scrna-cell-type-annotator

What This Skill Does

The scrna-cell-type-annotator is a specialized OpenClaw skill designed to automate the labor-intensive process of cell type identification in single-cell RNA sequencing (scRNA-seq) datasets. By leveraging a structured marker gene database, this tool processes cluster-specific gene expression data to assign biological identities to distinct cell populations. It acts as an automated bioinformatician, reducing subjective bias by providing a clear, evidence-based annotation workflow that ensures reproducibility across different experiments. The skill utilizes a primary script, scripts/main.py, which is engineered to handle input validation, threshold-based assignment, and structured report generation. It is built for researchers who require consistent, audit-ready annotations for complex biological datasets.

Installation

To integrate this skill into your local environment, use the OpenClaw repository management tools. Execute the following command in your terminal:

clawhub install openclaw/skills/skills/aipoch-ai/scrna-cell-type-annotator

Ensure that you have Python 3.10+ installed and that the pandas library is available in your environment. Verification of the installation can be performed by running python -m py_compile scripts/main.py from within the skill directory.

Use Cases

• High-throughput analysis: Quickly annotating clusters across dozens of samples with consistent marker gene criteria.
• Preliminary data exploration: Generating initial cell type hypotheses during the exploratory phase of a study.
• Reproducibility assurance: Maintaining a fixed logic set for annotations that can be audited by peers or journals.
• Comparative genomics: Applying identical classification logic to disparate datasets to determine cell type composition differences between cohorts.

Example Prompts

1. "Analyze the cell clusters in data/sample_01.csv and annotate cell types using the standard marker gene set provided in the configuration."
2. "Perform auto-annotation on the provided scRNA data using the threshold settings defined in config/annotation_params.yaml, ensuring that all unassigned clusters are flagged for review."
3. "Run the cell type annotator on the processed_clusters.h5ad file and produce a summary report detailing the marker genes used for each identified cell type."

Tips & Limitations

• Data quality: Always ensure your input data is normalized and cleaned prior to running this tool, as noisy expression data can lead to inaccurate annotations.
• Bounded scope: This skill relies on predefined marker genes; if your dataset includes rare or highly specific cell types, you may need to supplement the default marker list.
• Threshold tuning: Always review the output reports. The confidence of the annotation is heavily dependent on the chosen thresholds; if too few genes match your markers, consider adjusting the sensitivity parameters in the configuration block.
• Audit your work: Always treat the output as a preliminary annotation that may require expert biological verification.