Deep Process Research Articles

DNA Binding Protein Prediction based on Multi-feature Deep Metatransfer Learning

Background: In recent years, the rapid development of deep learning technology has had a significant impact on the prediction of DNA-binding proteins. Deep neural networks can automatically learn complex features in protein and DNA sequences, improving prediction accuracy and generalization capabilities. Objective: This article mainly establishes a meta-migration model and combines it with a deep learning model to predict DNA-binding proteins. Methods: This study introduces a meta-learning algorithm based on transfer learning, which helps achieve rapid learning and adaptation to new tasks. In addition, normalized Moreau-Broto autocorrelation attributes (NMBAC), position-specific scoring matrix-discrete cosine transform (PSSMDCT), and position-specific scoring matrix-discrete wavelet transform (PSSM-DWT) are also used for feature extraction. Finally, the prediction of DBP is achieved through the deep neural network model based on the attention mechanism. Results: This paper first establishes the basis of deep meta-transfer learning and uses the PDB186 data set as the benchmark to extract features using NMBAC, PSSM-DCT, and PSSM-DWT, respectively, and compare the fused features in pairs, and finally obtain the fused feature process. Through deep learning processing, it is concluded that the fused feature prediction effect is the best. At the same time, compared with the currently popular models, there are obvious improvements in the ACC, MCC, SN and Spec evaluation indicators. Conclusion: Finally, it was concluded that the method used in this article can effectively predict DNA-binding proteins and show more significant performance.

Application of deep learning-based speech signal processing technology in electronic communication

In recent years, the artificial intelligence boom triggered by deep learning is influencing and changing peoples lifestyles. People are no longer satisfied with human-computer interaction through simple text commands; instead, they look forward to more convenient and faster communication methods like voice interaction. Against the backdrop of innovative development, the application of speech signal processing systems is becoming increasingly widespread. Therefore, it is necessary to study the application of deep learning-based speech signal processing technology in electronic communication. This can provide more valuable references and assistance for future development, promoting the better development of deep learning-based speech signal processing technology in electronic communication. In this paper, we first review the application of deep learning in speech signal enhancement, speech recognition, and speech synthesis from a theoretical analysis perspective. Then, we discuss the application of deep learning-based speech signal processing in electronic communication, including the application of models such as Transformer, LAS (Listen, Attend and Spell), and GFT-conformer in speech signal processing. We also discuss some application scenarios of deep learning-based speech signal processing in electronic communication. Finally, we identify the need for deeper application of deep learning technology in speech signal processing and electronic communication, with continuous optimization and adjustment.

Anomalously fertile subcontinental lithospheric mantle beneath the intracontinental Canning Basin, Western Australia

SUMMARY Many intracontinental basins form as broad depressions through prolonged, slow subsidence of the continental lithosphere. Such long-lived basins can record lithospheric processes over hundreds of millions of years, serving as important archives of lithospheric evolution. Since continental amalgamation in the Mesoproterozoic, the lithosphere beneath the intracontinental Canning Basin has been subject to several tectonic events, with extensive crustal reworking evidenced through different upper crust data sets. However, knowledge of the structure of the subcontinental lithospheric mantle is lacking. As a consequence, understanding the coupled evolution between surface and deep lithospheric processes, crucial to resolving basin formation, development and survival, remains problematic. Here, we combine geochemical, geophysical and petrophysical data within a thermodynamic modelling framework to determine the thermochemical properties, rheology, density and seismic structure of the lithospheric and sublithospheric mantle beneath the Canning Basin. The results indicate a thick, rigid lithosphere with a maximum thickness of 185 km and strength of ca. 1 × 1013 Pa m, and an anomalously Fe-enriched subcontinental lithospheric mantle with a Mg# of 88.6. This mantle structure is not consistent with pre-collisional fragments or a Precambrian collisional setting and may reflect magmatic refertilization during high-volume mafic magmatic events. Potential candidate events are the ∼1070 Ma Warakurna, ∼825 Ma Gairdner and ∼510 Ma Kalkarindji Large Igneous Provinces. The youngest of these is temporally and spatially correlated with and therefore interpreted to have influenced the Canning Basin formation. We propose that refertilization caused a negatively buoyant subcontinental lithospheric mantle and prolonged subsidence and preservation of the basin, while the strong lithosphere ensured lithospheric stability and longevity.

Four Transformer-Based Deep Learning Classifiers Embedded with an Attention U-Net-Based Lung Segmenter and Layer-Wise Relevance Propagation-Based Heatmaps for COVID-19 X-ray Scans.

Background: Diagnosing lung diseases accurately is crucial for proper treatment. Convolutional neural networks (CNNs) have advanced medical image processing, but challenges remain in their accurate explainability and reliability. This study combines U-Net with attention and Vision Transformers (ViTs) to enhance lung disease segmentation and classification. We hypothesize that Attention U-Net will enhance segmentation accuracy and that ViTs will improve classification performance. The explainability methodologies will shed light on model decision-making processes, aiding in clinical acceptance. Methodology: A comparative approach was used to evaluate deep learning models for segmenting and classifying lung illnesses using chest X-rays. The Attention U-Net model is used for segmentation, and architectures consisting of four CNNs and four ViTs were investigated for classification. Methods like Gradient-weighted Class Activation Mapping plus plus (Grad-CAM++) and Layer-wise Relevance Propagation (LRP) provide explainability by identifying crucial areas influencing model decisions. Results: The results support the conclusion that ViTs are outstanding in identifying lung disorders. Attention U-Net obtained a Dice Coefficient of 98.54% and a Jaccard Index of 97.12%. ViTs outperformed CNNs in classification tasks by 9.26%, reaching an accuracy of 98.52% with MobileViT. An 8.3% increase in accuracy was seen while moving from raw data classification to segmented image classification. Techniques like Grad-CAM++ and LRP provided insights into the decision-making processes of the models. Conclusions: This study highlights the benefits of integrating Attention U-Net and ViTs for analyzing lung diseases, demonstrating their importance in clinical settings. Emphasizing explainability clarifies deep learning processes, enhancing confidence in AI solutions and perhaps enhancing clinical acceptance for improved healthcare results.

A novel framework inspired by human behavior for peg-in-hole assembly

PurposeThe purpose of this paper is to enhance the performance of robots in peg-in-hole assembly tasks, enabling them to swiftly and robustly accomplish the task. It also focuses on the robot’s ability to generalize across assemblies with different hole sizes.Design/methodology/approachHuman behavior in peg-in-hole assembly serves as inspiration, where individuals visually locate the hole firstly and then continuously adjust the peg pose based on force/torque feedback during the insertion process. This paper proposes a novel framework that integrate visual servo and adjustment based on force/torque feedback, the authors use deep neural network (DNN) and image processing techniques to determine the pose of hole, then an incremental learning approach based on a broad learning system (BLS) is used to simulate human learning ability, the number of adjustments required for insertion process is continuously reduced.FindingsThe author conducted experiments on visual servo, adjustment based on force/torque feedback, and the proposed framework. Visual servo inferred the pixel position and orientation of the target hole in only about 0.12 s, and the robot achieved peg insertion with 1–3 adjustments based on force/torque feedback. The success rate for peg-in-hole assembly using the proposed framework was 100%. These results proved the effectiveness of the proposed framework.Originality/valueThis paper proposes a framework for peg-in-hole assembly that combines visual servo and adjustment based on force/torque feedback. The assembly tasks are accomplished using DNN, image processing and BLS. To the best of the authors’ knowledge, no similar methods were found in other people’s work. Therefore, the authors believe that this work is original.

MFTP-Tool: A Wide & Deep Learning Framework for Multi-Functional Therapeutic Peptides Prediction

Background: The identification and functional prediction of Multifunctional Therapeutic Peptides (MFTP) play a pivotal role in drug discovery, particularly for conditions such as inflammation and hyperglycemia. Current computational methods exhibit limitations in their ability to accurately predict the multifunctionality of these peptides. Methods: We propose a novel Wide and Deep Learning Framework that integrates both deep learning and machine learning approaches. The deep learning segment processes word vectors using a neural network model, while the wide segment utilizes the physicochemical properties of peptides in a random forest-based model. This hybrid approach aims to enhance the accuracy of MFTP function prediction. Results: Our framework outperformed the existing PrMFTP predictor in terms of precision, coverage, accuracy, and absolute true values. The evaluation was conducted on both training and independent testing datasets, demonstrating the robustness and generalizability of our model. Conclusion: The proposed Wide & Deep Learning Framework offers a significant advancement in the computational prediction of MFTP functions. The availability of our model through a userfriendly web interface at MFTP-Tool.m6aminer.cn provides a valuable tool for researchers in the field of therapeutic peptide-based drug discovery, potentially accelerating the development of new treatments.

Semantic-to-autobiographical memory priming: the role of stimulus processing.

It is now well established that semantic processing can cause the activation of memories in the autobiographical memory system. Studies have shown that semantic processing of words, sounds, objects, or pictures primes autobiographical memories on voluntary and involuntary autobiographical memory tasks (the Crovitz cue-word task and the vigilance task). Known as semantic-to-autobiographical memory priming, our goal in the current study was to demonstrate that this form of priming occurs under different forms of processing (i.e., shallow versus deep), and that some forms of processing (e.g., visual mental imagery) may enhance priming in this domain. In Experiment 1, equivalent semantic-to-autobiographical priming was obtained on the vigilance task following shallow (e-counting) and deep (meaning judgements) word processing. In Experiment 2, word meaning judgements were compared to visual imagery of word meanings, and visual imagery led to more semantic-to-autobiographical priming on the vigilance task than meaning judgements. The results of these experiments support the idea that semantic-to-autobiographical priming occurs under a wide range of processing conditions, supporting a ubiquity claim, with some conditions producing more priming than others, and they further support the idea that this form of may play an important role in the production of involuntary memories in everyday life.

The Role of Self-directed Learning in Promoting Deep Learning Processes: A Systematic Literature Review

In current educational research, deep learning is widely considered a key approach to fostering the development of students’ comprehensive abilities. However, effectively promoting deep learning, especially across various educational settings, remains a challenge. Especially in the context of Self-Directed Learning (SDL) environments, current research does not specifically detail how SDL facilitates the deep learning process or how students experience and achieve deep learning within SDL environments. Addressing this research gap, this study explores the promotional effect of SDL on deep learning through a systematic literature review and analyzes how self-directed learning strategies and environments support the deep learning process of students. This article, based on Self-Determination Theory (SDT) and meta-cognitive theory, delves into the intrinsic factors and external conditions under SDL environments that promote deep learning. The research found that self-directed learning significantly promotes deep learning by fostering students’ active participation, self-management, and the development of metacognitive skills. Students’ interests and intrinsic motivation, along with reflective and evaluative activities, play a crucial role in deepening the understanding of knowledge. The effective use of technology provides the necessary support for self-directed learning, further facilitating students’ deep learning. Despite the close relationship between SDL and deep learning, effectively integrating these two modes of learning within today’s educational environment remains a challenge. The future requires educators to continue exploring innovative teaching methods and learning environments to promote the effective integration of SDL and deep learning, thereby improving the quality of education to meet future challenges.

Formation and hydrothermal alteration of a volcanic center: Melt pooling and mass transfers at Langseth Ridge (Gakkel Ridge, Arctic Ocean)

Volcanic centers are characteristic features of ultraslow-spreading mid-ocean ridges, the least-explored parts of the global ridge system. Volcanic centers can provide insights into deep magmatic and metamorphic processes at these ridges. Here, we present data from the largest volcanic center on the Gakkel Ridge, the Langseth Ridge, situated at 60–62°E. Langseth is ∼10 km wide, consisting of three peaks that rise to 585 m water depth, some 3–4 km above the surrounding seafloor. It strikes perpendicular to Gakkel's spreading direction and can be traced for ∼40 km, which translates to an age of ∼8 Myr. Seafloor imaging revealed abundant (pillow) basalt but also fissures and geologic faults across the Langseth Ridge. Basaltic rocks were sampled at all summits and diabase at the slope of the northern summit that dips into the rift valley.Our samples are of normal to depleted mid-ocean ridge basalt composition and exhibit a wide range of major and trace element contents, due to magmatic processes, accumulation of macrocrysts, and hydrothermal alteration. Radiogenic isotope ratios, most notably 143Nd/144Nd and 208Pb/206Pb, trend from typical rift valley compositions to isotopically enriched values with increasing distance to the rift valley. This trend may imply melt pooling from different sources, potentially representing a shift from shallow melting beneath the rift valley to deeper melting of enriched sources and higher degrees of melting underneath Langseth. Mineral compositions and plagioclase sieve textures imply prolonged storage of magma at depth prior to eruption. Hydrothermal alteration occurred over a range of conditions. Basalt from the summits is weakly altered at temperatures ≪100 °C, which likely occurred in situ at the summit sites. Diabase samples experienced chloritization and albitization and display epidote and quartz veins, which formed at >300 °C. These assemblages and temperatures are typical for lower crustal levels and imply uplift of the samples of >1 km. Diabase samples from the Afanasenkov Seamount, another volcanic center on the Gakkel Ridge that we investigated for comparison, were altered under comparable conditions.Our findings suggest a combined volcanic–tectonic origin of the studied volcanic centers, potentially implying that such complexes may generally form due to the interplay of magmatism and tectonics. Researching volcanic centers has the potential to further our understanding of both deep and shallow crustal processes at ultraslow-spreading ridges, providing further insights into the role of these centers as linkages between lithosphere and hydrosphere and the (deep) biosphere they sustain.

Singing voice separation based on U-NET neural network and its application

Singing Voice separation is a very important method that helps people to obtain a vocal or accompaniment for further usage like music composing. In this paper, I developed a Song accompaniment separation application based on Convolutional neural network Unet model and Deep learning (DL). Basically, the U-net convolutional network is applied to the two-dimensional spectrum of a music audio in order to achieve a more precise separation effect. During this deep learning process, a significant amount of data was inserted into the model to get the best fit parameters. Then, after all, the model can automatically generate a mask that filters out the accompaniment or vocal part that users want.. This completed model is then transmitted to the terminal of a internet server to enable users to get access to the model through internet. After the test of both quantitative evaluation and subjective assessment, the application impressively achieved the Song accompaniment Separation that provides a good quality of both vocal and song.

Early parafoveal semantic integration in natural reading.

Humans can read and comprehend text rapidly, implying that readers might process multiple words per fixation. However, the extent to which parafoveal words are previewed and integrated into the evolving sentence context remains disputed. We investigated parafoveal processing during natural reading by recording brain activity and eye movements using MEG and an eye tracker while participants silently read one-line sentences. The sentences contained an unpredictable target word that was either congruent or incongruent with the sentence context. To measure parafoveal processing, we flickered the target words at 60 Hz and measured the resulting brain responses (i.e. Rapid Invisible Frequency Tagging, RIFT) during fixations on the pre-target words. Our results revealed a significantly weaker tagging response for target words that were incongruent with the previous context compared to congruent ones, even within 100ms of fixating the word immediately preceding the target. This reduction in the RIFT response was also found to be predictive of individual reading speed. We conclude that semantic information is not only extracted from the parafovea but can also be integrated with the previous context before the word is fixated. This early and extensive parafoveal processing supports the rapid word processing required for natural reading. Our study suggests that theoretical frameworks of natural reading should incorporate the concept of deep parafoveal processing.

Effect of levels-of-processing on rates of forgetting.

The levels-of-processing (LOP) framework, proposing that deep processing yields superior retention, has provided an important paradigm for memory research and a practical means of improving learning. However, the available levels-of-processing literature focuses on immediate memory performance. It is assumed within the LOP framework that deep processing will lead to slower forgetting than will shallow processing. However, it is unclear whether, or how, the initial level of processing affects the forgetting slopes over longer retention intervals. The present three experiments were designed to explore whether items encoded at qualitatively different LOP are forgotten at different rates. In the first two experiments, depth of processing was manipulated within-participants at encoding under deep and shallow conditions (semantic vs. rhyme judgement in Experiment 1; semantic vs. consonant-vowel pattern decision in Experiment 2). Recognition accuracy (d prime) was measured between-participants immediately after learning and at 30-min, 2-h, and 24-h delays. The third experiment employed a between-participants design, contrasting the rates of forgetting following semantic and phonological (rhyme) processing at immediate, 30-min, 2-h, and 6-h delays. Results from the three experiments consistently demonstrated a large effect size of levels of processing on immediate performance and a medium-to-large level effect size on delayed recognition, but crucially no LOP × delay group interaction. Analysis of the retention curves revealed no significant differences between the slopes of forgetting for deep and shallow processing. These results suggest that the rates of forgetting are independent of the qualitatively distinct encoding operations manipulated by levels of processing.

Research Overview on Sugarcane Deep Processing and Comprehensive Utilization in China

Research Overview on Sugarcane Deep Processing and Comprehensive Utilization in China

Deep Learning Enhanced Label-Free Action Potential Detection Using Plasmonic-Based Electrochemical Impedance Microscopy.

Measuring neuronal electrical activity, such as action potential propagation in cells, requires the sensitive detection of the weak electrical signal with high spatial and temporal resolution. None of the existing tools can fulfill this need. Recently, plasmonic-based electrochemical impedance microscopy (P-EIM) was demonstrated for the label-free mapping of the ignition and propagation of action potentials in neuron cells with subcellular resolution. However, limited by the signal-to-noise ratio in the high-speed P-EIM video, action potential mapping was achieved by averaging 90 cycles of signals. Such extensive averaging is not desired and may not always be feasible due to factors such as neuronal desensitization. In this study, we utilized advanced signal processing techniques to detect action potentials in P-EIM extracted signals with fewer averaged cycles. Matched filtering successfully detected action potential signals with as few as averaging five cycles of signals. Long short-term memory (LSTM) recurrent neural network achieved the best performance and was able to detect single-cycle stimulated action potential successfully [satisfactory area under the receiver operating characteristic curve (AUC) equal to 0.855]. Therefore, we show that deep learning-based signal processing can dramatically improve the usability of P-EIM mapping of neuronal electrical signals.

Mesh Neural Networks Based on Dual Graph Pyramids.

Deep neural networks (DNNs) have been widely used for mesh processing in recent years. However, current DNNs can not process arbitrary meshes efficiently. On the one hand, most DNNs expect 2-manifold, watertight meshes, but many meshes, whether manually designed or automatically generated, may have gaps, non-manifold geometry, or other defects. On the other hand, the irregular structure of meshes also brings challenges to building hierarchical structures and aggregating local geometric information, which is critical to conduct DNNs. In this paper, we present DGNet, an efficient, effective and generic deep neural mesh processing network based on dual graph pyramids; it can handle arbitrary meshes. First, we construct dual graph pyramids for meshes to guide feature propagation between hierarchical levels for both downsampling and upsampling. Second, we propose a novel convolution to aggregate local features on the proposed hierarchical graphs. By utilizing both geodesic neighbors and euclidean neighbors, the network enables feature aggregation both within local surface patches and between isolated mesh components. Experimental results demonstrate that DGNet can be applied to both shape analysis and large-scale scene understanding. Furthermore, it achieves superior performance on various benchmarks, including ShapeNetCore, HumanBody, ScanNet and Matterport3D. Code and models will be available at https://github.com/li-xl/DGNet.

Исследование дезинфицирующей способности феррата натрия

More and more attention has been recently paid to the disinfection of industrial wastewater. Food, petrochemical, pharmaceutical, and other industries cannot directly dump wastewater into the environment, therefore, the wastewater is fed to the deep biological treatment lines. A high level of biological threat and the expanding list of potential pathogens cause the need to search for new highly efficient disinfecting agents. The use of ferrates can be a new step in the improvement of the wastewater disinfection process due to the combination of processes of inactivation of pathogenic microflora and the reactant coagulation on ferric hydroxides generated during decontamination (oxidation). Within the framework of the study, the potential use of sodium ferrate as a disinfecting agent obtained by electrochemical dissolution of iron electrodes in alkali solutions has been confirmed. It has been established that sodium ferrate can effectively inactivate conditionally pathogenic cultures A. niger, B. subtillis which are part of activated sludge symbiosis. The efficient dosing of ferrate for complete inhibition of B. subtillis is 0.5 % mass, whereas for A. Niger it is 15 % mass. It has been proved that microflora inhibition occurs directly under the action of sodium ferrate and not due to the alkali reaction of the medium. Industrial tests of the disinfecting ability of ferrates in relation to wasted activated sludge of industrial wastewater deep biological treatment systems have confirmed the results obtained for individual objects (A. niger, B. subtillis). Based on the results, a conclusion about the potential use of ferrates in the deep biological treatment processes for highly bacteriologically contaminated industrial wastewater has been made. The main advantages of the suggested agent include the low toxicity of metabolites, minimum secondary contamination of treated water, and possible reactant advanced treatment (dephosphorization).

Design and implementation of deep learning-based object detection and tracking system

Many human tracking methods by deep learning rely on powerful computing resources. For embedded platforms with limited resources, efficient use of resources is a priority. In this paper, we design an object detection and tracking system based on deep learning methods. We propose an efficient system with software and hardware design. We apply the framework of Vitis AI and its Deep Learning Processing Unit using a hardware/software co-design approach. This approach capitalizes on a higher-level acceleration design framework, where the convolutional models can be updated more flexibly and rapidly. This design approach not only provides a fast design flow but also has good performance in terms of throughput. We facilitate the design and accelerate the object detection model YOLO v3 to achieve higher throughput and energy efficiency. Our tracking method achieves a 1.27x improvement in processing speed with the addition of a single-object tracker. Our proposed human tracking methods can achieve better performance than the others in precision with the same test sequences.

Vulnerability to semantic and phonological interference in normal aging and amnestic mild cognitive impairment (aMCI).

To determine whether the increased vulnerability to semantic interference previously observed in amnestic mild cognitive impairment (aMCI) is specifically associated with semantic material or if it also affects other types of material, suggesting generalized executive and inhibitory impairment. Seventy-two participants divided into two groups (33 aMCI, 39 normal control [NC]) matched for age and education were included. They completed a comprehensive neuropsychological examination, the French version of the Loewenstein Acevedo Scale for Semantic Interference and Learning (LASSI-L; semantic interference test), and a homologous experimental phonological test, the phonological interference and learning test. Independent sample t tests, mixed analysis of variance (ANOVA), and analysis of covariance (ANCOVA) on memory and interference scores were conducted to compare memory and interference in both conditions for both groups. For memory scores, results revealed significant main effects of group (NC > aMCI) and condition (semantic > phonological) and significant interactions (poorer performance in the semantic condition for aMCI). aMCI committed more phonological false recognition errors, were disproportionately more vulnerable to retroactive semantic interference, and showed a higher percentage of intrusion errors associated with proactive semantic interference than NC. To our knowledge, this is the first study to compare vulnerability to interference in aMCI and normal aging with two similarly designed semantic and phonological word list learning tasks. Taken together, our results suggest that aMCI present with broad difficulties in source memory and inhibition, but that impaired deep semantic processing results in additional semantic intrusion errors during proactive interference and impacts their ability to show good recall after an interference list (greater semantic retroactive interference). Results are discussed according to the level-of-processing and activation/monitoring theories. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Probabilistic Back Analysis Based on Nadam, Bayesian, and Matrix-Variate Deep Gaussian Process for Rock Tunnels

Probabilistic Back Analysis Based on Nadam, Bayesian, and Matrix-Variate Deep Gaussian Process for Rock Tunnels

Phylogeography of mammals in Southeast Alaska and implications for management of the Tongass National Forest

AbstractInsular evolution on archipelagos generates a significant proportion of global biodiversity, yet islands are among the ecosystems most sensitive to accelerating anthropogenic disturbance, introductions of non‐native species, and emerging pathogens, among other conservation challenges. The Alexander and Haida Gwaii archipelagos along North America's North Pacific Coast support a disproportionate number of endemic taxa compared to other high‐latitude terrestrial ecosystems. In this region, endemics in Canada are explicitly protected, but in the United States, endemics have been operationally ignored. We reviewed regional research on terrestrial mammals and endemics from 2000–2022 to guide wildlife management. Elevated regional endemism is due to a combination of deep and shallow temporal processes (i.e., long‐term refugial isolation vs. recent colonization). With adequate sampling, genomic analyses are well‐suited to identifying nuanced patterns of divergence and endemism, thereby facilitating a deeper understanding of regional diversity. We identified 18 mammalian endemics in Southeast Alaska, USA, at varying taxonomic scales, but research effort has significant taxonomic biases and sampling infrastructure remains inadequate. Of the 66 terrestrial and aquatic mammal species in Southeast Alaska, only 55% are represented by ≥10 archived samples over the last 2 decades. Across taxa, major spatial and temporal sampling gaps limit interpretations of wildlife responses to changing environmental conditions. The Tongass National Forest is spread across an island archipelago, and climate change is projected to have disproportionate impacts on island endemics worldwide. In this case, the United States Forest Service is not closely monitoring endemic taxa, as was required by the Tongass Land Management Plan in 1997. Our review underscores a need for increased consideration of how endemism can be incorporated into land and wildlife management across the Alexander Archipelago. Moving forward, we encourage state and federal agencies, Indigenous communities, and international collaborators to continue to partner with natural history biorepositories to ensure strategic wildlife sampling infrastructure is built and made accessible to the broader scientific community as part of the land management process.