Bump version to 0.8.1 and clean up dependencies

- Remove unused dependencies: pyqt6, noise
- Move scipy-stubs to dev dependencies only
- Clean up Watanabe examples: remove colloquial references, use proper
  citation format "Watanabe and Kohn (2015)"

Author: RaulSimpetru

CHANGELOG.md

@@ -7,6 +7,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.8.1] - 2025-12-28
+
+### Changed
+- **Dependencies**: Removed unused dependencies
+  - Removed `pyqt6` (not used in codebase)
+  - Moved `scipy-stubs` to dev dependencies only (was duplicated)
+
 ## [0.8.0] - 2025-12-28
 
 ### Added

README.md

@@ -8,7 +8,7 @@
 
   [![Documentation](https://img.shields.io/badge/docs-latest-blue.svg)](https://nsquaredlab.github.io/MyoGen/)
   [![Python 3.12+](https://img.shields.io/badge/python-3.12+-blue.svg)](https://www.python.org/downloads/)
-  [![Version](https://img.shields.io/badge/version-0.8.0-orange.svg)](https://github.com/NsquaredLab/MyoGen)
+  [![Version](https://img.shields.io/badge/version-0.8.1-orange.svg)](https://github.com/NsquaredLab/MyoGen)
 
   [Installation](https://nsquaredlab.github.io/MyoGen/#installation) •
   [Documentation](https://nsquaredlab.github.io/MyoGen/) •

examples/03_papers/watanabe/01_compute_baseline_force.py

@@ -1,5 +1,5 @@
 """
-Compute Reference Force with Constant Drive (Watanabe Network)
+Compute Reference Force with Constant Drive
 ===============================================================
 
 This example computes the reference muscle force using constant descending drive.
@@ -10,7 +10,7 @@
 
     - 800 motor neurons
     - 400 descending drive neurons
-    - 30% connectivity (Watanabe specification)
+    - 30% connectivity
     - Constant drive frequency (configurable, default 40 Hz)
     - Poisson process (order 1)
 
@@ -29,7 +29,7 @@
   These represent cortical/brainstem input to motor neurons.
   ``poisson_batch_size=1`` creates order-1 (renewal) Poisson processes.
 
-- :class:`~myogen.simulator.neuron.Network`:
+- :class:`~myogen.simulator.Network`:
   Container that manages populations and synaptic connections between them.
   Supports probabilistic connectivity (``connect()``) and external input (``connect_from_external()``).
 
@@ -71,7 +71,7 @@
 from myogen import RANDOM_GENERATOR
 from myogen.simulator import RecruitmentThresholds
 from myogen.simulator.core.force.force_model import ForceModel
-from myogen.simulator.neuron import Network
+from myogen.simulator import Network
 from myogen.simulator.neuron.populations import AlphaMN__Pool, DescendingDrive__Pool
 from myogen.utils.helper import calculate_firing_rate_statistics
 from myogen.utils.nmodl import load_nmodl_mechanisms
@@ -121,7 +121,7 @@
 print(f"DD neurons: {N_DD_NEURONS}")
 print(f"Connection probability: {DD_CONNECTIVITY:.1%}")
 print(f"Drive frequency: {DD_DRIVE_HZ:.1f} Hz (constant)")
-print(f"Process type: Poisson (batch size: 1)")
+print("Process type: Poisson (batch size: 1)")
 print("=" * 60 + "\n")
 
 ##############################################################################
@@ -293,7 +293,7 @@
 n_active = int(stats["n_active"])
 
 print("Firing Rate Statistics:")
-print(f"  Active units: {n_active}/{N_MOTOR_UNITS} ({n_active/N_MOTOR_UNITS*100:.1f}%)")
+print(f"  Active units: {n_active}/{N_MOTOR_UNITS} ({n_active / N_MOTOR_UNITS * 100:.1f}%)")
 print(f"  Mean firing rate: {fr_mean:.2f} Hz")
 print(f"  Std deviation: {fr_std:.2f} Hz")
 print(f"  Coefficient of variation: {fr_std / fr_mean:.3f}\n")
@@ -313,9 +313,9 @@
 
 force_model = ForceModel(
     recruitment_thresholds=recruitment_thresholds,
-    recording_frequency__Hz=2048 * pq.Hz,        # Output sampling rate
-    longest_duration_rise_time__ms=90.0 * pq.ms, # Slowest unit's rise time
-    contraction_time_range_factor=3,              # Ratio of slowest/fastest T
+    recording_frequency__Hz=2048 * pq.Hz,  # Output sampling rate
+    longest_duration_rise_time__ms=90.0 * pq.ms,  # Slowest unit's rise time
+    contraction_time_range_factor=3,  # Ratio of slowest/fastest T
 )
 
 # Generate force from spike trains

examples/03_papers/watanabe/02_optimize_oscillating_dc.py

@@ -1,12 +1,12 @@
 """
-Optimize Oscillating DC Offset to Match Reference Force (Watanabe Network)
+Optimize Oscillating DC Offset to Match Reference Force
 ===========================================================================
 
 This example optimizes the DC offset component of an oscillating descending drive to
 match the reference force produced by constant drive (from script 01).
 The drive pattern is: `DC_offset + 20*sin(2π*20*t)`.
 
-This finds the equivalent of Watanabe's 58 pps value (Phase 3) that produces the same
+This finds the equivalent of the original 58 pps value (Phase 3) that produces the same
 force as constant drive (Phase 1) when combined with 20 Hz oscillation.
 
 .. note::
@@ -31,7 +31,7 @@
   Poisson spike generators driven by time-varying rate: ``DC_offset + amplitude*sin(2πft)``.
   The ``integrate()`` method advances the internal Poisson process by one timestep.
 
-- :class:`~myogen.simulator.neuron.Network`:
+- :class:`~myogen.simulator.Network`:
   Rebuilds the network for each trial with same connectivity parameters (30%).
 
 - :class:`~myogen.simulator.core.force.force_model.ForceModel`:
@@ -41,7 +41,7 @@
 
 **Key Concept**: The oscillation itself contributes to motor neuron activation, so
 the DC offset in Phase 3 must be *lower* than the constant drive in Phase 1 to
-produce the same mean force. Watanabe found 58 Hz offset vs 65 Hz constant.
+produce the same mean force. The original paper found 58 Hz offset vs 65 Hz constant.
 
 **Workflow Position**: Step 2 of 6
 
@@ -130,8 +130,7 @@
 
 if not reference_file.exists():
     raise FileNotFoundError(
-        f"Reference force not found: {reference_file}\n"
-        f"Run 01_compute_baseline_force.py first!"
+        f"Reference force not found: {reference_file}\nRun 01_compute_baseline_force.py first!"
     )
 
 with open(reference_file, "r") as f:
@@ -530,5 +529,7 @@ def objective(trial):
 
 print("\n[DONE] DC offset optimization complete!")
 print(f"Best DC offset: {best_trial.user_attrs['dc_offset__Hz']:.2f} Hz")
-print(f"Original Watanabe: 65 Hz constant → 58 Hz + oscillation (ratio: {58/65:.3f})")
-print(f"This implementation: {REFERENCE_DRIVE__HZ:.1f} Hz constant → {best_trial.user_attrs['dc_offset__Hz']:.1f} Hz + oscillation (ratio: {best_trial.user_attrs['dc_offset__Hz']/REFERENCE_DRIVE__HZ:.3f})")
+print(f"Original Watanabe: 65 Hz constant → 58 Hz + oscillation (ratio: {58 / 65:.3f})")
+print(
+    f"This implementation: {REFERENCE_DRIVE__HZ:.1f} Hz constant → {best_trial.user_attrs['dc_offset__Hz']:.1f} Hz + oscillation (ratio: {best_trial.user_attrs['dc_offset__Hz'] / REFERENCE_DRIVE__HZ:.3f})"
+)

examples/03_papers/watanabe/03_10pct_mvc_simulation.py

@@ -1,16 +1,16 @@
 """
-Watanabe Spinal Network Simulation with Optimized DC Offset
+Spinal Network Simulation with Optimized DC Offset
 ============================================================
 
-This example runs the Watanabe et al. (2015) spinal network simulation using:
+This example runs the Watanabe and Kohn (2015) spinal network simulation using:
 - Constant drive (Phase 1) - baseline from script 01
 - Constant + oscillation (Phase 2) - oscillation added to baseline
 - Optimized DC + oscillation (Phase 3) - DC offset matches Phase 1 force
 
 .. note::
     **Simulation Phases** (5 seconds each):
 
-    - **Phase 1**: Constant drive (Watanabe baseline, e.g., 40-65 Hz)
+    - **Phase 1**: Constant drive (baseline, e.g., 40-65 Hz)
     - **Phase 2**: Constant + 20*sin(20Hz) (oscillation added to baseline)
     - **Phase 3**: Optimized DC + 20*sin(20Hz) (DC offset matches Phase 1 force)
 
@@ -22,7 +22,7 @@
 
 **Scientific Context**: Demonstrates how shared descending drive creates motor unit
 synchronization, while independent noise reduces it. Phase 3 uses lower DC offset than
-Phase 1 (like Watanabe's 58 < 65) because oscillation contributes additional activation.
+Phase 1 (e.g., 58 < 65) because oscillation contributes additional activation.
 
 **MyoGen Components Used**:
 
@@ -54,8 +54,8 @@
   Essential for 15-second simulations with 800+ neurons.
 
 **Key Concept**: The ratio of shared vs independent input determines motor unit
-synchronization. Watanabe showed that 20 Hz cortical oscillations are transmitted
-through the spinal network and can be detected in EMG coherence spectra.
+synchronization.  Watanabe and Kohn (2015) showed that 20 Hz cortical oscillations are
+transmitted through the spinal network and can be detected in EMG coherence spectra.
 
 **Workflow Position**: Step 3 of 6
 
@@ -113,8 +113,7 @@
 
 if not dc_offset_file.exists():
     raise FileNotFoundError(
-        f"Optimized DC offset not found: {dc_offset_file}\n"
-        f"Run 02_optimize_oscillating_dc.py first!"
+        f"Optimized DC offset not found: {dc_offset_file}\nRun 02_optimize_oscillating_dc.py first!"
     )
 
 with open(dc_offset_file) as f:
@@ -134,7 +133,9 @@
 print("  Phase 1: 65 pps")
 print("  Phase 2: 65 + 20*sin(20Hz) pps")
 print("  Phase 3: 58 + 20*sin(20Hz) pps")
-print(f"\nOptimization ratio: {DC_OFFSET_OPTIMIZED/DD_DRIVE_CONSTANT:.3f} (Watanabe: {58/65:.3f})")
+print(
+    f"\nOptimization ratio: {DC_OFFSET_OPTIMIZED / DD_DRIVE_CONSTANT:.3f} (Watanabe: {58 / 65:.3f})"
+)
 print(f"Current setup uses {DD_DRIVE_CONSTANT:.1f} Hz baseline (configurable in script 11)")
 print("=" * 70 + "\n")
 
@@ -211,8 +212,8 @@
 plt.xlabel("Time (s)")
 plt.ylabel("DD Drive (Hz)")
 plt.title("Descending Drive Pattern (Optimized)")
-plt.axvline(segment_duration__s, color='r', linestyle='--', alpha=0.5, label='Phase boundary')
-plt.axvline(2*segment_duration__s, color='r', linestyle='--', alpha=0.5)
+plt.axvline(segment_duration__s, color="r", linestyle="--", alpha=0.5, label="Phase boundary")
+plt.axvline(2 * segment_duration__s, color="r", linestyle="--", alpha=0.5)
 plt.legend()
 plt.grid()
 
@@ -352,9 +353,9 @@ def eachStep(
 #       Creates NetCon objects for external event delivery (e.g., commands).
 #
 # The ratio of shared (DD) to private (IN) input determines the degree of
-# motor unit synchronization - a key parameter in Watanabe's analysis.
+# motor unit synchronization - a key parameter in the original analysis.
 
-# DD connectivity: 30% probability (Watanabe specification)
+# DD connectivity: 30% probability (original specification)
 # This means each motor neuron receives input from ~30% of DD neurons
 DD_connectivity = 0.3
 expected_DD_per_MN = int(DD_connectivity * nDD)  # ~120 connections per MN
@@ -514,4 +515,4 @@ def step_callback(step_counter):
 print(f"  Spikes: {save_path / 'watanabe_spikes.pkl'}")
 print(f"\nPhase 1 constant drive: {DD_DRIVE_CONSTANT:.2f} Hz")
 print(f"Phase 3 DC offset: {DC_OFFSET_OPTIMIZED:.2f} Hz (optimized to match Phase 1 force)")
-print(f"Ratio: {DC_OFFSET_OPTIMIZED/DD_DRIVE_CONSTANT:.3f} (Watanabe: {58/65:.3f})")
+print(f"Ratio: {DC_OFFSET_OPTIMIZED / DD_DRIVE_CONSTANT:.3f} (Watanabe: {58 / 65:.3f})")

examples/03_papers/watanabe/04_load_and_analyze_results.py

@@ -2,7 +2,7 @@
 Load and Analyze Membrane Potentials and Spike Trains
 ======================================================
 
-This example demonstrates **post-simulation analysis** of the Watanabe et al. (2015)
+This example demonstrates **post-simulation analysis** of the Watanabe and Kohn (2015)
 spinal network model. It shows how to load chunked simulation data, convert it to
 NEO format, and analyze membrane potentials, spike trains, and population dynamics.
 
@@ -52,7 +52,7 @@
 - ``neo.AnalogSignal``: Continuous data (e.g., membrane potential) with sampling rate
 
 **Use Case**: Analyze motor unit recruitment, synchronization, and firing patterns
-to validate model predictions against experimental data from Watanabe et al. (2015).
+to validate model predictions against experimental data from Watanabe and Kohn (2015).
 
 **Workflow Position**: Step 4 of 6
 
@@ -197,8 +197,20 @@
 
         # Highlight the three experimental phases
         ax.axvspan(0, phase1_end, alpha=0.2, color=phase_colors[0], label="Phase 1: Constant")
-        ax.axvspan(phase1_end, phase2_end, alpha=0.2, color=phase_colors[1], label="Phase 2: Sinusoid DC=65")
-        ax.axvspan(phase2_end, phase3_end, alpha=0.2, color=phase_colors[2], label="Phase 3: Sinusoid DC=58")
+        ax.axvspan(
+            phase1_end,
+            phase2_end,
+            alpha=0.2,
+            color=phase_colors[1],
+            label="Phase 2: Sinusoid DC=65",
+        )
+        ax.axvspan(
+            phase2_end,
+            phase3_end,
+            alpha=0.2,
+            color=phase_colors[2],
+            label="Phase 3: Sinusoid DC=58",
+        )
 
         if ax == axes[0]:
             ax.legend(loc="upper right", framealpha=1.0, edgecolor="none")
@@ -267,8 +279,12 @@
 
     # Highlight the three experimental phases
     ax.axvspan(0, phase1_end, alpha=0.2, color=phase_colors[0], label="Phase 1: Constant")
-    ax.axvspan(phase1_end, phase2_end, alpha=0.2, color=phase_colors[1], label="Phase 2: Sinusoid DC=65")
-    ax.axvspan(phase2_end, phase3_end, alpha=0.2, color=phase_colors[2], label="Phase 3: Sinusoid DC=58")
+    ax.axvspan(
+        phase1_end, phase2_end, alpha=0.2, color=phase_colors[1], label="Phase 2: Sinusoid DC=65"
+    )
+    ax.axvspan(
+        phase2_end, phase3_end, alpha=0.2, color=phase_colors[2], label="Phase 3: Sinusoid DC=58"
+    )
     ax.legend(loc="upper right", framealpha=1.0, edgecolor="none")
 
     plt.tight_layout()

examples/03_papers/watanabe/05_compute_force_from_spinal_network.py

@@ -60,7 +60,7 @@
 
 **Use Case**: Generate realistic muscle force output from motor neuron spike trains
 for biomechanical analysis, force-EMG relationship studies, and validation against
-experimental force recordings from Watanabe et al. (2015).
+experimental force recordings from Watanabe and Kohn (2015).
 
 **Workflow Position**: Step 5 of 6
 

examples/03_papers/watanabe/06_visualize.py

@@ -2,7 +2,7 @@
 Generate Publication-Quality Figures and Analysis Plots
 ========================================================
 
-This example generates **publication-quality figures** for the Watanabe et al. (2015)
+This example generates **publication-quality figures** for the Watanabe and Kohn (2015)
 spinal network model reproduction. It creates comprehensive visualizations including
 coherence analysis, force timeseries, and detailed raster plots.
 

examples/03_papers/watanabe/README.rst

@@ -1,6 +1,6 @@
 .. _watanabe-reproduction:
 
-Watanabe and Kohn (2015)
+ Watanabe and Kohn (2015)
 ======================
 
 Reproduces findings from:

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "MyoGen"
-version = "0.8.0"
+version = "0.8.1"
 description = "Modular and extensible neuromuscular simulation framework for generating physiologically grounded motor-unit activity, muscle force, and EMG signals (surface and intramuscular)"
 readme = "README.md"
 requires-python = ">=3.12"
@@ -37,17 +37,14 @@ dependencies = [
     "neuron==8.2.7 ; sys_platform == 'linux' or sys_platform == 'darwin'",
     #"neuron-gpu-nightly>=9.0a0",
     #"neuron-nightly>=9.0a1.dev1492",
-    "noise>=1.2.2",
     "numba>=0.61.2",
     "numpy>=1.26,<2.0",
     "optuna>=4.5.0",
-    "pyqt6>=6.9.0",
     "quantities>=0.16.2",
     "scienceplots>=2.1.1",
     "scikit-fmm>=2025.1.29",
     "scikit-learn>=1.6.1",
     "scipy>=1.15.3",
-    "scipy-stubs==1.16.1.0",
     "seaborn>=0.13.2",
     "setuptools>=80.9.0",
     "tqdm>=4.67.1",