メニュー
AI-powered RNA velocity analysis for predicting cellular state transitions, differentiation trajectories, and dynamic gene regulation from single-cell RNA sequencing data.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills- --skill rna-velocity-agentコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
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.
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.
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.
SOC 職業分類に基づく
SKILL.md を表示中
| name | rna-velocity-agent |
| description | AI-powered RNA velocity analysis for predicting cellular state transitions, differentiation trajectories, and dynamic gene regulation from single-cell RNA sequencing data. |
The RNA Velocity Agent analyzes RNA velocity from single-cell RNA sequencing to predict cellular state transitions, differentiation trajectories, and dynamic transcriptional regulation. It implements velocyto, scVelo, and deep learning approaches for trajectory inference.
Splicing-Based Velocity: Calculate RNA velocity from spliced/unspliced transcript ratios.
Dynamic Modeling: Deep learning models (scVelo dynamical mode) for accurate velocity estimation.
Trajectory Inference: Project velocity vectors onto UMAP/PCA for differentiation flow visualization.
Driver Gene Identification: Identify genes driving cell state transitions.
Latent Time Estimation: Reconstruct cellular pseudotime from velocity fields.
Multi-Modal Velocity: Integrate protein (CITE-seq) or chromatin (ATAC) velocity.
Transcription → Unspliced RNA → Splicing → Spliced (mature) mRNA → Degradation
Velocity = d[spliced]/dt = β[unspliced] - γ[spliced]
- Positive velocity: Gene upregulating
- Negative velocity: Gene downregulating
- Zero velocity: Steady state
Input: scRNA-seq data with spliced/unspliced counts (from STARsolo, velocyto, kallisto-bustools).
Quality Control: Filter genes by splice detection rates and expression levels.
Velocity Computation: Calculate velocity using steady-state or dynamical models.
Embedding Projection: Project velocity onto low-dimensional representations.
Trajectory Analysis: Identify root cells, terminal states, and differentiation paths.
Driver Analysis: Rank genes by velocity-based contribution to transitions.
Output: Velocity vectors, trajectory plots, driver genes, latent time estimates.
User: "Analyze RNA velocity in this hematopoiesis scRNA-seq dataset to map differentiation trajectories."
Agent Action:
python3 Skills/Genomics/RNA_Velocity_Agent/velocity_analyzer.py \
--adata hematopoiesis.h5ad \
--spliced_layer spliced \
--unspliced_layer unspliced \
--model dynamical \
--n_top_genes 2000 \
--identify_roots true \
--output velocity_results/
| Model | Method | Best For | Limitations |
|---|---|---|---|
| velocyto (steady-state) | Linear regression | Quick overview | Assumes equilibrium |
| scVelo stochastic | Moment-based | General use | Limited dynamics |
| scVelo dynamical | Likelihood-based | Complex trajectories | Computationally intensive |
| UniTVelo | Deep learning | Multi-lineage | Training requirements |
| veloVI | Variational inference | Uncertainty quantification | Complex |
1. Root Cell Identification
2. Terminal State Detection
3. Driver Gene Analysis
4. Latent Time
| Metric | Threshold | Interpretation |
|---|---|---|
| Fraction unspliced | >10% of reads | Adequate capture |
| Genes with velocity | >1000 genes | Sufficient coverage |
| Velocity confidence | >0.8 | Reliable estimates |
| Coherence score | >0.3 | Consistent trajectories |
Cell Cycle Analysis:
Perturbation Effects:
Disease Dynamics:
AI Group - Biomedical AI Platform