TMAP API Reference

This is a practical reference for the public API. For a guided workflow, start with molecule_tutorial.md.

`TMAP`

High-level estimator for building a tree-shaped 2D map.

Constructor

TMAP(
    n_neighbors=20,
    metric="jaccard",
    n_permutations=512,
    kc=50,
    seed=42,
    layout_iterations=1000,
    layout_config=None,
    store_index=False,
)

Key Methods

Method	What it does
`fit(X)`	Build the graph, tree, and 2D embedding
`fit_transform(X)`	Fit and return `(x, y, s, t)`
`transform(X_new)`	Place new points on the existing map without changing the model
`add_points(X_new)`	Add new points into the fitted model
`to_tmapviz()`	Create a `TmapViz` object for notebook or HTML output
`to_html(path)`	Write an HTML visualization
`serve(port=8050)`	Start a local HTTP server for the visualization
`save(path)` / `load(path)`	Save or load the fitted model

Key Attributes

Attribute	Meaning
`embedding_`	2D coordinates, shape `(n, 2)`
`tree_`	Tree extracted from the kNN graph
`graph_`	k-nearest-neighbor graph
`lsh_forest_`	Jaccard search index
`index_`	Dense ANN index when `store_index=True`

Metrics

Metric	Input	Backend
`jaccard`	Binary matrix	MinHash + LSHForest
`cosine`	Dense float matrix	USearch
`euclidean`	Dense float matrix	USearch
`precomputed`	Distance matrix	Direct graph construction

`TmapViz`

Visualization object returned by model.to_tmapviz().

Common Methods

Method	What it does
`add_color_layout(name, values, ...)`	Add a colorable column
`add_label(name, values)`	Add a tooltip column
`add_smiles(values)`	Add molecule structures to tooltips
`to_widget(...)`	Build a Jupyter widget
`write_html(path)`	Write one self-contained HTML file
`write_static(path)`	Write static assets for hosting
`serve(port=8050)`	Start a local HTTP server

Minimal Example

viz = model.to_tmapviz()
viz.title = "My Map"
viz.add_color_layout("Score", scores.tolist(), color="viridis")
viz.add_label("Name", names)
viz.write_html("my_map.html")

Chemistry Helpers

These live in tmap.utils.

Function	What it does
`fingerprints_from_smiles(smiles, fp_type="morgan", ...)`	Build fingerprints from SMILES
`molecular_properties(smiles, properties=None)`	Compute RDKit properties
`murcko_scaffolds(smiles)`	Compute Murcko scaffold strings

Chemistry Example

from tmap.utils import fingerprints_from_smiles, molecular_properties

fps = fingerprints_from_smiles(smiles, fp_type="morgan", radius=2, n_bits=2048)
props = molecular_properties(smiles, properties=["mw", "logp", "n_rings"])

Power-User Layout API

If you need lower-level control, these functions are still available:

Function	Use it when
`layout_from_lsh_forest(...)`	You want the classic MinHash + LSH workflow directly
`layout_from_knn_graph(...)`	You already have a `KNNGraph`
`tree_from_knn_graph(...)`	You want a `Tree` from a custom kNN graph
`layout_from_edge_list(...)`	You already have weighted edges

For the parameter details, see layout_guide.md.