Reverse-engineering flow-cytometry gating strategies for phenotypic labelling and high-performance cell sorting.

Area of Science:

• Immunology
• Computational Biology
• Bioinformatics

Background:

• High-dimensional cytometry datasets (20-40 dimensions, ~1 million cells) present challenges in identifying novel cell populations.
• Optimizing gating strategies for cell identification becomes exponentially more complex with increasing data dimensionality.

Purpose of the Study:

• To develop an algorithm, Hypergate, that identifies optimized gating strategies for specific cell populations.
• To improve the yield and purity of cell population identification in high-dimensional datasets.

Main Methods:

• Developed Hypergate, an R-implemented algorithm for automated gating strategy optimization.
• Evaluated Hypergate against human experts, Support Vector Machines, and Random Forests on public datasets.
• Applied Hypergate to identify innate lymphoid cells, comparing its strategies to established methods.

Main Results:

• Hypergate demonstrated superior yield and purity compared to human experts and machine learning methods.
• Revisited innate lymphoid cell identification, yielding concise and efficient gating strategies with higher performance than current standards.
• Phenotypic descriptions from Hypergate were consistent with existing knowledge and sparser than a competing method.

Conclusions:

• Hypergate effectively optimizes gating strategies for high-dimensional cytometry data, enhancing cell population identification.
• The algorithm offers a more efficient and accurate approach for analyzing complex biological datasets.
• Hypergate provides a valuable tool for researchers in immunology and computational biology.