ceLLama is an automation pipeline for cell type annotations utilizing large-language models (LLMs). The program offers advantages such as privacy, comprehensive analysis, speed, and extensive reporting. It operates locally to prevent data leaks, considers negative genes, processes efficiently, and generates customized reports. Ideal for quick cell type checks, ceLLama requires the installation of Ollama and the selection of a preferred model. The process involves annotating cell types, identifying cluster markers, running ceLLama, transferring labels, and creating detailed reports. Users can customize annotations by adjusting temperature and base prompts. It is important to note that LLMs can make mistakes, so it is advised to check information thoroughly.
https://github.com/CelVoxes/ceLLama