This repository serves as a comprehensive catalog of algorithms for estimating pseudotime in single-cell RNA-sequencing (scRNA-seq) data, also known as trajectory inference or cell ordering. It targets researchers and bioinformaticians working with scRNA-seq data who need to understand the landscape of available methods for analyzing cellular differentiation or dynamic processes. The primary benefit is providing a centralized, unopinionated list of methods, their software links, and associated publications, facilitating method selection and comparative analysis.

How It Works

The project functions as a curated list, detailing various algorithms used for single-cell trajectory inference. These methods typically involve reducing the dimensionality of gene expression data, then identifying a smooth progression through this low-dimensional space. Approaches include graph algorithms (e.g., minimum spanning trees), principal curves, and probabilistic models. Some methods incorporate prior knowledge like marker genes or capture times, while others focus on specific biological processes like branching or cyclic differentiation.

Quick Start & Requirements

This repository is a reference list and does not require installation or execution. It provides links to external software and publications for users to explore individual methods.

Highlighted Details

• Comprehensive catalog of over 200 pseudotime inference and RNA velocity estimation algorithms.
• Includes method names, software links, and manuscript citations for each entry.
• Covers a wide range of algorithmic approaches, from graph-based methods to probabilistic models and deep learning.
• Lists related methods for RNA velocity and specialized trajectory analysis.

Maintenance & Community

The initial list was created by Anthony Gitter, with contributions welcomed via pull requests. The repository is a static reference, with no active development or community channels mentioned.

Licensing & Compatibility

The repository itself is licensed under the MIT License, allowing for broad use and modification. However, users must adhere to the licenses of the individual software packages and publications linked within the repository.

Limitations & Caveats

This repository is a passive catalog and does not provide any benchmarking, comparative analysis, or commentary on the methods listed. Users must independently evaluate the suitability and performance of each algorithm for their specific research needs.