C2S-Pythia-410m Open-source Cell Type Prediction Model -Precise Analysis of Single-cell RNA Sequencing Data

C2S Pythia 410m Cell Type Prediction

Developed by vandijklab

A cell type prediction model based on the Pythia-410m architecture, fine-tuned by converting scRNA-seq data into 'cell sentences' through the Cell2Sentence method, focusing on cell type prediction in single-cell RNA sequencing data.

Large Language Model

Transformers

English#Single-cell RNA sequencing #Cell type prediction #Gene ranking sentence

Downloads 266

Release Time : 9/3/2024

Model Overview

This model leverages the natural language processing capabilities of large language models, transforming single-cell RNA sequencing data into sequences of gene names via the innovative Cell2Sentence method to achieve cell type prediction.

Model Features

Cell2Sentence Innovative Method

Converts scRNA-seq data into sequences of gene names sorted by expression levels ('cell sentences'), enabling LLMs to adapt to single-cell biology research.

Large-scale Training Data

Trained on over 57 million human and mouse cells from more than 800 single-cell RNA sequencing datasets, covering various tissues and cell types.

Cross-species Applicability

Training data includes human and mouse cells, potentially enabling cross-species cell type prediction.

Model Capabilities

Single-cell RNA sequencing data analysis

Cell type prediction

Gene expression pattern recognition

Use Cases

Biomedical Research

Cell Atlas Construction

Helps researchers quickly classify and annotate cell types in large-scale single-cell sequencing data

Accelerates the cell annotation process for projects like the Human Cell Atlas

Disease Research

Identifies abnormal cell types or states in disease samples

May discover new disease-related cell subpopulations

🚀 C2S-Pythia-410m Cell Type Prediction Model

This model is based on the Pythia-410m architecture, fine - tuned for cell type prediction in single - cell biology using Cell2Sentence on scRNA - seq datasets.

🚀 Quick Start

This section provides a high - level introduction to the C2S - Pythia - 410m cell - type - prediction model, its underlying architecture, and the innovative approach used for fine - tuning.

✨ Features

Base Architecture: Built on the Pythia - 410m architecture developed by EleutherAI.
Fine - Tuning Approach: Utilizes Cell2Sentence (C2S) to transform scRNA - seq data into "cell sentences", enabling large language models to handle single - cell tasks.
Task Focus: Specifically designed for cell type prediction based on scRNA - seq data.

📦 Installation

No installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

No code examples are provided in the original document, so this section is skipped.

📚 Documentation

Overview

This is the C2S - Pythia - 410m cell - type - prediction model, based on the Pythia - 410m architecture developed by EleutherAI, fine - tuned using Cell2Sentence (C2S) on a diverse set of single - cell RNA sequencing (scRNA - seq) datasets from CellxGene and the Human Cell Atlas. Cell2Sentence is an innovative approach for adapting large language models (LLMs) to single - cell biology by transforming scRNA - seq data into "cell sentences"—sequences of gene names ordered by expression levels. This transformation enables LLMs to leverage their natural language processing capabilities for various single - cell tasks, with a focus on cell type prediction in this model.

Training Data

This model was trained on over 57 million human and mouse cells gathered from over 800 single - cell RNA sequencing datasets from CellxGene and the Human Cell Atlas. This dataset covers a broad range of cell types and conditions from multiple tissues in both human and mouse.

This model was trained with the top 200 genes per cell sentence.

Tasks

This model is designed for:

Cell type prediction: Predicting the cell type based on the "cell sentence" generated from scRNA - seq data.

🔧 Technical Details

No in - depth technical implementation details (more than 50 words) are provided in the original document, so this section is skipped.

📄 License

No license information is provided in the original document, so this section is skipped.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご