imputomics: web server and R package for missing values imputation in metabolomics data

Jarosław Chilimoniuk ¹, Krystyna Grzesiak ^1,2, Jakub Kała ¹

⁰Clinical Research Centre, Medical University of Białystok, Białystok, Poland.

Bioinformatics +

|

February 20, 2024

View abstract on PubMed

Summary

This study introduces "imputomics," an R package and web app simplifying missing value imputation for mass spectrometry metabolomics data. It offers 41 algorithms and a novel selection tool to improve data quality and analysis.

Area Of Science

Metabolomics
Bioinformatics
Data Science

Background

Missing values are prevalent in mass spectrometry-based metabolomics data.
Accurate imputation of missing values is essential for robust statistical analysis, machine learning, and data completeness.
Existing Missing Value Imputation Algorithms (MVIAs) are numerous but challenging to use due to dependency, documentation, and stability issues.

Purpose Of The Study

To develop a user-friendly tool for imputing missing values in metabolomics data.
To provide a comprehensive solution addressing the challenges associated with existing MVIAs.
To introduce a novel method for selecting optimal MVIAs based on performance and execution time.

Main Methods

Developed the 'imputomics' R package and a Shiny web application.
Integrated 41 established Missing Value Imputation Algorithms (MVIAs) and a random imputation baseline.
Implemented a new functionality for recommending MVIAs tailored for metabolomics data.

Main Results

The 'imputomics' package offers a convenient wrapper for 41 MVIAs.
A web application and command-line tool are available for easy access.
A novel selection functionality aids in choosing the best-performing MVIAs for specific metabolomics datasets.

Conclusions

'imputomics' simplifies the complex landscape of missing value imputation in metabolomics.
The package enhances data quality and facilitates more reliable downstream analyses.
'imputomics' is freely available, promoting wider adoption and reproducibility in the field.