Profiling Cell Type Abundance and Expression in Bulk Tissues with CIBERSORTx
Chloé B Steen1, Chih Long Liu1,2, Ash A Alizadeh3,4,5,6
1Division of Oncology, Department of Medicine, Stanford Cancer Institute, Stanford University, Stanford, CA, USA.
Methods in Molecular Biology (Clifton, N.J.)
|January 22, 2020
Summary
CIBERSORTx is a machine learning tool that analyzes gene expression data to identify different cell types and their abundance in tissues. This method allows for high-throughput cell profiling without physical cell isolation.
Area of Science:
- Computational biology
- Bioinformatics
- Genomics
Background:
- Bulk tissue transcriptome profiles contain complex cellular information.
- Distinguishing cell type-specific gene expression is crucial for understanding tissue function and disease.
- Current methods often require physical cell isolation, which can be laborious and introduce biases.
Purpose of the Study:
- To provide a comprehensive primer on CIBERSORTx, a machine learning framework.
- To demonstrate the capabilities of CIBERSORTx for assessing cellular abundance and gene expression patterns.
- To showcase its application in high-throughput profiling of cell types and states in normal and neoplastic tissues.
Main Methods:
- Utilizing CIBERSORTx, a suite of machine learning tools.
- Learning molecular signatures of distinct cell types from single-cell or bulk-sorted RNA sequencing data.
- Applying learned signatures to bulk tissue transcriptomes to characterize cellular heterogeneity without physical cell isolation.
Main Results:
- CIBERSORTx enables the assessment of cellular abundance from transcriptome data.
- It facilitates the identification of cell type-specific gene expression patterns.
- The framework allows for repeated application to characterize cellular heterogeneity in various tissue types.
Conclusions:
- CIBERSORTx is a powerful tool for deconvoluting cellular composition from bulk tissue transcriptomes.
- It offers a high-throughput and efficient method for cell type profiling.
- This approach is valuable for studying cellular heterogeneity in both normal and diseased tissues.
Keywords:
Cellular heterogeneityDeconvolutionDigital cytometryGene expressionTumor microenvironmentscRNA-seq
