# panchenko-lab/nupose architecture

> Architecture diagram of the GitHub repository panchenko-lab/nupose, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-01). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/panchenko-lab/nupose
- Repository: https://github.com/panchenko-lab/nupose

## Overview

NuPoSe is a nucleosome-positioning research and prediction framework. Its documented research workflow goes from MNase-seq dyad identification and DNA-set construction through sequence-feature extraction, feature selection, and ResNet-based training. It also distributes two standalone prediction models. The ZIP contents were not sampled, so their internal components and connections are represented only at archive level; workflow links follow the README rather than inspected archive code.

Main entry points: [`FeatureExtraction.py`](https://github.com/panchenko-lab/nupose/blob/main/FeatureExtraction.py) (Feature extraction), [`Trader.py`](https://github.com/panchenko-lab/nupose/blob/main/Trader.py) (Feature selection), [`Res.py`](https://github.com/panchenko-lab/nupose/blob/main/Res.py) (ResNet training), [`NuPoSe.zip`](https://github.com/panchenko-lab/nupose/blob/main/NuPoSe.zip) (NuPoSe model)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_data["Data preparation"]
  node_mnase["MNase processing<br/>[DyadMNase.zip]"]
  node_dna_sets["DNA datasets<br/>[DyadMNase.zip]"]
end

subgraph group_analysis["Feature analysis"]
  node_extract["Feature extraction"]
  node_feature_data["Feature datasets"]
  node_select["Feature selection<br/>[Trader.py]"]
  node_selected_sets["Feature subsets"]
end

subgraph group_models["Prediction models"]
  node_train["ResNet training<br/>[Res.py]"]
  node_trained_model["Trained model"]
  node_nupose["NuPoSe model<br/>[NuPoSe.zip]"]
  node_star["NuPoSe* model<br/>[NuPoSeStar.zip]"]
end

node_user(("Researcher"))
node_input["201-base sequences"]
node_output["Prediction scores"]

node_user -->|"runs preprocessing"| node_mnase
node_mnase -->|"produces DNA sets"| node_dna_sets
node_user -->|"runs extraction"| node_extract
node_dna_sets -->|"provides sequences"| node_extract
node_extract -->|"writes features"| node_feature_data
node_user -->|"runs selection"| node_select
node_feature_data -.->|"provides features"| node_select
node_select -->|"generates subsets"| node_selected_sets
node_user -->|"runs training"| node_train
node_selected_sets -->|"provides subsets"| node_train
node_train -->|"saves model"| node_trained_model
node_user -->|"runs prediction"| node_nupose
node_user -->|"runs prediction"| node_star
node_input -->|"supplies sequences"| node_nupose
node_input -->|"supplies sequences"| node_star
node_nupose -->|"writes scores"| node_output
node_star -->|"writes scores"| node_output

click node_mnase "https://github.com/panchenko-lab/nupose/blob/main/DyadMNase.zip"
click node_dna_sets "https://github.com/panchenko-lab/nupose/blob/main/DyadMNase.zip"
click node_extract "https://github.com/panchenko-lab/nupose/blob/main/FeatureExtraction.py"
click node_select "https://github.com/panchenko-lab/nupose/blob/main/Trader.py"
click node_train "https://github.com/panchenko-lab/nupose/blob/main/Res.py"
click node_nupose "https://github.com/panchenko-lab/nupose/blob/main/NuPoSe.zip"
click node_star "https://github.com/panchenko-lab/nupose/blob/main/NuPoSeStar.zip"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_mnase,node_dna_sets toneBlue
class node_extract,node_feature_data,node_select,node_selected_sets toneAmber
class node_train,node_trained_model,node_nupose,node_star toneMint
class node_user,node_input,node_output toneIndigo
```

## Components

### Data preparation

- [MNase processing](https://github.com/panchenko-lab/nupose/blob/main/DyadMNase.zip): `DyadMNase.zip`
- [DNA datasets](https://github.com/panchenko-lab/nupose/blob/main/DyadMNase.zip): `DyadMNase.zip`

### Feature analysis

- [Feature extraction](https://github.com/panchenko-lab/nupose/blob/main/FeatureExtraction.py): `FeatureExtraction.py`
- **Feature datasets**
- [Feature selection](https://github.com/panchenko-lab/nupose/blob/main/Trader.py): `Trader.py`
- **Feature subsets**

### Prediction models

- [ResNet training](https://github.com/panchenko-lab/nupose/blob/main/Res.py): `Res.py`
- **Trained model**
- [NuPoSe model](https://github.com/panchenko-lab/nupose/blob/main/NuPoSe.zip): `NuPoSe.zip`
- [NuPoSe* model](https://github.com/panchenko-lab/nupose/blob/main/NuPoSeStar.zip): `NuPoSeStar.zip`

### Other components

- **Researcher**
- **201-base sequences**
- **Prediction scores**

## Connections

- Researcher → MNase processing: runs preprocessing (evidence: `README.md`)
- MNase processing → DNA datasets: produces DNA sets (evidence: `README.md`)
- Researcher → Feature extraction: runs extraction (evidence: `README.md`)
- DNA datasets → Feature extraction: provides sequences (evidence: `README.md`)
- Feature extraction → Feature datasets: writes features (evidence: `FeatureExtraction.py`)
- Researcher → Feature selection: runs selection (evidence: `README.md`)
- Feature datasets → Feature selection: provides features
- Feature selection → Feature subsets: generates subsets (evidence: `README.md`)
- Researcher → ResNet training: runs training (evidence: `README.md`)
- Feature subsets → ResNet training: provides subsets (evidence: `README.md`)
- ResNet training → Trained model: saves model (evidence: `Res.py`)
- Researcher → NuPoSe model: runs prediction (evidence: `README.md`)
- Researcher → NuPoSe* model: runs prediction (evidence: `README.md`)
- 201-base sequences → NuPoSe model: supplies sequences (evidence: `README.md`)
- 201-base sequences → NuPoSe* model: supplies sequences (evidence: `README.md`)
- NuPoSe model → Prediction scores: writes scores (evidence: `README.md`)
- NuPoSe* model → Prediction scores: writes scores (evidence: `README.md`)
