Tuning hyperparameters of doublet-detection methods for single-cell RNA sequencing data

TL;DR

This paper presents a hyperparameter tuning strategy for doublet-detection methods in single-cell RNA sequencing data, improving detection accuracy across diverse datasets and offering a general approach for optimizing computational methods.

Contribution

The study introduces a systematic hyperparameter tuning approach using factorial design and response surface modeling for doublet detection in scRNA-seq data, enhancing performance and generalizability.

Findings

Optimal hyperparameters improve detection accuracy across datasets

The tuning strategy is applicable to other doublet-detection methods

Insights into hyperparameter effects inform broader scRNA-seq analysis

Abstract

The existence of doublets in single-cell RNA sequencing (scRNA-seq) data poses a great challenge in downstream data analysis. Computational doublet-detection methods have been developed to remove doublets from scRNA-seq data. Yet, the default hyperparameter settings of those methods may not provide optimal performance. Here, we propose a strategy to tune hyperparameters for a cutting-edge doublet-detection method. We utilize a full factorial design to explore the relationship between hyperparameters and detection accuracy on 16 real scRNA-seq datasets. The optimal hyperparameters are obtained by a response surface model and convex optimization. We show that the optimal hyperparameters provide top performance across scRNA-seq datasets under various biological conditions. Our tuning strategy can be applied to other computational doublet-detection methods. It also offers insights into hyperparameter tuning for broader computational methods in scRNA-seq data analysis.