sc-velocity-prep

Start here for RNA velocity when you have Cell Ranger BAM, loom, or STARsolo Velocyto output. Creates the spliced and unspliced layers needed by scVelo.

Jump to install

Repository: mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills-
Last source activity: April 30, 2026 at 21:29
Detected SKILL.md language: English
Stars: 31
Forks: 8

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

A direct command skips the review prompt. Inspect the source before running it.

npx skills add https://github.com/mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills- --skill sc-velocity-prep

The command stays on one line. Scroll horizontally to inspect it before copying.

Prefer a local copy? Download the files currently available to SkillsMP.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

SKILL.md

Source instructions · Read-only preview

Back to search Back to installation

medsam2-3d-segmentation

Operate MedSAM2 for promptable segmentation of 3D medical images and medical videos, including CT lesion propagation, MRI volumes, RECIST-guided prompts, efficient CPU-oriented variants, training, and 3D Slicer integration. Use when generating or validating volumetric masks from sparse prompts or propagating masks through image slices or video frames.

2026-06-18

monai-medical-imaging

Build reproducible healthcare imaging pipelines with Project MONAI for DICOM, NIfTI, pathology, and multidimensional imaging tasks including preprocessing, augmentation, training, sliding-window inference, evaluation, model bundles, labeling, and deployment. Use when implementing medical image classification, segmentation, registration, detection, generative, or foundation-model workflows in PyTorch.

2026-06-18

txgemma-therapeutics

Operate Google TxGemma prediction and chat models for therapeutic property prediction across small molecules, proteins, nucleic acids, diseases, targets, and cell lines. Use when formatting Therapeutics Data Commons tasks, choosing TxGemma model size or variant, running local or Model Garden inference, fine-tuning on private therapeutic data, or evaluating TxGemma in drug-discovery workflows.

2026-06-18

View all from this repository

🌀 Single-Cell Velocity Prep

You are SC Velocity Prep, a specialized OmicsClaw agent for producing velocity-ready spliced and unspliced layers before running sc-velocity.

Why This Exists

Without it: users have BAM files or STARsolo outputs but still need to manually convert them into loom or h5ad objects with velocity layers.

With it: one wrapper creates velocity_input.h5ad that is ready for the existing sc-velocity skill.

Why OmicsClaw: it keeps the heavy upstream extraction step separate from scVelo modeling while preserving an AnnData-first downstream contract.

Core Capabilities

Cell Ranger + velocyto path: generate a loom file from BAM and barcode whitelist, then import it.

STARsolo path: import existing STARsolo Velocyto matrices or run a velocity-oriented STARsolo job from FASTQ.

Optional merge path: add velocity layers into an existing base h5ad.

Standard output layer: writes processed.h5ad, a compatibility alias velocity_input.h5ad, summary figures, and figure-ready exports.

Reproducibility layer: writes README.md, report.md, result.json, and rerun commands.

Input Formats

Format

Extension

Required Fields / Structure

Example

Cell Ranger output

directory

contains outs/possorted_genome_bam.bam and filtered barcodes

sample_count/

STARsolo output

directory

contains Solo.out/Velocyto/... or spliced.mtx / unspliced.mtx

starsolo_pbmc/

Loom

.loom

velocyto loom file with spliced/unspliced layers

sample.loom

FASTQ directory

🌀 Single-Cell Velocity Prep

You are SC Velocity Prep, a specialized OmicsClaw agent for producing velocity-ready spliced and unspliced layers before running sc-velocity.

Why This Exists

Without it: users have BAM files or STARsolo outputs but still need to manually convert them into loom or h5ad objects with velocity layers.

With it: one wrapper creates velocity_input.h5ad that is ready for the existing sc-velocity skill.

Why OmicsClaw: it keeps the heavy upstream extraction step separate from scVelo modeling while preserving an AnnData-first downstream contract.

Core Capabilities

Cell Ranger + velocyto path: generate a loom file from BAM and barcode whitelist, then import it.

STARsolo path: import existing STARsolo Velocyto matrices or run a velocity-oriented STARsolo job from FASTQ.

Optional merge path: add velocity layers into an existing base h5ad.

Standard output layer: writes processed.h5ad, a compatibility alias velocity_input.h5ad, summary figures, and figure-ready exports.

Reproducibility layer: writes README.md, report.md, result.json, and rerun commands.

Input Formats

Format

Extension

Required Fields / Structure

Example

Cell Ranger output

directory

contains outs/possorted_genome_bam.bam and filtered barcodes

sample_count/

STARsolo output

directory

contains Solo.out/Velocyto/... or spliced.mtx / unspliced.mtx

starsolo_pbmc/

Loom

.loom

velocyto loom file with spliced/unspliced layers

sample.loom

FASTQ directory

name	sc-velocity-prep
description	Start here for RNA velocity when you have Cell Ranger BAM, loom, or STARsolo Velocyto output. Creates the spliced and unspliced layers needed by scVelo.
version	0.1.0
author	OmicsClaw
license	MIT
tags	["singlecell","scrna","velocity","velocyto","starsolo","scvelo"]
metadata	{"omicsclaw":{"domain":"singlecell","requires":{"bins":"[Truncated]","env":"[Truncated]","config":"[Truncated]"},"emoji":"🌀","homepage":"https://github.com/TianGzlab/OmicsClaw","os":["macos","linux"],"install":["[Truncated]"],"trigger_keywords":["RNA velocity prep","prepare spliced unspliced layers","velocyto","starsolo velocyto","velocity-ready AnnData"],"allowed_extra_flags":["--base-h5ad","--chemistry","--gtf","--method","--read2","--reference","--sample","--threads","--whitelist","--r-enhanced"],"legacy_aliases":["scrna-velocity-prep"],"saves_h5ad":true,"requires_preprocessed":false,"param_hints":{"velocyto":"[Truncated]","starsolo":"[Truncated]"}}}

name	sc-velocity-prep
description	Start here for RNA velocity when you have Cell Ranger BAM, loom, or STARsolo Velocyto output. Creates the spliced and unspliced layers needed by scVelo.
version	0.1.0
author	OmicsClaw
license	MIT
tags	["singlecell","scrna","velocity","velocyto","starsolo","scvelo"]
metadata	{"omicsclaw":{"domain":"singlecell","requires":{"bins":"[Truncated]","env":"[Truncated]","config":"[Truncated]"},"emoji":"🌀","homepage":"https://github.com/TianGzlab/OmicsClaw","os":["macos","linux"],"install":["[Truncated]"],"trigger_keywords":["RNA velocity prep","prepare spliced unspliced layers","velocyto","starsolo velocyto","velocity-ready AnnData"],"allowed_extra_flags":["--base-h5ad","--chemistry","--gtf","--method","--read2","--reference","--sample","--threads","--whitelist","--r-enhanced"],"legacy_aliases":["scrna-velocity-prep"],"saves_h5ad":true,"requires_preprocessed":false,"param_hints":{"velocyto":"[Truncated]","starsolo":"[Truncated]"}}}

Requirement	Where it should exist	Why it matters
Spliced/unspliced source	BAM + GTF, STARsolo Velocyto output, or loom	required to construct RNA velocity layers
Optional base object	`--base-h5ad`	useful when velocity layers should be merged back into an existing AnnData

Parameter	Default	Description
`--method`	`velocyto`	choose the Cell Ranger+velocyto path or the STARsolo Velocyto path
`--gtf`	none	required when running velocyto from Cell Ranger BAM
`--base-h5ad`	none	merge velocity layers into an existing AnnData
`--reference`	none	STAR genome directory for FASTQ-backed STARsolo runs
`--chemistry`	`auto`	STARsolo FASTQ runs require explicit supported chemistry
`--whitelist`	none	STARsolo barcode whitelist file

Flag	Type	Default	Description	Validation
`--input`	str	None	Cell Ranger output dir, STARsolo output dir, FASTQ path, or loom file	Required unless `--demo`
`--output`	str	—	Output directory	Required
`--demo`	flag	off	Run with built-in demo data	—
`--method`	str	`velocyto`	Velocity preparation backend	Choices: `velocyto`, `starsolo`
`--gtf`	str	None	GTF annotation file required for BAM-backed velocyto runs	Required when input is a Cell Ranger BAM directory
`--base-h5ad`	str	None	Existing AnnData to merge velocity layers into	Optional — useful to align velocity layers with a preprocessed object
`--reference`	str	None	STAR genome directory for FASTQ-backed STARsolo runs	Required for STARsolo FASTQ runs
`--sample`	str	None	Choose one sample from a multi-sample FASTQ directory	—
`--read2`	str	None	Explicit mate FASTQ when `--input` points to one file	—
`--threads`	int	8	Backend thread count	—
`--chemistry`	str	`auto`	STARsolo chemistry; real FASTQ runs require `10xv2`, `10xv3`, or `10xv4`	—
`--whitelist`	str	None	STARsolo barcode whitelist file	Strongly recommended for STARsolo real FASTQ runs
`--r-enhanced`	flag	off	Accepted for CLI consistency; no R Enhanced plots for this skill	No-op

sc-velocity-prep

Install options

Review the source files

🌀 Single-Cell Velocity Prep

Why This Exists

Core Capabilities

Input Formats

🌀 Single-Cell Velocity Prep

Why This Exists

Core Capabilities

Input Formats

Data / State Requirements

Beginner Setup Hint

Workflow

CLI Reference

Example Queries

Algorithm / Methodology

velocyto Path

STARsolo Path

Visualization Contract

Output Structure

Reproducibility Contract

Knowledge Companions

Dependencies

Environment Management

CLI Parameters

R Enhanced Plots

Special Requirements

Input Sources and Required Assets

GTF File

Merging into an Existing AnnData

Workflow Position

Input source	`--method`	Also needs
Cell Ranger BAM directory	`velocyto`	`--gtf` (matching annotation GTF)
Existing loom file	`velocyto`	— (imported directly)
STARsolo Velocyto output dir	`starsolo`	— (imported directly)
FASTQ directory (STARsolo)	`starsolo`	`--reference` (STAR genome dir), `--chemistry`, `--whitelist`

sc-velocity-prep

Source facts

Install options

Review the source files

More from this repository

🌀 Single-Cell Velocity Prep

Why This Exists

Core Capabilities

Input Formats

More from this repository

🌀 Single-Cell Velocity Prep

Why This Exists

Core Capabilities

Input Formats

Data / State Requirements

Beginner Setup Hint

Workflow

CLI Reference

Example Queries

Algorithm / Methodology

velocyto Path

STARsolo Path

Visualization Contract

Output Structure

Reproducibility Contract

Knowledge Companions

Dependencies

Environment Management

CLI Parameters

R Enhanced Plots

Special Requirements

Input Sources and Required Assets

GTF File

Merging into an Existing AnnData

Workflow Position