Learn Starplast step by step. Each tutorial comes as a GUI walkthrough -- the application's own panels, with every setting and console line explained on hover -- and as a notebook doing the same from Python, with every line explained on hover and its real output. The notebooks are also provided as runnable .ipynb files.
Find a gene, color the map, read the evidence, and look at the tables from Python.
GUINotebook.ipynbStrategy 01: hide a label and everything restating it, walk map settings, keep the map whose clusters recover it.
GUINotebook.ipynbStrategies 02, 20, 24 and 25: a gene list in, a cluster, a ranking and a profile out.
GUINotebook.ipynbSelf-tests, nulls, noise tables and the calibration of every strategy over its settings.
GUINotebook.ipynbSix silent screen captures of the real application, one to two minutes each. Every panel, number and table in them was produced by the program while the clip was being recorded; the captions are burned in, so there is nothing to listen to. Click a poster to play.
The find box, the evidence panel, and the rule that a dash is not a zero.
The guided tab: what you have, which genes, which label, what you want to know -- then the ranked strategies and Run.
A strategy card, its four bars, the hold-out test, and the verdict with its chance level.
The pregenerated gallery, sorted by how well a label maps onto each, and the score table that says so.
The star map: one gene's measured links by source, widened to its neighbourhood and re-centred on a neighbour.
Question 13 of instruction 59: which never-published proteins get a confident compartment call, from the question to the named genes.
Every feature of Starplast in one long page: the window, finding genes and reading evidence, coloring and edges, every tab of the analysis panel with every control explained, the gallery and annotations, the guided workflows, your own data, the Strategies tab, the two organisms, preferences, the Python API, and how to read the numbers honestly.