using handgrip dynamometry and diaphragmatic ultrasound in intensive care unit-acquired weakness (ICU-AW). Methods: This study included patients who received mechanical ventilation in the ICU at the Fourth Hospital of Hebei Medical University from July to December 2020. We collected comprehensive demographic data and selected conscious patients for muscle strength and ICU-AW assessments. The evaluation comprised grip strength measurement and bedside ultrasound for diaphragmatic excursion (DE) and thickening fraction (DTF). Results were documented for comparative analysis between patient groups, focusing on the diagnostic efficacy of grip strength, DE, DTF, and their combined application in diagnosing ICU-AW. Results: A total of 95 patients were initially considered for inclusion in this study. Following the exclusion of 20 patients, a final cohort of 75 patients were enrolled, comprising of 32 patients (42.6%) diagnosed with ICU-AW and 43 patients (57.4%) classified as non-ICU-AW. Comparative analysis revealed that grip strength, DE, and DTF were significantly lower in the ICU-AW group (P < 0.05). Subgroup analysis specific to male patients demonstrated a noteworthy decrease in grip strength, DE, and DTF within the ICU-AW group (P < 0.05). Receiver operating characteristic curve analysis indicated statistically significant diagnostic value for ICU-AW with grip strength, DE, DTF, and grip strength and diaphragmatic ultrasound (P < 0.01). Furthermore, it was observed that the amalgamation of grip strength and diaphragmatic ultrasound significantly enhanced the diagnostic accuracy of ICU-AW in patients who are critically ill. Conclusion: Grip strength, DE, DTF, and the combined use of grip strength with diaphragm ultrasound demonstrated diagnostic efficacy in ICU-AW. Notably, the integration of grip strength with diaphragm ultrasound exhibited a heightened capacity to enhance the diagnostic value specifically in patients diagnosed who are critically ill with ICU-AW.

BiFTransNet: A unified and simultaneous segmentation network for gastrointestinal images of CT & MRI.

Gastrointestinal (GI) cancer is a malignancy affecting the digestive organs. During radiation therapy, the radiation oncologist must precisely aim the X-ray beam at the tumor while avoiding unaffected areas of the stomach and intestines. Consequently, accurate, automated GI image segmentation is urgently needed in clinical practice. While the fully convolutional network (FCN) and U-Net framework have shown impressive results in medical image segmentation, their ability to model long-range dependencies is constrained by the convolutional kernel's restricted receptive field. The transformer has a robust capacity for global modeling owing to its inherent global self-attention mechanism. The TransUnet model leverages the strengths of both the convolutional neural network (CNN) and transformer models through a hybrid CNN-transformer encoder. However, the concatenation of high- and low-level features in the decoder is ineffective in fusing global and local information. To overcome this limitation, we propose an innovative transformer-based medical image segmentation architecture called BiFTransNet, which introduces a BiFusion module into the decoder stage, enabling effective global and local feature fusion by enabling feature integration from various modules. Further, a multilevel loss (ML) strategy is introduced to oversee the learning process of each decoder layer and optimize the use of globally and locally fused contextual features at different scales. Our method achieved a Dice score of 89.51% and an intersection-over-union (IoU) score of 86.54% on the UW-Madison Gastrointestinal Segmentation dataset. Moreover, our method attained a Dice score of 78.77% and a Hausdorff distance (HD) of 27.94% on the Synapse Multi-organ Segmentation dataset. Compared with the state-of-the-art methods, our proposed method achieves superior segmentation performance in gastrointestinal segmentation tasks. More significantly, our method can be easily extended to medical segmentation in different modalities such as CT and MRI. Our method achieves clinical multimodal medical segmentation and provides decision supports for clinical radiotherapy plans.

Self-supervised deep learning for joint 3D low-dose PET/CT image denoising.

Deep learning (DL)-based denoising of low-dose positron emission tomography (LDPET) and low-dose computed tomography (LDCT) has been widely explored. However, previous methods have focused only on single modality denoising, neglecting the possibility of simultaneously denoising LDPET and LDCT using only one neural network, i.e., joint LDPET/LDCT denoising. Moreover, DL-based denoising methods generally require plenty of well-aligned LD-normal-dose (LD-ND) sample pairs, which can be difficult to obtain. To this end, we propose a self-supervised two-stage training framework named MAsk-then-Cycle (MAC), to achieve self-supervised joint LDPET/LDCT denoising. The first stage of MAC is masked autoencoder (MAE)-based pre-training and the second stage is self-supervised denoising training. Specifically, we propose a self-supervised denoising strategy named cycle self-recombination (CSR), which enables denoising without well-aligned sample pairs. Unlike other methods that treat noise as a homogeneous whole, CSR disentangles noise into signal-dependent and independent noises. This is more in line with the actual imaging process and allows for flexible recombination of noises and signals to generate new samples. These new samples contain implicit constraints that can improve the network's denoising ability. Based on these constraints, we design multiple loss functions to enable self-supervised training. Then we design a CSR-based denoising network to achieve joint 3D LDPET/LDCT denoising. Existing self-supervised methods generally lack pixel-level constraints on networks, which can easily lead to additional artifacts. Before denoising training, we perform MAE-based pre-training to indirectly impose pixel-level constraints on networks. Experiments on an LDPET/LDCT dataset demonstrate its superiority over existing methods. Our method is the first self-supervised joint LDPET/LDCT denoising method. It does not require any prior assumptions and is therefore more robust.