Protocol for deep-learning-driven cell type label transfer in single-cell RNA sequencing data.

Publication Year:

2025

Authors:

Zoe Zabetian, Jesus Gonzalez-Ferrer, Julian Lehrer, Vanessa D Jonsson, Mircea Teodorescu, David Haussler, Mohammed A Mostajo-Radji

PubMed ID:

40232935

Funding Grants:

Deep phenotyping of human brain organoid models of autism spectrum disorder to unravel disease heterogeneity and develop biomarkers and treatments

Public Summary:

Identifying cell types from single cell sequencing data of complex cell cultures is often slow and prone to error. This protocol provides a standardized, deep-learning method to automatically label cells based on their genetic signatures. By streamlining how researchers identify cells across different experiments, this tool provides speed and accuracy in identifying cells in brain organoids and ensures that lab-grown tissues accurately mimic the human body for drug testing and disease modeling.

Scientific Abstract:

Here, we present a protocol for using SIMS (scalable, interpretable machine learning for single cell) to transfer cell type labels in single-cell RNA sequencing data. This protocol outlines data preparation, model training with labeled data or inference using pretrained models, and methods for visualizing, downloading, and interpreting predictions. We provide stepwise instructions for accessing SIMS through the application programming interface (API), GitHub Codespaces, and a web application. For complete details on the use and execution of this protocol, please refer to Gonzalez-Ferrer et al.(1).