mxnorm: An R Package to Normalize Multiplexed Imaging Data.

Area of Science:

• Single-cell biology
• Computational pathology
• Bioinformatics

Background:

• Multiplexed imaging assays (e.g., CODEX, MIBI, MxIF) provide high-resolution spatial information on cellular interactions in complex diseases.
• These assays generate massive datasets (terabytes), necessitating advanced imaging analysis pipelines.
• Technical variations, such as batch and slide effects, complicate data analysis and require robust normalization methods.

Purpose of the Study:

• To introduce mxnorm, an open-source R package for implementing, evaluating, and visualizing normalization techniques for multiplexed imaging data.
• To establish a foundation for evaluating multiplexed imaging normalization methods in R, addressing an understudied area.
• To provide a flexible framework for users to assess technical variability and normalization efficacy.

Main Methods:

• Development of mxnorm, an R package utilizing S3 methods for normalization technique evaluation.
• Implementation of a framework to assess normalization methods, including user-defined algorithms.
• Leveraging existing statistical methodologies for normalization and evaluation.

Main Results:

• mxnorm provides a standardized approach to evaluate normalization methods for multiplexed imaging.
• The package allows for the assessment of technical variability reduction and biological signal preservation.
• User-defined normalization and thresholding algorithms can be integrated for flexible analysis.

Conclusions:

• mxnorm offers a valuable tool for the multiplexed imaging community to address technical variability through normalization.
• The R package facilitates robust evaluation and comparison of diverse normalization strategies.
• This work promotes standardized and reproducible analysis of complex multiplexed imaging data.