Biology subjects

Wang, D. D.

Publications and source records attributed to Wang, D. D..

6 recordsLinked to original sources

bioRxiv2026

Pallidal beta oscillations underlying locomotor adaptation in Parkinsons disease

BackgroundLocomotor adaptation is essential for adjusting walking patterns to complex environments. This study investigated locomotor adaptation deficits in people with Parkinsons disease (PD) and examined oscillatory activity in the globus pallidus internus (GPi) during walking adaptation. We hypothesized that elevated beta-band activity in the GPi is associated with reduced locomotor adaptability in PD. MethodsTwelve PD patients with GPi deep brain stimulation (DBS) (eleven bilateral and one unilateral) were included. Local field potentials (LFPs) were recorded from DBS electrodes during split-belt treadmill walking. Patients were tested in the medication-off, DBS-off state. Locomotor adaptation was measured as the change in step length asymmetry during split-belt walking, with smaller changes indicating greater adaptation deficits. ResultsWe found that GPi high beta (20-30 Hz) and low gamma (30-60 Hz) oscillations were modulated during split-belt walking. Compared to adapters, non-adapters showed decreased movement-related beta suppression during walking. Across participants, beta activity in the GPi contralateral to the fast leg was negatively associated with adaptation magnitude (Spearmans {rho} = -0.65 to -0.75). ConclusionsGPi oscillations are dynamically modulated during locomotor adaptation in PD. Increased beta activity may underlie impaired sensorimotor adaptation during walking. These findings provide novel insight into basal ganglia mechanisms of gait adaptation in PD and suggest that elevated GPi beta activity may serve as a marker of locomotor adaptation deficits.

neuroscience↗

bioRxiv2025

Dynamics-enhanced Molecular Property Prediction Guided by Deep Learning

Molecular property prediction (MPP) is a key challenge in computational biology and drug discovery. Traditional approaches mostly rely on feature representations yielded from static structures of molecules, ignoring their dynamic nature. The scarcity of dynamics data in public databases and the complexity of learning such high-dimensional data made it more difficult for dynamics-involved studies. Accordingly, we built a series of dynamics datasets for MPP tasks by performing comprehensive molecular dynamics (MD) simulations on different molecules. In addition, we proposed a dynamically enhanced molecular representation (DEMR) method with multiple sampling strategies for the dynamics frames. Besides, two deep learning pipelines were employed for mapping DEMR to the molecular properties in various tasks. Our models achieved better performance in different MPP tasks, with a practical guidance in efficient frame selection. This study highlights the significance of integrating MD data into MPP tasks and opens new avenues for structure-based drug design. The generated MD datasets are publicly available in a Zenodo repository at https://doi.org/10.5281/zenodo.15788151.

bioinformatics↗

bioRxiv2025

Predicting Risk of Transfusion-Induced Red Blood Cell Alloimmunization Using Statistical and Machine Learning Approaches in the Recipient Epidemiology and Donor Evaluation Study (REDS-III) Database

Red blood cell (RBC) alloimmunization is a common complication from blood transfusion, often resulting in accelerated donor RBC destruction. Patients show substantial variation in their predisposition to RBC alloimmunization. Previous studies have identified several risk factors, but to our knowledge, there have been no studies that predict risk of RBC alloimmunization by modeling multiple potential risk factors simultaneously. Here, our study represents the first attempt to build prediction models for RBC alloimmunization using the large sample size and rich set of potential risk factors available in the Recipient Epidemiology and Donor Evaluation Study (REDS-III) recipient database. To develop the prediction models, we applied a range of approaches, including traditional statistical models (logistic regression), and modern machine learning (including gradient boosting, random forest, and XGBoost), deep learning (the multilayer perceptron method), and large language models (LLM). XGBoost demonstrates the overall best performance among models providing uncertainty quantification (F1= 0.672 and area under the ROC curve [AUC-ROC]=0.752). LLMs show promising results with the best F1 scores (0.677-0.687), though they are limited by their inability to provide uncertainty estimates, they hold the potential for use as an interactive chatbot for patients. Although there is ample room for performance improvement, limiting the analysis to patients predicted with >80% confidence by XGBoost resulted in a substantially improved AUC-ROC of 0.919, which can be of potential clinical significance.

immunology↗

bioRxiv2024

Learning-related oscillatory dynamics in the human cortico-basal ganglia network during motor sequence initiation in Parkinson's Disease

Learning dexterous motor sequences is crucial to autonomy and quality of life but can be altered in Parkinsons Disease (PD). Learning involves optimizing pre-movement planning (preplanning) of multiple sequence elements to reduce computational overhead during active movement. However, it is unclear which brain regions mediate preplanning or how this process evolves with learning. Recording cortico-basal ganglia field potentials during a multi-day typing task in four individuals with PD, we found evidence for network-wide multi-element preplanning that improved with learning, facilitated by functional connectivity. In both cortex and basal ganglia, pre-movement gamma ({gamma}, 30-250 Hz) activity, historically linked to population spiking, distinguished between future action sequences and became increasingly predictive with learning. For motor cortex {gamma}, this increase was tied to learning-related cross-frequency coupling led by cortically-driven network delta ({delta}, 0.5-4 Hz) synchrony. More generally, coordinated network {delta} supported a complex pattern of learning-driven cross-frequency couplings within and between cortex and basal ganglia, including striatal lead of cortical beta ({beta}, 12-30 Hz) activity, reflecting the specialized roles of these brain regions in motor preparation. In contrast, impaired learning was characterized by practice-driven decreases in {gamma}s predictive value, limited cross-frequency coupling and absent network {delta} synchrony, with network dynamics possibly altered by pathologically high inter-basal ganglia {delta} synchrony. These results suggest that cortically-led {delta} phase coordination optimized cortico-basal ganglia multi-element preplanning through enhanced recruitment of higher-frequency neural activity. Neurostimulation that enhances cortico-basal ganglia {delta} synchrony may thus hold potential for improving skilled fine motor control in PD.

neuroscience↗

bioRxiv2024

Weighted variance component test for the integrative multi-omics analysis of microbiome data

Metabolic dysregulation and alterations have been linked to various diseases and conditions. Innovations in high-throughput technology now allow rapid profiling of the metabolome and metagenome -- often the gene content of bacterial populations -- for characterizing metabolism. Due to the small sample sizes and high dimensionality of the data, pathway analysis (wherein the effect of multiple genes or metabolites on an outcome is cumulatively assessed) of metabolomic data is commonly conducted and also represents a standard for metagenomic analysis. However, how to integrate both data types remains unclear. Recognizing that a metabolic pathway can be complementarily characterized by both metagenomics and metabolomics, we propose a weighted variance components framework to test if the joint effect of genes and metabolites in a biological pathway is associated with outcomes. The approach allows analytic p-value calculation, correlation between data types, and optimal weighting. Power simulations show that our approach often outperforms other strategies while maintaining type I error. The approach is illustrated on real data.

bioinformatics↗

bioRxiv2020

Chronic wireless streaming of invasive neural recordings at home for circuit discovery and adaptive stimulation

Invasive neural recording in humans shows promise for understanding the circuit basis of brain disorders. Most recordings have been done for short durations from externalized brain leads in hospital settings, or from first-generation implantable sensing devices that offer only intermittent brief streaming of time series data. Here we report the first human use of an implantable neural interface for wireless multichannel streaming of field potentials over long periods, with and without simultaneous therapeutic neurostimulation, untethered to receiving devices. Four Parkinsons disease patients streamed bilateral 4-channel motor cortical and basal ganglia field potentials at home for over 500 hours, paired with wearable monitors that behaviorally categorize states of inadequate or excessive movement. Motor state during normal home activities was efficiently decoded using either supervised learning or unsupervised clustering algorithms. This platform supports adaptive deep brain stimulation, and may be widely applicable to brain disorders treatable by invasive neuromodulation.

neuroscience↗