Search bioRxiv⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/2025.06.30.662262

In silico model of basal ganglia Deep Brain Stimulation in Parkinsons Disease captures range of effective parameters for pathological beta power suppression

Ahmadipour, M.·Mazzoni, A.
View original source ↗Export metadata ↓

Abstract

In Parkinsons disease (PD), the beta band ([12 30] Hz) component of basal ganglia activity is pathologically high. Deep brain stimulation (DBS) is an effective treatment to suppress symptoms of PD and is known to suppress pathological beta activity. However, the mechanism underlying this effect is not completely understood. Here, we tested the circuital effects of DBS in a computational model of the basal ganglia network in dopamine-depleted condition mimicking PD. Our model reproduces suppression of beta pathological oscillations in the basal ganglia network induced by subthalamic nucleus (STN) DBS. Crucially, this occurs for realistic levels of DBS intensity only if we incorporate short-term plasticity in STN projections to their downstream targets. STN stimulation hampers beta oscillations in the subthalamo-pallidal beta loop. This induces a progressive dephasing between this loop and the striato-pallidal beta loop, which leads in turn to a network-wide suppression of beta oscillations. This is also reflected in a restoration of the balance between D1 and D2 firing rates, which was altered by dopamine depletion. Moreover, the model also reproduces the DBS-induced gamma activity associated with symptom recovery. Finally, we explored the circuital effects of a broad range of DBS parameters in suppressing beta oscillations, focusing on the clinically relevant range of 60-150 Hz stimulation. The model suggests that 60-80 Hz stimulation frequencies might achieve beta desynchronization for an intensity even lower than the one needed at the standard 130 Hz frequency. Overall, our model lays the ground for in-silico tests of a broad spectrum of stimulation patterns. Author SummaryDeep brain stimulation (DBS) is an effective treatment for alleviating symptoms of Parkinsons disease (PD), but its underlying mechanism is unclear. Motor symptoms of PD are linked to exaggerated beta-band oscillations in a particular brain area called basal ganglia (BG). In this study, we developed a computational model to simulate how DBS affects brain circuits involved in PD. We found that stimulation can reduce pathological beta rhythms at realistic intensities only when we include realistic synaptic dynamics that temporarily adjust connection strength based on recent activity. Our model shows that stimulation decouples two available beta loops in the BG network and helps restore normal patterns of activity. Interestingly, we also found that stimulation at 60-80 Hz might be effective at lower intensities than the standard 130 Hz. Our work provides new insights into how DBS functions and supports safe and efficient exploration of a wide range of stimulation patterns.

Source connections

Explore related subjects

bioengineeringPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Ahmadipour, M., Mazzoni, A.. 2025-07-04. In silico model of basal ganglia Deep Brain Stimulation in Parkinsons Disease captures range of effective parameters for pathological beta power suppression. https://doi.org/10.1101/2025.06.30.662262

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

AtomWeaver: Multi-Component Flow Matching with a Structured Geometric Prior Facilitates Non-Canonical Peptide Design

Fixed-backbone sequence discovery, or inverse folding, is a critical recurring task in the development of new polypeptide therapeutics. Once promising backbones are established for a target pocket, computational inverse folding methods greatly help accelerate generation of candidate sequences. Such methods are mature for the traditional case of limiting to the fixed twenty-letter canonical vocabulary; however, they cannot access the broader space of non-canonical amino acids (NCAAs). This design constraint is exacerbated for peptide binders, a fast-growing modality that readily incorporates NCAAs, though in practice non-canonical design frequently depends on laborious medicinal-chemistry campaigns. An extension of inverse folding to NCAAs is thus critical to accelerating design of novel therapeutic peptides. AtomWeaver uses a joint all-site, atom-level generative scheme that does not restrict side-chain categorical assignment by either predetermined or co-resolving residue identity. Conditioned only on a fixed peptide backbone and its target protein, its multi-component flow guides side-chain atoms as unlabeled points in R3 from a nested shell prior to a variable-count final atom cloud. Identity is then read by matching each predicted cloud against a reference library of canonical and non-canonical templates. Since identity is decided only at decode time, the addressable vocabulary is a property of the library rather than of the trained weights: a new NCAA costs one reference structure and no retraining, and the model can select residues it was never prompted for and never saw in training. On a deep mutational scan of two peptide-target systems, AtomWeaver's canonical readout shows high observed mean agreement with experimental values among the compared inverse-folding methods. In the mixed canonical-noncanonical setting that canonical-only baselines cannot support at all, it likewise retains ranking signal across both systems. AtomWeaver also displayed self-consistent designs on de novo binder backbones, with the highest interface confidence among compared methods. Notably, it reached these metrics while achieving broad empirical coverage of our 300-residue vocabulary, including four non-canonical types never visible in training. AtomWeaver thus serves canonical and non-canonical peptide design alike, while transforming residue vocabulary to an expandable inference-time choice.

bioengineering↗
bioRxiv2026

The Influence of Obesity and Body Shape on Sagittal Plane Knee Kinematics and Kinetics during Obstacle Crossing

Altered walking mechanics in individuals with obesity can contribute to knee osteoarthritis. The gait deviations may become more pronounced during obstacle crossing. In women, body fat distribution may further influence knee load, especially when excess fat accumulates in the thighs and hips. However, relatively little is known about how regional fat distribution affects gait in women with obesity. This study investigated how obesity and, among women, different fat distributions (Apple: more abdominal fat; Pear: more lower-limb fat) influence knee biomechanics during walking with and without obstacle crossing. Participants were 15 controls without obesity (NB) and 27 with obesity (OB). Within female participants, 10 without obesity (fNB) were compared with 20 with obesity, stratified by waist-hip ratio (Apple:10, Pear:10). Speed-adjusted statistical parametric mapping applied a general linear model (NB vs. OB) and an analysis of covariance (fNB vs. Apple vs. Pear). OB exhibited a significantly greater late-stance knee extension moment than NB across all tasks, and this difference persisted among fNB, Apple, and Pear in obstacle tasks (p<0.05). OB walked with reduced knee flexion during the early-stance leading limb after crossing a medium-height obstacle (p=0.048) and a high-height obstacle (p=0.008) compared to NB. There were significant body-shape effects (p<0.05), and post-hoc comparisons confirmed that Pear had lower knee angles than fNB in both leading-limb conditions after crossing medium- and high-height obstacles (p=0.008 and p=0.001, respectively). These findings suggest that obstacle crossing helps illuminate how excess weight influences knee biomechanics, and how regional fat distribution modulates the degree of this alteration.

bioengineering↗
bioRxiv2026

RNASeek: A Cross-Phyla Generative Foundation Model for Multipurpose RNA Modeling and Reinforcement Learning-Based Design

RNA plays central roles in regulating information flow and provides a versatile substrate for engineering biological functions. While large language models (LLMs) have transformed natural language processing and protein design, a general framework connecting RNA foundation models to functional sequence design remains limited. Here, we present RNASeek, a 1.6-billion-parameter generative foundation model built on a DeepSeek architecture and trained on a cross-phyla transcriptomic corpus for RNA sequence representation and generation. Natural-language tokens enable flexible conditional prediction and sequence design using a unified backbone. RNASeek captures species-specific transcript features and intron-exon boundaries in a zero-shot setting. We then fine-tune RNASeek to predict ribozyme self-cleavage activity and viral mRNA stability, revealing interpretable sequence features associated with function, including ribozyme loop flexibility and stem stability, as well as AU-rich motifs associated with mRNA stability. We use these functional predictors as reward models and apply Group Relative Policy Optimization (GRPO) to update the generation policy of RNASeek toward sequences with desired properties. GRPO-guided generation produces faster-cleaving ribozymes and stability-enhancing 3' UTRs while satisfying user-specified IUPAC constraints. Experimentally validated RNASeek-generated ribozymes achieve wild-type levels of activity, while RNASeek-generated 3' UTR sequences exceed the performance of the training data and benchmarked AI-generated 3' UTRs. Together, RNASeek establishes a unified pretrain-predict-optimize framework that connects learned RNA function to controllable de novo sequence design and provides a general strategy for engineering regulatory RNAs with desired properties.

bioengineering↗