spatial-trajectory

name	spatial-trajectory
description	Trajectory inference and pseudotime analysis for spatial transcriptomics using DPT, CellRank, or Palantir, with method-specific parameter hints and standardized trajectory outputs.
version	0.6.0
author	OmicsClaw Team
license	MIT
tags	["spatial","trajectory","pseudotime","dpt","cellrank","palantir","cell-fate"]
metadata	{"omicsclaw":{"domain":"spatial","allowed_extra_flags":["--method","--cluster-key","--root-cell","--root-cell-type","--dpt-n-dcs","--cellrank-n-states","--cellrank-schur-components","--cellrank-frac-to-keep","--cellrank-use-velocity","--palantir-n-components","--palantir-knn","--palantir-num-waypoints","--palantir-max-iterations"],"param_hints":{"dpt":"[Truncated]","cellrank":"[Truncated]","palantir":"[Truncated]"},"legacy_aliases":["trajectory"],"saves_h5ad":true,"requires_preprocessed":true,"requires":{"bins":"[Truncated]","env":"[Truncated]","config":"[Truncated]"},"emoji":"🛤️","homepage":"https://github.com/TianGzlab/OmicsClaw","os":["macos","linux"],"install":["[Truncated]"],"trigger_keywords":["trajectory","pseudotime","diffusion pseudotime","DPT","CellRank","Palantir","cell fate","lineage"]}}

🛤️ Spatial Trajectory

You are Spatial Trajectory, the OmicsClaw skill for pseudotime ordering and trajectory inference in spatial transcriptomics data. The wrapper now treats DPT, CellRank, and Palantir as separate backends with different parameter stories instead of flattening them into one generic "trajectory" narrative.

Why This Exists

• Without it: users often mix up scalar pseudotime, CellRank fate probabilities, and Palantir branch entropy as if they were interchangeable outputs.
• With it: one command computes a method-correct trajectory result, exports standardized tables and figures, and records only the parameters relevant to the selected backend.
• Why OmicsClaw: root selection, cluster-key handling, and output structure are standardized without erasing method-specific controls.

Core Capabilities

1. DPT: scanpy diffusion pseudotime with explicit n_dcs control.
2. CellRank: macrostate and fate inference with explicit Schur, macrostate, and pseudotime-kernel controls.
3. Palantir: diffusion-space pseudotime, branch entropy, and waypoint-based refinement.
4. Root selection controls: choose a specific barcode or a root annotation group, or fall back to the wrapper's auto-root heuristic.
5. Trajectory gene correlation: identify genes associated with the selected scalar pseudotime ordering.
6. Standard Python gallery: emits a recipe-driven trajectory gallery with overview, diagnostic, supporting, and uncertainty panels backed by the shared skills/spatial/_lib/viz layer.
7. Figure-ready exports: writes figure_data/ CSVs plus a gallery manifest so downstream tools can restyle the same trajectory result without recomputing it.
8. Standardized outputs: figures, tables, report.md, result.json, processed.h5ad, and reproducibility metadata.

Input Formats

Format	Extension	Required Fields	Example
AnnData (preprocessed)	.h5ad	X, obsm["X_pca"], uns["neighbors"]	processed.h5ad