-
Bioinformatics (Oxford, England) Jun 2024Electronic health records (EHRs) represent a comprehensive resource of a patient's medical history. EHRs are essential for utilizing advanced technologies such as deep...
MOTIVATION
Electronic health records (EHRs) represent a comprehensive resource of a patient's medical history. EHRs are essential for utilizing advanced technologies such as deep learning (DL), enabling healthcare providers to analyze extensive data, extract valuable insights, and make precise and data-driven clinical decisions. DL methods such as recurrent neural networks (RNN) have been utilized to analyze EHR to model disease progression and predict diagnosis. However, these methods do not address some inherent irregularities in EHR data such as irregular time intervals between clinical visits. Furthermore, most DL models are not interpretable. In this study, we propose two interpretable DL architectures based on RNN, namely time-aware RNN (TA-RNN) and TA-RNN-autoencoder (TA-RNN-AE) to predict patient's clinical outcome in EHR at the next visit and multiple visits ahead, respectively. To mitigate the impact of irregular time intervals, we propose incorporating time embedding of the elapsed times between visits. For interpretability, we propose employing a dual-level attention mechanism that operates between visits and features within each visit.
RESULTS
The results of the experiments conducted on Alzheimer's Disease Neuroimaging Initiative (ADNI) and National Alzheimer's Coordinating Center (NACC) datasets indicated the superior performance of proposed models for predicting Alzheimer's Disease (AD) compared to state-of-the-art and baseline approaches based on F2 and sensitivity. Additionally, TA-RNN showed superior performance on the Medical Information Mart for Intensive Care (MIMIC-III) dataset for mortality prediction. In our ablation study, we observed enhanced predictive performance by incorporating time embedding and attention mechanisms. Finally, investigating attention weights helped identify influential visits and features in predictions.
AVAILABILITY AND IMPLEMENTATION
https://github.com/bozdaglab/TA-RNN.
Topics: Electronic Health Records; Neural Networks, Computer; Humans; Deep Learning; Alzheimer Disease
PubMed: 38940180
DOI: 10.1093/bioinformatics/btae264 -
Frontiers in Bioscience (Landmark... Jun 2024
Topics: Humans; Biomarkers; Dementia; Proteostasis Deficiencies; Protein Folding; Alzheimer Disease
PubMed: 38940055
DOI: 10.31083/j.fbl2906227 -
Frontiers in Bioscience (Landmark... Jun 2024Alzheimer's disease is characterized by extracellular beta-amyloid plaques, intraneuronal tau neurofibrillary tangles and excessive neurodegeneration. The mechanisms of...
BACKGROUND
Alzheimer's disease is characterized by extracellular beta-amyloid plaques, intraneuronal tau neurofibrillary tangles and excessive neurodegeneration. The mechanisms of neuron degeneration and the potential of these neurons to form new nerve fibers for compensation remain elusive. The present study aimed to evaluate the impact of beta-amyloid and tau on new formations of nerve fibers from mouse organotypic brain slices connected to collagen-based microcontact prints.
METHODS
Organotypic brain slices of postnatal day 8-10 wild-type mice were connected to established collagen-based microcontact prints loaded with polyornithine to enhance nerve fiber outgrowth. Human beta-amyloid(42) or P301S mutated aggregated tau was co-loaded to the prints. Nerve fibers were immunohistochemically stained with neurofilament antibodies. The physiological activity of outgrown neurites was tested with neurotracer MiniRuby, voltage-sensitive dye FluoVolt, and calcium-sensitive dye Rhod-4.
RESULTS
Immunohistochemical staining revealed newly formed nerve fibers extending along the prints derived from the brain slices. While collagen-only microcontact prints stimulated nerve fiber growth, those loaded with polyornithine significantly enhanced nerve fiber outgrowth. Beta-amyloid(42) significantly increased the neurofilament-positive nerve fibers, while tau had only a weak effect. MiniRuby crystals, retrogradely transported along these newly formed nerve fibers, reached the hippocampus, while FluoVolt and Rhod-4 monitored electrical activity in newly formed nerve fibers.
CONCLUSIONS
Our data provide evidence that intact nerve fibers can form along collagen-based microcontact prints from mouse brain slices. The Alzheimer's peptide beta-amyloid(42) stimulates this growth, hinting at a neuroprotective function when physiologically active. This "brain-on-chip" model may offer a platform for screening bioactive factors or testing drug effects on nerve fiber growth.
Topics: Animals; Amyloid beta-Peptides; Mice; Nerve Fibers; Brain; tau Proteins; Humans; Immunohistochemistry; Peptide Fragments; Alzheimer Disease; Mice, Inbred C57BL
PubMed: 38940051
DOI: 10.31083/j.fbl2906232 -
Sheng Li Xue Bao : [Acta Physiologica... Jun 2024The purpose of the study was to investigate the mechanism of TFEB activator 1 (TA1) improving the autophagic degradation of oligomeric amyloid-β (oAβ) in microglia,...
The purpose of the study was to investigate the mechanism of TFEB activator 1 (TA1) improving the autophagic degradation of oligomeric amyloid-β (oAβ) in microglia, and to explore the therapeutic effect of TA1 on an in vitro model of microglia in Alzheimer's disease (AD). Primary microglia were exposed to 1 μmol/L oAβ for 0, 3, 12, and 24 h respectively to construct the in vitro model of microglia in AD. In order to explore the therapeutic effect of TA1, primary microglia were co-treated with 1 μmol/L oAβ and 1 μmol/L TA1 for 12 h. To determine the autophagy flux, the above cells were further treated with 100 nmol/L Bafilomycin A1 for 1 h before fixation. Fluorescent probes were used to detect the endocytosis or degradation of oAβ by microglia. The autophagic flux was determined by infection of lentivirus mCherry-EGFP-LC3. The nuclear TFEB intensity, the autophagosomes number, and the colocalization ratio of oAβ with lysosome-associated membrane protein 1 (LAMP1) or microtubule-associated protein light chain 3 (LC3), were detected by immunofluorescence assay. Expressions of autophagy-related-genes, including Lamp1, Atg5, and Map1lc3b, were detected by qRT-PCR. Results showed that prolonged oAβ exposure inhibited the endocytosis and degradation of oAβ by microglia. Meanwhile, the number of autophagosomes and autophagy flux in microglia decreased after 12 h of oAβ treatment. We further found that the nuclear expression of autophagy regulator TFEB decreased after 12 h of oAβ exposure, resulting in the decrease of autophagy genes, thus leading to the damage of autophagic degradation of oAβ. Therefore, long-term oAβ exposure was considered to construct the in vitro model of microglia in AD. After TA1 treatment, the nuclear expression of TFEB in cells was obviously upregulated. TA1 treatment upregulated the expressions of autophagy-related genes, leading to the recovery of autophagy flux. TA1 also recovered the endocytosis and degradation of oAβ by microglia. In conclusion, TA1 could improve oAβ clearance by microglia in AD by upregulating microglial TFEB-mediated autophagy, suggesting TA1 as a potential therapeutic drug for AD.
Topics: Microglia; Amyloid beta-Peptides; Autophagy; Animals; Basic Helix-Loop-Helix Leucine Zipper Transcription Factors; Alzheimer Disease; Cells, Cultured; Mice
PubMed: 38939931
DOI: No ID Found -
Exploration (Beijing, China) Jun 2024The fibrillation of amyloid-β (Aβ) is the critical causal factor in Alzheimer's disease (AD), the dissolution and clearance of which are promising for AD therapy....
The fibrillation of amyloid-β (Aβ) is the critical causal factor in Alzheimer's disease (AD), the dissolution and clearance of which are promising for AD therapy. Although many Aβ inhibitors are developed, their low Aβ-binding affinity results in unsatisfactory effect. To solve this challenge, the Aβ sequence-matching strategy is proposed to tail-design dissociable nanosystem (B6-PNi NPs). Herein, B6-PNi NPs aim to improve Aβ-binding affinity for effective dissolution of amyloid fibrils, as well as to interfere with the in vivo fate of amyloid for Aβ clearance. Results show that B6-PNi NPs decompose into small nanostructures and expose Aβ-binding sites in response to AD microenvironment, and then capture Aβ via multiple interactions, including covalent linkage formed by nucleophilic substitution reaction. Such high Aβ-binding affinity disassembles Aβ fibrils into Aβ monomers, and induces the reassembly of Aβ&nanostructure composite, thereby promoting microglial Aβ phogocytosis/clearance via Aβ receptor-mediated endocytosis. After B6-PNi NPs treatment, the Aβ burden, neuroinflammation and cognitive impairments are relieved in AD transgenic mice. This work provides the Aβ sequence-matching strategy for Aβ inhibitor design in AD treatment, showing meaningful insight in biomedicine.
PubMed: 38939864
DOI: 10.1002/EXP.20230048 -
Cell culture-derived extracellular vesicles: Considerations for reporting cell culturing parameters.Journal of Extracellular Biology Oct 2023Cell culture-conditioned medium (CCM) is a valuable source of extracellular vesicles (EVs) for basic scientific, therapeutic and diagnostic applications. Cell culturing...
Cell culture-conditioned medium (CCM) is a valuable source of extracellular vesicles (EVs) for basic scientific, therapeutic and diagnostic applications. Cell culturing parameters affect the biochemical composition, release and possibly the function of CCM-derived EVs (CCM-EV). The CCM-EV task force of the Rigor and Standardization Subcommittee of the International Society for Extracellular Vesicles aims to identify relevant cell culturing parameters, describe their effects based on current knowledge, recommend reporting parameters and identify outstanding questions. While some recommendations are valid for all cell types, cell-specific recommendations may need to be established for non-mammalian sources, such as bacteria, yeast and plant cells. Current progress towards these goals is summarized in this perspective paper, along with a checklist to facilitate transparent reporting of cell culturing parameters to improve the reproducibility of CCM-EV research.
PubMed: 38939735
DOI: 10.1002/jex2.115 -
JACC. Advances Feb 2024Previous studies have linked cardiovascular risk factors during midlife to cognitive function in later life. However, few studies have looked at the association between...
BACKGROUND
Previous studies have linked cardiovascular risk factors during midlife to cognitive function in later life. However, few studies have looked at the association between cardiac function, brain structure, and cognitive function and even less have included diverse middle-aged populations.
OBJECTIVES
The objective of this study was to determine associations between cardiac and brain structure and function in a multiethnic cohort of middle-aged adults.
METHODS
A cross-sectional study was conducted in participants of the Dallas Heart Study phase 2 (N = 1,919; 46% Black participants). Left ventricular (LV) mass, LV ejection fraction, LV concentricity, and peak systolic strain (LV E) were assessed by cardiac magnetic resonance imaging. White matter hyperintensities (WMH) volume was measured by fluid attenuated inversion recovery magnetic resonance imaging. The Montreal Cognitive Assessment was used to measure cognitive functioning. Associations between cardiac and brain measures were determined using multivariable linear regression after adjusting for cardiovascular risk factors, education level, and physical activity.
RESULTS
LV ejection fraction was associated with total Montreal Cognitive Assessment score (β = 0.06 [95% CI: 0.003-0.12], = 0.042) and LV E was associated with WMH volume (β = 0.08 [95% CI: 0.01-0.14], = 0.025) in the overall cohort without significant interaction by race/ethnicity. Higher LV mass and concentricity were associated with larger WMH volume in the overall cohort (β = 0.13 [95% CI: 0.03-0.23], = 0.008 and 0.10 [95% CI: 0.03-0.17], = 0.005). These associations were more predominant in Black than White participants (β = 0.17 [95% CI: 0.04-0.30] vs β = -0.009 [95% CI: -0.16 to 0.14], = 0.036 and β = 0.22 [95% CI: 0.13-0.32] vs β = -0.11 [95% CI: -0.21 to -0.01], < 0.0001, for LV mass and concentricity, respectively).
CONCLUSIONS
Subclinical cardiac dysfunction indicated by LVEF was associated with lower cognitive function. Moreover, LV mass and concentric remodeling were associated with higher WMH burden, particularly among Black individuals.
PubMed: 38939405
DOI: 10.1016/j.jacadv.2023.100777 -
JACC. Advances Feb 2024
PubMed: 38939404
DOI: 10.1016/j.jacadv.2023.100778 -
Biochemistry Research International 2024Eg5 is a protein encoded by KIF11 gene and is primarily involved in correct mitotic cell division. It is also involved in nonmitotic processes such as polypeptide... (Review)
Review
Eg5 is a protein encoded by KIF11 gene and is primarily involved in correct mitotic cell division. It is also involved in nonmitotic processes such as polypeptide synthesis, protein transport, and angiogenesis. The scientific literature sheds light on the ubiquitous functions of KIF11 and its involvement in the onset and progression of different pathologies. This review focuses attention on two main points: (1) the correlation between Eg5 and cancer and (2) the involvement of Eg5 in noncancerous conditions. Regarding the first point, several tumors revealed an overexpression of this kinesin, thus pushing to look for new Eg5 inhibitors for clinical practice. In addition, the evaluation of Eg5 expression represents a crucial step, as its overexpression could predict a poor prognosis for cancer patients. Referring to the second point, in specific pathological conditions, the reduced activity of Eg5 can be one of the causes of pathological onset. This is the case of Alzheimer's disease (AD), in which A and Tau work as Eg5 inhibitors, or in acquired immune deficiency syndrome (AIDS), in which Tat-mediated Eg5 determines the loss of CD T-lymphocytes. Reduced Eg5 activity, due to mutations of KIF11 gene, is also responsible for pathological conditions such as microcephaly with or without chorioretinopathy, lymphedema, or intellectual disability (MCLRI) and familial exudative vitreous retinopathy (FEVR). In conclusion, this review highlights the double impact that overexpression or loss of function of Eg5 could have in the onset and progression of different pathological situations. This emphasizes, on one hand, a possible role of Eg5 as a potential biomarker and new target in cancer and, on the other hand, the promotion of Eg5 expression/activity as a new therapeutic strategy in different noncancerous diseases.
PubMed: 38939361
DOI: 10.1155/2024/3649912 -
Health Care Science Feb 2024Alzheimer's disease (AD) is a progressive brain disorder that impairs cognitive functions, behavior, and memory. Early detection is crucial as it can slow down the...
INTRODUCTION
Alzheimer's disease (AD) is a progressive brain disorder that impairs cognitive functions, behavior, and memory. Early detection is crucial as it can slow down the progression of AD. However, early diagnosis and monitoring of AD's advancement pose significant challenges due to the necessity for complex cognitive assessments and medical tests.
METHODS
This study introduces a data acquisition technique and a preprocessing pipeline, combined with multivariate long short-term memory (M-LSTM) and AdaBoost models. These models utilize biomarkers from cognitive assessments and neuroimaging scans to detect the progression of AD in patients, using The AD Prediction of Longitudinal Evolution challenge cohort from the Alzheimer's Disease Neuroimaging Initiative database.
RESULTS
The methodology proposed in this study significantly improved performance metrics. The testing accuracy reached 80% with the AdaBoost model, while the M-LSTM model achieved an accuracy of 82%. This represents a 20% increase in accuracy compared to a recent similar study.
DISCUSSION
The findings indicate that the multivariate model, specifically the M-LSTM, is more effective in identifying the progression of AD compared to the AdaBoost model and methodologies used in recent research.
PubMed: 38939169
DOI: 10.1002/hcs2.84