CRISPR-based editing of the ω- and γ-gliadin gene clusters reduces wheat immunoreactivity without affecting grain protein quality.

Wheat immunotoxicity is associated with abnormal reaction to gluten-derived peptides. Attempts to reduce immunotoxicity using breeding and biotechnology often affect dough quality. Here, the multiplexed CRISPR-Cas9 editing of cultivar Fielder was used to modify gluten-encoding genes, specifically focusing on ω- and γ-gliadin gene copies, which were identified to be abundant in immunoreactive peptides based on the analysis of wheat genomes assembled using the long-read sequencing technologies. The whole-genome sequencing of an edited line showed mutation or deletion of nearly all ω-gliadin and half of the γ-gliadin gene copies and confirmed the lack of editing in the α/ß-gliadin genes. The estimated 75% and 64% reduction in ω- and γ-gliadin content, respectively, had no negative impact on the end-use quality characteristics of grain protein and dough. A 47-fold immunoreactivity reduction compared to a non-edited line was demonstrated using antibodies against immunotoxic peptides. Our results indicate that the targeted CRISPR-based modification of the ω- and γ-gliadin gene copies determined to be abundant in immunoreactive peptides by analysing high-quality genome assemblies is an effective mean for reducing immunotoxicity of wheat cultivars while minimizing the impact of editing on protein quality.

Multiplexed promoter and gene editing in wheat using a virus-based guide RNA delivery system.

The low efficiency of genetic transformation and gene editing across diverse cultivars hinder the broad application of CRISPR technology for crop improvement. The development of virus-based methods of CRISPR-Cas system delivery into the plant cells holds great promise to overcome these limitations. Here, we perform direct inoculation of wheat leaves with the barley stripe mosaic virus (BSMV) transcripts to deliver guide RNAs (sgRNA) into the Cas9-expressing wheat. We demonstrate that wheat inoculation with the pool of BSMV-sgRNAs could be used to generate heritable precise deletions in the promoter region of a transcription factor and to perform multiplexed editing of agronomic genes. We transfer the high-expressing locus of Cas9 into adapted spring and winter cultivars by marker-assisted introgression and use of the BSMV-sgRNAs to edit two agronomic genes. A strategy presented in our study could be applied to any adapted cultivar for creating new cis-regulatory diversity or large-scale editing of multiple genes in biological pathways or QTL regions, opening possibilities for the effective engineering of crop genomes, and accelerating gene discovery and trait improvement efforts.

A High-Throughput Nanofluidic Device for Exosome Nanoporation to Develop Cargo Delivery Vehicles.

Efficient loading of various exogenous cargos into exosomes while not affecting their integrity and functionalities remains a major challenge. Here, a nanofluidic device named "exosome nanoporator (ENP)" is presented for high-throughput loading of various cargos into exosomes. By transporting exosomes through nanochannels with height comparable to their dimension, exosome membranes are permeabilized by mechanical compression and fluid shear, allowing the influx of cargo molecules into the exosomes from the surrounding solution while maintaining exosome integrity. The ENP consisting of an array of 30 000 nanochannels demonstrates a high sample throughput, and the working mechanism of the device is elucidated through experimental and numerical study. Further, the exosomes treated by the ENP can deliver their drug cargos to human non-small cell lung cancer cells and induce cell death, indicating the potential opportunities of the device for developing new exosome-based delivery vehicles for medical and biological applications.

New insights into the evolution of wheat avenin-like proteins in wild emmer wheat (Triticum dicoccoides).

Fifteen full-length wheat grain avenin-like protein coding genes (TaALP) were identified on chromosome arms 7AS, 4AL, and 7DS of bread wheat with each containing five genes. Besides the a- and b-type ALPs, a c type was identified in the current paper. Both a and b types have two subunits, named x and y types. The five genes on each of the three chromosome arms consisted of two x-type genes, two y-type genes, and one c-type gene. The a-type genes were typically of 520 bp in length, whereas the b types were of 850 bp in length, and the c type was of 470 bp in length. The ALP gene transcript levels were significantly up-regulated in Blumeria graminis f. sp. tritici (Bgt)-infected wheat grain caryopsis at early grain filling. Wild emmer wheat [(WEW), Triticum dicoccoides] populations were focused on in our paper to identify allelic variations of ALP genes and to study the influence of natural selection on certain alleles. Consequently, 25 alleles were identified for TdALP-bx-7AS, 13 alleles were identified for TdALP-ax-7AS, 7 alleles were identified for TdALP-ay-7AS, and 4 alleles were identified for TdALP-ax-4AL Correlation studies on TdALP gene diversity and ecological stresses suggested that environmental factors contribute to the ALP polymorphism formation in WEW. Many allelic variants of ALPs in the endosperm of WEW are not present in bread wheat and therefore could be utilized in breeding bread wheat varieties for better quality and elite plant defense characteristics.

Characterising avenin-like proteins (ALPs) from albumin/globulin fraction of wheat grains by RP-HPLC, SDS-PAGE, and MS/MS peptides sequencing.

BACKGROUND: Wheat grain avenin-like proteins (ALPs) belong to a recently discovered class of wheat grain storage protein. ALPs in wheat grains not only have beneficial effects on dough quality but also display antifungal activities, which is a novel observation for wheat storage proteins. Previous studies have shown that ALPs are likely present in the albumin/globulin fractions of total protein extract from wheat flour. However, the accumulation characteristics of these ALPs in the mature wheat grain remains unknown. RESULTS: In the present study, a total of 13 ALPs homologs were isolated and characterized in the albumin/globulin fractions of the wheat protein extract. A combination of multiple techniques including RP-HPLC, SDS-PAGE, MALDI-TOF and peptide sequencing were used for accurate separation and identification of individual ALP homolog. The C-terminal TaALP-by-4AL/7DS, TaALP-by-4AL/7AS/7DS, TaALP-bx/4AL/7AS/7DS, TaALP-ay-7DS, TaALP-ay-4AL, TaALP-ax-4AL, TaALP-ax-7AS, and TaALP-ax-7DS, were separated as individual protein bands from wheat flour for the first time. These unique ALPs peptides were mapped to the latest wheat genome assembly in the IWGSC database. The characteristic defence related proteins present in albumin and globulin fractions, such as protein disulfide-isomerase (PDI), grain softness protein (GSP), alpha-amylase inhibitors (AAIs) and endogenous alpha-amylase/subtilisin inhibitor were also found to co-segregate with these identified ALPs, avenin-3 and α-gliadins. The molecular weight range and the electrophoresis segregation properties of ALPs were characterised in comparison with the proteins containing the tryp_alpha_amyl domain (PF00234) and the gliadin domain (PF13016), which play a role in plant immunity and grain quality. We examined the phylogenetic relationships of the AAIs, GSP, avenin-3, α-gliadins and ALPs, based on the alignment of their functional domains. MALDI-TOF profiling indicated the occurrence of certain post-translations modifications (PTMs) in some ALP subunits. CONCLUSIONS: We reported for the first time the complete profiling of ALPs present in the albumin/globulin fractions of wheat grain protein extracts. We concluded that majority of the ALPs homologs are expressed in wheat grains. We found clear evidence of PTMs in several ALPs peptides. The identification of both gliadin domain (PF13016) and Tryp_alpha_amyl domain (PF00234) in the mature forms of ALPs highlighted the multiple functional properties of ALPs in grain quality and disease resistance.

Introgression of an expressed HMW 1Ay glutenin subunit allele into bread wheat cv. Lincoln increases grain protein content and breadmaking quality without yield penalty.

KEY MESSAGE: An expressed HMW glutenin subunit Glu-Ay showed positive impacts on a range of wheat processing quality and yield traits. The grain protein compositions are significantly optimised for baking, resulting in a better breadmaking quality. The unique breadmaking properties of wheat flour are related to the quality and quantity of high-molecular weight glutenin subunits (HMW-GSs) present in the grain. In the current study, the silent 1Ay HMW-GS allele, present in most bread wheat cultivars, was replaced by the expressed 1Ay21* allele, which was introgressed into Australian bread wheat cultivar Lincoln by a backcrossing and selfing scheme. Stability of gene expression and the effect of the introgressed 1Ay21* subunit on protein composition, agronomic traits, flour functionality, and breadmaking quality were studied using BC4F5 grain grown in glasshouse and field. Field phenotyping and grain quality testing showed that the 1Ay21* gene conferred significant improvements to a range of traits, including an increase in grain protein content by up to 9%, UPP% by up to 24%, bread volume by up to 28%. The glasshouse experiment and one of the field trials showed positive 1Ay21* effects on yield, while one field trial showed one significant effects. This indicates that expression of the 1Ay21* gene has the potential of simultaneously increasing protein content and grain yield under certain environment. The qualitative improvements of the grain also led to a reduction of the energy required during the baking process in addition to the significant positive effects on bread quality.

Aptasensor-enabled quantitative analysis of nano-sized extracellular vesicles by flow cytometry.

Extracellular vesicles (EVs) represent an important mode of intercellular communication in both disease and developmental biology, exposing their potential in diagnostics and therapeutics. Recently, aptamer-based sensors, i.e. aptasensors, have been gradually applied in EV analysis due to their high selectivity and sensitivity. A fluorescent aptasensor enables easy readout by flow cytometry (FCM) and has more accuracy and convenience than conventional immunoassays for EV analysis. Here, we develop a fluorescent aptasensor-based method for quantitative analysis of nano-sized membrane vesicles by using high-resolution FCM. EVs as small as 100 nm are detected and quantified using a dual-staining procedure with the fluorescent aptasensor targeting CD63 and a cytoplasmic dye. Nano-sized EVs derived from bone marrow mesenchymal stem cells, human neural stem cells and human cornea epithelial cells are analyzed, and the result shows that their amount varies from 6.79 × 106 mL-1 to 2.08 × 108 mL-1 in culture media. The technique is also used to evaluate the bioactivity of EVs and, in the future, it may develop into a versatile tool to analyze and quantify EVs from a variety of biological objects with conventional cytometric instruments.

Current Progress in Understanding and Recovering the Wheat Genes Lost in Evolution and Domestication.

The modern cultivated wheat has passed a long evolution involving origin of wild emmer (WEM), development of cultivated emmer, formation of spelt wheat and finally establishment of modern bread wheat and durum wheat. During this evolutionary process, rapid alterations and sporadic changes in wheat genome took place, due to hybridization, polyploidization, domestication, and mutation. This has resulted in some modifications and a high level of gene loss. As a result, the modern cultivated wheat does not contain all genes of their progenitors. These lost genes are novel for modern wheat improvement. Exploring wild progenitor for genetic variation of important traits is directly beneficial for wheat breeding. WEM wheat (Triticum dicoccoides) is a great genetic resource with huge diversity for traits. Few genes and quantitative trait loci (QTL) for agronomic, quantitative, biotic and abiotic stress-related traits have already been mapped from WEM. This resource can be utilized for modern wheat improvement by integrating identified genes or QTLs through breeding.

Diurnal Changes in Water Soluble Carbohydrate Components in Leaves and Sucrose Associated TaSUT1 Gene Expression during Grain Development in Wheat.

In plant tissues, sugar levels are determined by the balance between sugar import, export, and sugar synthesis. So far, water soluble carbohydrate (WSC) dynamics have not been investigated in a diurnal context in wheat stems as compared to the dynamics in flag leaves during the terminal phases of grain filling. Here, we filled this research gap and tested the hypothesis that WSC dynamics interlink with gene expression of TaSUT1. The main stems and flag leaves of two genotypes, Westonia and Kauz, were sampled at four hourly intervals over a 24 h period at six developmental stages from heading to 28 DAA (days after anthesis). The total levels of WSC and WSC components were measured, and TaSUT1 gene expression was quantified at 21 DAA. On average, the total WSC and fructan levels in the stems were double those in the flag leaves. In both cultivars, diurnal patterns in the total WSC and sucrose were detected in leaves across all developmental stages, but not for the fructans 6-kestose and bifurcose. However, in stems, diurnal patterns of the total WSC and fructan were only found at anthesis in Kauz. The different levels of WSC and WSC components between Westonia and Kauz are likely associated with leaf chlorophyll levels and fructan degradation, especially 6-kestose degradation. High correlation between levels of TaSUT1 expression and sucrose in leaves indicated that TaSUT1 expression is likely to be influenced by the level of sucrose in leaves, and the combination of high levels of TaSUT1 expression and sucrose in Kauz may contribute to its high grain yield under well-watered conditions.

Wheat grain protein accumulation and polymerization mechanisms driven by nitrogen fertilization.

In wheat (Triticum aestivum) grain yield and grain protein content are negatively correlated, making the simultaneous increase of the two traits challenging. Apart from genetic approaches, modification of nitrogen fertilization offers a feasible option to achieve this aim. In this study, a range of traits related to nitrogen-use efficiency in six Australian bread wheat varieties were investigated under different nitrogen treatments using 3-year multisite field trials. Changes in the individual storage protein composition were detected by high-performance liquid chromatography. Our results indicated that wheat grain yield and grain protein content reacted similarly to nitrogen availability, with grain yield being slightly more sensitive than grain protein content, and that genotype is a vital determinant of grain protein yield. Measurement of the glutamine synthetase activity of flag leaves and developing grains revealed that high nitrogen availability prompted the participation of glutamine in biological processes. In addition, a more significant accumulation of gluten macropolymer was observed under the high-nitrogen treatment from 21 days post-anthesis, and the underlying mechanism was elucidated by a comparative proteomics study. A yeast two-hybrid experiment confirmed this mechanism. The results of this study revealed that peptidyl-prolyl cis-trans isomerase (PPIase) was SUMOylated with the assistance of small ubiquitin-related modifier 1 and that high nitrogen availability facilitated this connection for the subsequent protein polymerization. Additionally, luminal-binding protein 2 in the endoplasmic reticulum played a similar role to PPIase in the aggregation of protein under high-nitrogen conditions.

Allelic variation of LMW-GS composition in Chinese wheat landraces of the Yangtze-River region detected by MALDI-TOF-MS.

Low molecular weight glutenin subunits are important components of wheat storage proteins, which play an important role in determining end-use quality of common wheat. A newly established matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) procedure was used to analyze 478 landraces of bread wheat collected from the Yangtze-River region in China. Results indicated that 17 alleles at three loci: Glu-A3, Glu-B3 and Glu-D3 were identified, resulting in 87 different allele combinations. Of the 17 alleles detected at all the Glu-3 loci, five belonged to Glu-A3, seven to Glu-B3 and five to Glu-D3 locus. MALDI-TOF-MS indicated Glu-A3a/c was present in 72.8%, Glu-A3b in 8.4%, Glu-A3d in 8.4%, Glu-A3f in 5.2% and Glu-A3e in 3.6% lines. Seven types of alleles were identified at the Glu-B3 locus: Glu-B3d/i (25.5%), Glu-B3b (21.3%), Glu-B3c (16.9%), Glu-B3h (13.8%), Glu-B3f (8.4%), Glu-B3a (8.2%), and Glu-B3g (5.2%). Five types of Glu-D3 alleles were detected: Glu-D3a (58.4%), Glu-D3c (22.6%), Glu-D3d (15.5%), Glu-D3b (3.3%) and Glu-D3f (0.2%). Four new alleles that showed abnormal MALDI-TOF spectrum patterns were identified at the Glu-A3 and Glu-B3 loci. A detailed study is needed to further characterize these alleles and their potential usage for wheat improvement.

Rapid separation of seed gliadins by reversed-phase ultra performance liquid chromatography (RP-UPLC) and its application in wheat cultivar and germplasm identification.

To separate gliadin from wheat flour, a novel and stability-indicating reversed-phase ultra performance liquid chromatography (RP-UPLC) method is established and optimized. A comparative analysis of routine capillary electrophoresis (CE), reversed-phase high-performance liquid chromatography (RP-HPLC), and RP-UPLC was performed and the results showed that the resolution and efficiency of RP-UPLC were significantly higher than those of CE and RP-HPLC. Characteristic RP-UPLC patterns of different bread wheat variety and related species were readily identified. These results demonstrated that our RP-UPLC procedure resulted in significant improvements in sensitivity, speed, and resolution, and thus is highly useful in wheat cultivar and germplasm identification.

Face anti-spoofing with cross-stage relation enhancement and spoof material perception.

Face Anti-Spoofing (FAS) seeks to protect face recognition systems from spoofing attacks, which is applied extensively in scenarios such as access control, electronic payment, and security surveillance systems. Face anti-spoofing requires the integration of local details and global semantic information. Existing CNN-based methods rely on small stride or image patch-based feature extraction structures, which struggle to capture spatial and cross-layer feature correlations effectively. Meanwhile, Transformer-based methods have limitations in extracting discriminative detailed features. To address the aforementioned issues, we introduce a multi-stage CNN-Transformer-based framework, which extracts local features through the convolutional layer and long-distance feature relationships via self-attention. Based on this, we proposed a cross-attention multi-stage feature fusion, employing semantically high-stage features to query task-relevant features in low-stage features for further cross-stage feature fusion. To enhance the discrimination of local features for subtle differences, we design pixel-wise material classification supervision and add a auxiliary branch in the intermediate layers of the model. Moreover, to address the limitations of a single acquisition environment and scarcity of acquisition devices in the existing Near-Infrared dataset, we create a large-scale Near-Infrared Face Anti-Spoofing dataset with 380k pictures of 1040 identities. The proposed method could achieve the state-of-the-art in OULU-NPU and our proposed Near-Infrared dataset at just 1.3GFlops and 3.2M parameter numbers, which demonstrate the effective of the proposed method.

Can patient-specific acquisition protocol improve performance on defect detection task in myocardial perfusion SPECT?

Myocardial perfusion imaging using single-photon emission computed tomography (SPECT), or myocardial perfusion SPECT (MPS) is a widely used clinical imaging modality for the diagnosis of coronary artery disease. Current clinical protocols for acquiring and reconstructing MPS images are similar for most patients. However, for patients with outlier anatomical characteristics, such as large breasts, images acquired using conventional protocols are often sub-optimal in quality, leading to degraded diagnostic accuracy. Solutions to improve image quality for these patients outside of increased dose or total acquisition time remain challenging. Thus, there is an important need for new methodologies to improve image quality for such patients. One approach to improving this performance is adapting the image acquisition protocol specific to each patient. For this study, we first designed and implemented a personalized patient-specific protocol-optimization strategy, which we term precision SPECT (PRESPECT). This strategy integrates ideal observer theory with the constraints of tomographic reconstruction to optimize the acquisition time for each projection view, such that MPS defect detection performance is maximized. We performed a clinically realistic simulation study on patients with outlier anatomies on the task of detecting perfusion defects on various realizations of low-dose scans by an anthropomorphic channelized Hotelling observer. Our results show that using PRESPECT led to improved performance on the defect detection task for the considered patients. These results provide evidence that personalization of MPS acquisition protocol has the potential to improve defect detection performance, motivating further research to design optimal patient-specific acquisition and reconstruction protocols for MPS, as well as developing similar approaches for other medical imaging modalities.

DEMIST: A Deep-Learning-Based Detection-Task-Specific Denoising Approach for Myocardial Perfusion SPECT.

There is an important need for methods to process myocardial perfusion imaging (MPI) single-photon emission computed tomography (SPECT) images acquired at lower-radiation dose and/or acquisition time such that the processed images improve observer performance on the clinical task of detecting perfusion defects compared to low-dose images. To address this need, we build upon concepts from model-observer theory and our understanding of the human visual system to propose a detection task-specific deep-learning-based approach for denoising MPI SPECT images (DEMIST). The approach, while performing denoising, is designed to preserve features that influence observer performance on detection tasks. We objectively evaluated DEMIST on the task of detecting perfusion defects using a retrospective study with anonymized clinical data in patients who underwent MPI studies across two scanners (N = 338). The evaluation was performed at low-dose levels of 6.25%, 12.5%, and 25% and using an anthropomorphic channelized Hotelling observer. Performance was quantified using area under the receiver operating characteristics curve (AUC). Images denoised with DEMIST yielded significantly higher AUC compared to corresponding low-dose images and images denoised with a commonly used task-agnostic deep learning-based denoising method. Similar results were observed with stratified analysis based on patient sex and defect type. Additionally, DEMIST improved visual fidelity of the low-dose images as quantified using root mean squared error and structural similarity index metric. A mathematical analysis revealed that DEMIST preserved features that assist in detection tasks while improving the noise properties, resulting in improved observer performance. The results provide strong evidence for further clinical evaluation of DEMIST to denoise low-count images in MPI SPECT.

3D printed VEGF-CPO biomaterial scaffold to promote subcutaneous vascularization and survival of transplanted islets for the treatment of diabetes.

Diabetes is a complex metabolic disease and islet transplantation is a promising approach for the treatment of diabetes. Unfortunately, the transplanted islets at the subcutaneous site are also affected by various adverse factors such as poor vascularization and hypoxia. In this study, we utilize biocompatible copolymers l-lactide and D,l-lactide to manufacture a biomaterial scaffold with a mesh-like structure via 3D printing technology, providing a material foundation for encapsulating pancreatic islet cells. The scaffold maintains the sustained release of vascular endothelial growth factor (VEGF) and a slow release of oxygen from calcium peroxide (CPO), thereby regulating the microenvironment for islet survival. This helps to improve insufficient subcutaneous vascularization and reduce islet death due to hypoxia post-transplantation. By pre-implanting VEGF-CPO scaffolds subcutaneously into diabetic rats, a sufficiently vascularized site is formed, thereby ensuring early survival of transplanted islets. In a word, the VEGF-CPO scaffold shows good biocompatibility both in vitro and in vivo, avoids the adverse effects on the implanted islets, and displays promising clinical transformation prospects.

Case study of financial leasing model driven by fuzzy logic control for alternative fuel vehicles operation.

Over the past decade, the alternative fuel vehicle industry in the world has sprung up with huge speed. For example, the annual output has increased from less than 2000 vehicles to now 3,500,000 vehicles in China. It enjoys more than 50% of the market share worldwide in the global market. A spurt of progress in the alternative fuel vehicle industry has built a foundation for carbon peaking and carbon neutrality goals. Financial leasing has unique advantages which not only can provide guarantees for this industry in many aspects concerning related equipment, systems and infrastructures but also offer financial support for green projects. Nevertheless, financial leasing firms are encountering a string of problems to solve, such as selecting optimal green projects and cooperative businesses, designing transaction structures, and controlling project risks. This study contains several main sections: connecting the incremental alternative fuel vehicle investment and purchase project of a leading regional enterprise; building the structure of the financial leasing project; and analyzing the project leasing property using a fuzzy logic model, the financial structure and the repayment capacity of the project main company so as to comprehensively evaluate the feasibility of the project. This paper aims to provide a reference for future financing of alternative fuel vehicle operation enterprises with a real case study. The case study results show that our introduced fuzzy logic method can obtain the satisfying performance and traffic allocation.

Protocol to develop a microfluidic human corneal barrier-on-a-chip to evaluate the corneal epithelial wound repair process.

Organs-on-chips are microfluidic devices for cell culturing to simulate tissue- or organ-level physiology, providing new solutions other than traditional animal tests. Here, we describe a microfluidic platform consisting of human corneal cells and compartmentalizing channels to achieve fully integrated human cornea's barrier effects on the chip. We detail steps to verify the barrier effects and physiological phenotypes of microengineered human cornea. Then, we use the platform to evaluate the corneal epithelial wound repair process. For complete details on the use and execution of this protocol, please refer to Yu et al. (2022).1.

Deep Learning for Face Anti-Spoofing: A Survey.

Face anti-spoofing (FAS) has lately attracted increasing attention due to its vital role in securing face recognition systems from presentation attacks (PAs). As more and more realistic PAs with novel types spring up, early-stage FAS methods based on handcrafted features become unreliable due to their limited representation capacity. With the emergence of large-scale academic datasets in the recent decade, deep learning based FAS achieves remarkable performance and dominates this area. However, existing reviews in this field mainly focus on the handcrafted features, which are outdated and uninspiring for the progress of FAS community. In this paper, to stimulate future research, we present the first comprehensive review of recent advances in deep learning based FAS. It covers several novel and insightful components: 1) besides supervision with binary label (e.g., '0' for bonafide versus '1' for PAs), we also investigate recent methods with pixel-wise supervision (e.g., pseudo depth map); 2) in addition to traditional intra-dataset evaluation, we collect and analyze the latest methods specially designed for domain generalization and open-set FAS; and 3) besides commercial RGB camera, we summarize the deep learning applications under multi-modal (e.g., depth and infrared) or specialized (e.g., light field and flash) sensors. We conclude this survey by emphasizing current open issues and highlighting potential prospects.

A task-specific deep-learning-based denoising approach for myocardial perfusion SPECT.

Deep-learning (DL)-based methods have shown significant promise in denoising myocardial perfusion SPECT images acquired at low dose. For clinical application of these methods, evaluation on clinical tasks is crucial. Typically, these methods are designed to minimize some fidelity-based criterion between the predicted denoised image and some reference normal-dose image. However, while promising, studies have shown that these methods may have limited impact on the performance of clinical tasks in SPECT. To address this issue, we use concepts from the literature on model observers and our understanding of the human visual system to propose a DL-based denoising approach designed to preserve observer-related information for detection tasks. The proposed method was objectively evaluated on the task of detecting perfusion defect in myocardial perfusion SPECT images using a retrospective study with anonymized clinical data. Our results demonstrate that the proposed method yields improved performance on this detection task compared to using low-dose images. The results show that by preserving task-specific information, DL may provide a mechanism to improve observer performance in low-dose myocardial perfusion SPECT.