Toy corpus tutorial#

This tutorial shows the intended idea of litmap without bundling real corpora.

Goal#

Use a tiny literature-like corpus to explain the difference between analysis space and display space.

Example corpus#

papers = [
    {"id": "P1", "title": "Checkpoint blockade biomarkers", "abstract": "PD-L1 and T cell infiltration"},
    {"id": "P2", "title": "Tumor microenvironment profiling", "abstract": "macrophage and stromal context"},
    {"id": "P3", "title": "ADC linker payload strategy", "abstract": "linker stability and payload release"},
]

Interpretation#

Even in a toy corpus, the package should keep the conceptual rule clear:

  • the analytical representation decides what is near what

  • the plotting representation exists only to make the result readable

This tutorial stops at the conceptual level on purpose. That keeps the repository small and avoids shipping heavy artifacts.