# kochgroup/famdlike architecture

> Architecture diagram of the GitHub repository kochgroup/famdlike, 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/kochgroup/famdlike
- Repository: https://github.com/kochgroup/famdlike

## Overview

FamDLike is a family-specific molecular drug-likeness scorer. Its main workflow accepts molecular records, optionally derives RDKit descriptors from SMILES, then produces family desirability scores and optional GAM and tree-ensemble probabilities for a composite score and class prediction. Separate experiment pipelines train the three scoring components. The README documents their roles and outputs; sampled source confirms descriptor calculation and model interfaces, but does not expose every inference call or artifact-loading detail, so those links are cited to documentation where appropriate.

Main entry points: [`main.py`](https://github.com/kochgroup/famdlike/blob/main/main.py) (Prediction API / CLI), [`desirability_functions/optimizer.py`](https://github.com/kochgroup/famdlike/blob/main/desirability_functions/optimizer.py) (Property distribution fitting), [`desirability_functions/desirability_functions.py`](https://github.com/kochgroup/famdlike/blob/main/desirability_functions/desirability_functions.py) (Desirability functions), [`desirability_functions/des_engine.py`](https://github.com/kochgroup/famdlike/blob/main/desirability_functions/des_engine.py) (FDS engine training)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_prediction["Prediction interface"]
  node_predict["Prediction API / CLI<br/>[main.py]"]
  node_settings["Feature configuration<br/>[config.py]"]
end

subgraph group_molecules["Molecular features"]
  node_descriptor_calc["RDKit descriptors<br/>[calc_desc.py]"]
end

subgraph group_fds["Family desirability"]
  node_fds_train["FDS engine training<br/>[des_engine.py]"]
  node_distribution_fit["Property distribution fitting<br/>[optimizer.py]"]
  node_desirability["Desirability functions"]
  node_des_fitting["Fitting utilities<br/>[des_fitting.py]"]
end

subgraph group_gam["GAM scoring"]
  node_gam_train["GAM training<br/>[gam_training.py]"]
  node_gam_model["GAM model<br/>[gam_model.py]"]
end

subgraph group_trees["Tree ensemble"]
  node_tree_train["Tree model training"]
  node_tree_trainer["Tree classifiers<br/>[treemodels.py]"]
  node_tree_utils["Training feature filters<br/>[utils.py]"]
end

node_user(("Molecule user"))
node_results["Scored results"]

node_user -->|"submits molecules"| node_predict
node_predict -->|"loads features"| node_settings
node_predict -->|"calculates descriptors"| node_descriptor_calc
node_predict -->|"scores FDS"| node_desirability
node_predict -->|"predicts probabilities"| node_gam_model
node_predict -->|"uses ensemble"| node_tree_trainer
node_predict -->|"returns scores"| node_results
node_fds_train -->|"fits distributions"| node_distribution_fit
node_fds_train -->|"builds engine"| node_desirability
node_distribution_fit -->|"configures functions"| node_desirability
node_des_fitting -->|"uses fitting"| node_distribution_fit
node_gam_train -->|"fits GAM"| node_gam_model
node_tree_train -->|"trains classifiers"| node_tree_trainer
node_tree_train -->|"uses filters"| node_tree_utils

click node_predict "https://github.com/kochgroup/famdlike/blob/main/main.py"
click node_settings "https://github.com/kochgroup/famdlike/blob/main/config.py"
click node_descriptor_calc "https://github.com/kochgroup/famdlike/blob/main/calc_desc.py"
click node_fds_train "https://github.com/kochgroup/famdlike/blob/main/desirability_functions/des_engine.py"
click node_distribution_fit "https://github.com/kochgroup/famdlike/blob/main/desirability_functions/optimizer.py"
click node_desirability "https://github.com/kochgroup/famdlike/blob/main/desirability_functions/desirability_functions.py"
click node_des_fitting "https://github.com/kochgroup/famdlike/blob/main/desirability_functions/des_fitting.py"
click node_gam_train "https://github.com/kochgroup/famdlike/blob/main/gam/gam_training.py"
click node_gam_model "https://github.com/kochgroup/famdlike/blob/main/gam/gam_model.py"
click node_tree_train "https://github.com/kochgroup/famdlike/blob/main/tree_models/training_treemodels.py"
click node_tree_trainer "https://github.com/kochgroup/famdlike/blob/main/tree_models/treemodels.py"
click node_tree_utils "https://github.com/kochgroup/famdlike/blob/main/tree_models/utils.py"

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_predict,node_settings,node_user toneBlue
class node_descriptor_calc toneAmber
class node_fds_train,node_distribution_fit,node_desirability,node_des_fitting toneMint
class node_gam_train,node_gam_model toneRose
class node_tree_train,node_tree_trainer,node_tree_utils,node_results toneIndigo
```

## Components

### Prediction interface

- [Prediction API / CLI](https://github.com/kochgroup/famdlike/blob/main/main.py): `main.py`
- [Feature configuration](https://github.com/kochgroup/famdlike/blob/main/config.py): `config.py`

### Molecular features

- [RDKit descriptors](https://github.com/kochgroup/famdlike/blob/main/calc_desc.py): `calc_desc.py`

### Family desirability

- [FDS engine training](https://github.com/kochgroup/famdlike/blob/main/desirability_functions/des_engine.py): `desirability_functions/des_engine.py`
- [Property distribution fitting](https://github.com/kochgroup/famdlike/blob/main/desirability_functions/optimizer.py): `desirability_functions/optimizer.py`
- [Desirability functions](https://github.com/kochgroup/famdlike/blob/main/desirability_functions/desirability_functions.py): `desirability_functions/desirability_functions.py`
- [Fitting utilities](https://github.com/kochgroup/famdlike/blob/main/desirability_functions/des_fitting.py): `desirability_functions/des_fitting.py`

### GAM scoring

- [GAM training](https://github.com/kochgroup/famdlike/blob/main/gam/gam_training.py): `gam/gam_training.py`
- [GAM model](https://github.com/kochgroup/famdlike/blob/main/gam/gam_model.py): `gam/gam_model.py`

### Tree ensemble

- [Tree model training](https://github.com/kochgroup/famdlike/blob/main/tree_models/training_treemodels.py): `tree_models/training_treemodels.py`
- [Tree classifiers](https://github.com/kochgroup/famdlike/blob/main/tree_models/treemodels.py): `tree_models/treemodels.py`
- [Training feature filters](https://github.com/kochgroup/famdlike/blob/main/tree_models/utils.py): `tree_models/utils.py`

### Other components

- **Molecule user**
- **Scored results**

## Connections

- Molecule user → Prediction API / CLI: submits molecules (evidence: `README.md`)
- Prediction API / CLI → Feature configuration: loads features (evidence: `main.py`)
- Prediction API / CLI → RDKit descriptors: calculates descriptors (evidence: `main.py`)
- Prediction API / CLI → Desirability functions: scores FDS (evidence: `README.md`)
- Prediction API / CLI → GAM model: predicts probabilities (evidence: `README.md`)
- Prediction API / CLI → Tree classifiers: uses ensemble (evidence: `README.md`)
- Prediction API / CLI → Scored results: returns scores (evidence: `README.md`)
- FDS engine training → Property distribution fitting: fits distributions (evidence: `desirability_functions/des_engine.py`)
- FDS engine training → Desirability functions: builds engine (evidence: `desirability_functions/des_engine.py`)
- Property distribution fitting → Desirability functions: configures functions (evidence: `desirability_functions/optimizer.py`)
- Fitting utilities → Property distribution fitting: uses fitting (evidence: `desirability_functions/des_fitting.py`)
- GAM training → GAM model: fits GAM (evidence: `README.md`)
- Tree model training → Tree classifiers: trains classifiers (evidence: `tree_models/training_treemodels.py`)
- Tree model training → Training feature filters: uses filters (evidence: `tree_models/training_treemodels.py`)
